Gosselin v. Quebec:
Back to the Poorhouse …
Natasha Kim and Tina Piper*
regulations
in Quebec during
The case of Gosselin v. Quebec (A.G) made its way to the
Supreme Court of Canada based on a claim that the social
assistance
the 1980s were
discriminstory, single individuals under the age of thirty, who
were considered employable, received only one third of the
assistance granted to their older counterparts. The Court’s
divergent judgments and its narrow five-to-four split reflect the
range and complexity of issues raised by this case.
This comment focuses on the Court’s treatment of the
discrimination analysis under section 15 of the Canadian
Charter of Rights and Freedoms. Through a corparison of the
varied decisions to previous equality jurisprudence, the authors
raise four criticisms of the section 15 analyses taken by the
judges in Gosselin. First, the case highlights how evidentiary
requirements can be used as a barrier to equality, Second,
repeated references to Law v. Canada (Minister of Employment
and Imnigration) suggest a regressive trend in the Court’s
definition of equality. Third, an insufficient consideration of
irersectionality is identified as a distorting factor in the Court’s
the equality analyses. Finally, the authors point to certain
assumptions underlying the judgments in Gosselin, which serve
to perpetuate stereotypes regarding social assistance that have
impaired the human dignity of those living in poverty.
including: institut
Recommendations to enhance flture equality analyses are
offered in conclusion,
clear evidetiary
requirements for claimants; maintaining a precise and conservative
use ofjudicial notice for facts in section 15(1) cases that operate to
the detriment of a claimnt; applying the section 15(1) test
established in Law in both a thorough and contextual maim,
avoiding the rigid categorization of claimamts and appointing
judges from diverse backgrounds combined with educating judicial
decision-makers ahout the stereotypes and the barriers to attaining
substantive equality in our society.
L’arr& Gosselin c. Qudbec (P.G.) a d6 entendu par Is
Cour suprdme do Canada en rdponse A Is prdtention selon
laquelle les rdglements r6gissant I’aide sociale au Qudbec
pendant les anods 80 6taient discriminatoires: les personnes
Agdes de moins de 30 ans, vivant seules et considdnres aptes au
travail recevaient seulement le tiers des prestations accorddes aux
bendficiaires de 30 ans et plus. Les jugements divergents
exprimnds par la Cour et sa d6cision serrde de cinq juges contre
quatre refl&te
bien l’ampleur et Is complexit6 des enjeux
sounlevds par cete cause.
la Cour
Ce cormentaire accorde ne attention particulidre au
traitement que
fait de I’analyse en matidre de
‘article 15 de la Charte canadienne des
discrimination sous
doits et libertA, Par une comparaison des diverses decisions
ant~fieures portant sur le droit A l’dgalitd, les auteurs formulent
quatre critiques A l’encontre des analyses effectu6es par les jugcs
sous l’article 15 dans I’arret Gosselin. Premidrement, l’arret met
en 6vidence comnent les exigences en matire de preuve
peuvent Ore utilisdes comme baries au droit is l’dgalit&
les constantes rdfdrences A Law c. Canada
Deuxidmement,
(Ministre de l’emploi et l’immigration) laissent supposer ne
le droit A
rigression dan
l’dgalitd. Troisinement, le manque de considdration accordde A
l’approche intersectioonelle est idontifat cormm dtant t
facteur
de distorsion dana les analyses que Is Cour fait du droit A
certaines
l’dgalitd.
prdsomptions sous-jacentes aux jugements rends dans l’amet
Gosselin, lesquelles ont pour effet de perptuer des stdrdoypes
relatifsi A l’aide sociale qui portent atteinte A la dignit6 humaine
de ceux qui vivent dam In pauvretd.
la manidre dont la Cour definit
Finalement,
identifient
les
auteurs
Enr conclusion, certaines recommandations sont proposdes
dans le but d’amliorer les fitres analyses en matidre de droit A
l’dgalitd: instituer des exigences claires amx reludrants sur le plan
de Is preuve ; maintiir une utilisation prdcise et conservatrice de Is
connaissance d’office pour les fairs dtai an dddtiment du requdran
dans les causes intents en vert de l’article 15 ; appliquer le test de
l’article 15 tel que d6veloppd dans l’arrt Law d’une nasiimre 4 Is
lois rigoureuse et counextuelle; diter ne catigoriation rigide des
requ6tas; et nommar des juges provenant de diffdrents milieux en
combinaison avec ne sensibilisatin des ddideurs judiciairs mix
stdotypes et aux barrires A ‘atteinte d’une dgat
relle dana
notre socidtd.
* Natasha Kim, LL.B. (Dalhousie); Tina Piper, LL.B. (Dalhousie), B.C.L. (Oxon), M.Phil. (cand.).
E-mails: natasha@rescommunes.ca; tina.piper@balliol.oxford.ac.uk. The authors are grateful to Vince
Calderhead, Richard Devlin, Wayne MacKay, and Simon Archer for their valuable comments,
suggestions and advice on earlier drafts of this paper. Responsibility for any remaining errors is
entirely ours.
McGill Law Journal 2003
Revue de droit de McGill 2003
To be cited as: (2003) 48 McGill L.J. 749
Mode de r~f6rence: (2003) 48 R.D. McGill 749
MCGILL LAW JOURNAL/ REVUE DEDROITDE MCGILL
[Vol. 48
Introduction
1. Gosselin v. Quebec (A.G.)
A. Background
B. The Supreme Court of Canada’s Decision on
Section 15(1)
II. Critique of the Section 15 Equality Analysis
A. Evidence
1. Problems with the Evidentiary Standard Required
2. The Use of Stereotypes
3. The Problem of Judicial Notice
B. Contrast with Past Equality Jurisprudence
1. Comparison between Gosselin and Law
2. Comparing Gosselin to Miron and Egan: Relevant
Distinctions and Legislative Purpose
a. Relevant Distinctions versus Irrelevant Personal
Characteristics
b. Legislative Purpose versus Discriminatory Intent
3. Lovelace: Affirmative Action and Ameliorative Purpose
C. Intersectionality
1. Intersectionality in Past Supreme Court Judgments
2.
Intersectionality in Gosselin
D. Stereotypes
Conclusion
751
752
752
753
756
756
756
758
760
762
763
764
764
766
769
770
770
772
776
2003]
N. KIM & T PIPER – GOSSELIN V. QUEBEC
Introduction
The Supreme Court of Canada released its decision in the highly anticipated case
of Gosselin v. Quebec (A.G.) on 19 December 2002.’ The reason for the unusually
long delay in the release of the judgment is apparent in the final result: a slim five-to-
four split, with four separate dissenting judgments and highly divergent positions
taken by the two factions of the Court.
The basis of this judicial debate was the court challenge by Louise Gosselin to
Quebec’s social assistance regulations of the 1980s. Under section 29(a) of the
Regulation respecting social aid,2 social assistance
treated
differentially on the basis of age and employability. Single individuals under thirty
thirty”), were given
years old, who were considered employable
approximately one third the assistance of their counterparts thirty years and over
(“thirty and over”): only 170 dollars per month.
recipients were
(“under
The case raises innumerable issues ranging from the technical, such as the
interpretation of section 45 of the Quebec Charter of Human Rights and Freedoms3
and section 33 of the Canadian Charter ofRights and Freedoms,4 to the legal, such as
the proper scope of section 7 of the Charter and the justiciability of “economic”
rights. Underpinning all of these, however, are the more nebulous normative issues
which touch the highly contested field of economic and social rights. These
normative issues include the extent to which a nation-state should be compelled to
provide for the basic necessities of its residents, and the reliability and legitimacy of
the judicial perspective in assessing right claims of the young and impoverished.
We are not so ambitious as to address all, or even most, of these difficult issues.
Rather, this case comment focusses on the Court’s section 15 discrimination analysis
and asks whether it serves the commitment to substantive equality that the Court has
repeatedly invoked in its equality jurisprudence. Following a brief history of the case,
we will discuss four criticisms of the section 15 analysis undertaken by the judges
both in majority and dissent. First, we critique the use of evidentiary requirements as
a barrier to equality. We then question the legitimacy of the repeated comparisons of
this case to Law v. Canada (Minister of Employment and Immigration),5 exposing the
seemingly regressive steps taken by the majority in its understanding of equality.
Third, we critique the inadequacy of the analyses in accounting for the intersections
earlier on 29 October 2001.
‘ [2002] 4 S.C.R. 429,221 D.L.R. (4th) 257,2002 SCC 84 [Gosselin]. The appeal was heard a year
2 1.1,Q. 1981, c. A-16, r. 1, s. 29(a), adopted under the Social AidAct, R.S.Q., c. A-16, as re-en. by
An Act respecting income security, R.S.Q., c. S-3.1.1, as re-en. by An Act respecting income support,
employment assistance and social solidarity, R.S.Q., c. S-32.001.
3 R.S.Q., c. C-12, s. 45 [Quebec Charter].
4 Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.), 1982, c. 11
[Charter].
‘ [1999] 1 S.C.R. 497, 170 D.L.R. (4th) 1 [Law].
752
MCGILL LAW JOURNAL / REVUE DE DROITDE MCGILL
[Vol. 48
of various identity markers. Finally, we extract the assumptions that underlie some
aspects of the judgments. We conclude with recommendations that will hopefully
avoid some of the pitfalls we have discussed.
I. Gosselin v. Quebec (A.G.)
A. Background
As noted above, the Regulation respecting social aid limited the amount of social
assistance payable to recipients under thirty to one-third of what was available to
recipients thirty and over. The regulation did, however, make it possible for those
under thirty to increase their payments to the level of social assistance paid to
recipients thirty and over if they participated in one of three programs: On-the-Job-
Training, Community Work, or Remedial Education.’ Ms. Gosselin brought the case
as a class action on behalf of all social assistance recipients subject to the differential
regime from 1985 to 1989,7 arguing that the differential regime violated sections 15
and 7 of the Charter and section 45 of the Quebec Charter. The claim failed on all
three grounds at all levels of court.
With respect to the section 15 claim, Reeves J. held at trial that the claim was not
supported by the evidence and that the distinction made under the Regulation
respecting social aid was not discriminatory under section 15(1) of the Charter.’ At
the Quebec Court of Appeal, two of the judges found a section 15(1) violation based
on age, but only Robert J.A. (as he was then) held that the violation was not justified
under section 1.’ Mailhot J.A. found no violation at all.
6 See Regulation to amend the Regulation respecting social aid, O.C. 872-84, 5 April 1984, G.O.Q.
1984.11.1687, s. 2, as am. by O.C. 1347-84, 6 June 1984, G.O.Q. 1984.11.2399. Note that participation
in the first two programs raised social assistance benefits up to the base amount available to those
thirty and over. Participation in Remedial Education only raised social assistance payments to within
100 dollars of the base amount payable to those thirty and over. The Regulation respecting social aid
was in force from 1984 until 1989, when it was replaced by legislation that does not make age-based
distinctions.
7 Calculated by the appellants to be 75,000 individuals (though unnamed in the appellant’s
submissions). See Gosselin, supra note I at para. 8.
8 Gosselin c. Qubec (Procureur Giniral) [1992] RJ.Q. 1647 (C.S.).
9 Gosselin c Quibec (Procureur Gdndral) [1999] RJ.Q. 1033 (C.A.). Mailhot J.A. dismissed the
case on the basis that it was indistinguishable from Law and that the scheme in its entirety and in
context did not violate the appellant group’s dignity or liberty (ibid at para. 9). Baudouin J.A. held
that the social assistance scheme violated section 15(1) as age discrimination that imposed economic
disadvantage on those under thirty. He found, however, the infringement to be justified under section I
of the Charter, because the government should be accorded a high level of deference in the allocation
of scarce economic resources (ibid. at paras. 17, 31). Robert J.A., in the minority, would have found a
violation of section 15(1) that was not justifiable by section 1, but would have dismissed the remedy
of damages as inappropriate. Ms. Gosselin had asked that the Quebec government be ordered to pay
2003]
N. KIM & T PIPER – GOSSELIN V. QUEBEC
753
B. The Supreme Court of Canada’s Decision on Section 15(1)
The Supreme Court of Canada held, by a slim majority of five to four, that the
social assistance scheme established by the Regulation respecting social aid did not
infringe section 15(1) of the Charter. The majority judgment was written by
McLachlin C.J. and was concurred in by Gonthier, Iacobucci, Major, and Binnie, JJ.
The dissenting judgments held that the regulation violated section 15(1) of the
Charter and that the infringement was not justified under section 1.
The main area of conflict between the majority and the minority was the
application of the Law section 15(1) test, in particular, the four contextual factors to
be considered under the third branch of the test. Since it will be referred to throughout
this piece, the Law test for discrimination is reproduced here:
[A] court that is called upon to determine a discrimination claim under s. 15(1)
should make the following three broad inquiries:
(A) Does the impugned law (a) draw a formal distinction between
the claimant and others on the basis of one or more personal
characteristics, or (b) fail to take into account the claimant’s
already disadvantaged position within Canadian society
resulting in substantively differential treatment between the
claimant and others on the basis of one or more personal
characteristics?
(B)
Is the claimant subject to differential treatment based on one
or more enumerated and analogous grounds?
and
reflects
(C) Does the differential treatment discriminate, by imposing a
burden upon or withholding a benefit from the claimant in a
manner which
the stereotypical application of
presumed group or personal characteristics, or which
otherwise has the effect of perpetuating or promoting the view
that the individual is less capable or worthy of recognition or
value as a human being or as a member of Canadian society,
equally deserving of concern, respect, and consideration?
Some important contextual factors influencing the determination of
whether s. 15(1) has been infringed are, among others:
(A) Pre-existing
disadvantage,
stereotyping, prejudice,
or
vulnerability experienced by the individual or group at issue.
The effects of a law as they relate to the important purpose of
s. 15(1)
individuals or groups who are
vulnerable, disadvantaged, or members of “discrete and
insular minorities” should always be a central consideration….
in protecting
almost 389 million dollars in benefits plus the interest accrued since 1985 (Gosselin, supra note 1 at
para. 4).
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 48
(B) The correspondence, or lack thereof, between the ground or
grounds on which the claim is based and the actual need,
capacity, or circumstances of the claimant or others. … [I]t will
generally be more difficult to establish discrimination to the
extent that the law takes into account the claimant’s actual
situation in a manner that respects his or her value as a human
being or member of Canadian society, and less difficult to do
so where the law fails to take into account the claimant’s
actual situation.
(C) The ameliorative purpose or effects of the impugned law upon
a more disadvantaged person or group in society. An
ameliorative purpose or effect which accords with the purpose
of s. 15(1) of the Charter will likely not violate the human
dignity of more advantaged individuals where the exclusion of
these more advantaged individuals largely corresponds to the
greater need or the different circumstances experienced by the
disadvantaged group being targeted by the legislation.
and
(D) The nature and scope of the interest affected by the impugned
law. The more severe and localized the consequences of the
legislation for the affected group, the more likely that the
differential treatment responsible for these consequences is
discriminatory within the meaning of s. 15(l).10
McLachlin C.J., writing for the majority, concluded, after an examination of the
four factors from Law, that there was no support for a finding that there had been
discrimination and a denial of human dignity to constitute a violation of section
15(1). First, she held that young people, as a group, had not suffered from historical
disadvantage and age distinctions were common and necessary for ordering society.
Second, she observed that there was a correspondence between the scheme and the
actual circumstances of the social assistance recipients: the provision of education
and training provided incentives for young people to work and affirmed their
potential and did not undermine their dignity. Third, she noted that the ameliorative
purpose factor was neutral in this case since the Regulation respecting social aid was
not designed to improve the condition of another group (e.g., recipients who are thirty
and older). McLachlin C.J. concluded that the impugned law did not adversely affect
the appellant’s dignity and that any adverse short-term effects were outweighed by the
legislation’s attempt to improve the self-reliance and dignity of the group.”
In dissent, Bastarache J. (Arbour and L’Heureux-Dub6 JJ. concurring in the
section 15(1) analysis) applied the Law criteria to different effect. First, he argued that
the majority mischaracterized the group (i.e., young people) and should have
t0 Law, supra note 5 at para. 88.
tGosselin, supra note 1 at para. 74.
2003]
N. KIM & T PPER – GOSSELIN V. QUEBEC
755
considered the special vulnerability of social assistance recipients under thirty.
Second, he held that there was a lack of correspondence between the differential
social assistance scheme and the actual needs, capacities, and circumstances of social
assistance recipients under and over thirty. Third, Bastarache J. rejected
the
ameliorative purpose factor in this case and found that the scheme had a severe effect
on an extremely important interest to the claimant by knowingly placing her and
others in a precarious and unliveable situation.
Bastarache J. (with the concurrence of LeBel J. in full and Arbour J. in part) then
found that the infringement was not justified under section 1 of the Charter. Although
he accepted that a certain level of deference should be paid to government in
reviewing this type of legislation, he held that this does not give the government carte
blanche to limit rights. Bastarache found the objectives-to facilitate the integration
of youth into the workforce and to “avoid attracting them to social assistance”– 2 to
be pressing and substantial and also found a rational connection between those
objectives and some differential treatment of the under thirty recipients. Bastarache J.
concluded, however, that the measures enacted by the government were not
minimally
impairing
alternatives to achieve the government’s goals. He further found that there was no
proportionality between the detrimental effects on the appellant’s self-worth and
equality and the government’s objectives.
impairing of the appellant’s rights and suggested
less
Arbour J. concurred with this minimal impairment analysis, and further argued
that there was no rational connection between denying social assistance recipients the
basic means of subsistence and promoting their dignity and liberty.
LeBel J. also dissented from the majority, holding that the differential legislative
treatment between social assistance recipients was based on stereotypes of youth and
discriminated against young recipients for no valid reason.
Finally, L’Heureux-Dubd J. held that the Court should not consider the
legislature’s good intentions
in determining whether the impugned legislation
violated section 15(1) and held that discrimination based on age did not operate only
in relation to old age. She found that the legislative distinction was discriminatory, in
part, because of the harm to Ms. Gosselin’s fundamental interests in physical and
psychological integrity, as a result of both a personal characteristic that could not be
changed and the pre-existing disadvantage of some members of the group.
12 Ibid. at para. 263 [emphasis added], quoting from Robert J.A.’s judgment in the Court of Appeal.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 48
II. Critique of the Section 15 Equality Analysis
A. Evidence
1. Problems with the Evidentiary Standard Required
Among the many issues discussed by the Supreme Court of Canada in Gosselin
one of the most contentious was the evidence required to support the appellant’s
claims. In considering the evidence, the Court arguably misapplied the standard
established in Law and previous jurisprudence. In the absence of what it viewed as
appropriate evidence, the majority seemed to resort to the use of stereotypical
generalizations to underpin its legal argument. These stereotypes are problematic, not
only because they went unacknowledged in the judgment and are incorrect, but also
because they have been effectively codified through judicial notice, and will now be
difficult to challenge.
McLachlin C.J. held that the claimant had not adduced sufficient evidence to
ground her claim. This position was made clear throughout her decision beginning
with her reference to the lack of “direct evidence of any other young person’s
experience with the government programs”‘ 3 provided by Ms. Gosselin. It was seen
again in her concern for making inferences about the program’s ability to respond to
the needs of a particular group “absent concrete evidence.”‘ 4 McLachlin C.J.
expressed further concern with the mode of evidence adduced by the plaintiff,
pointing out that Ms. Gosselin “alone provided first-hand evidence and testimony as a
class member”‘ and that there was “no indication that Ms. Gosselin [could] be
considered representative of the [under thirty] class.” 6
These dicta are troubling for the following reasons. First, McLachlin C.J. appears
to have applied a higher evidentiary standard to the claimant in this case than is
generally required in section 15 challenges. In Law, lacobucci J., speaking for the
Court, expressly warned against imposing too heavy a burden on claimants and
clarified that claimants would not be required to adduce social science evidence or
other data “not generally available, in order to show a violation of the claimant’s
dignity or freedom.”‘ 7 Rather, they should be allowed, if appropriate, to rely on
judicial notice and logical reasoning to establish their claims.’
3 Ibid. at para. 8.
14Ibid at para. 54.
lbid. at para. 8.
‘6Ibid. at para. 47.
17 Law, supra note 5 at para. 77.
18 Ibid. at para. 78, citing cases such as Andrews v. Law Society of British Columbia, [1989] 1 S.C.R.
143, 56 D.L.R. (4th) [Andrews cited to S.C.R.]; R. v. Turpin, [1989] 1 S.C.R. 1296, 96 N.R. 115;
Weatherall v. Canada (A.G.), [1993] 2 S.C.R. 872, 105 D.L.R. (4th) 210. We could also add to this list
2003]
N. KIM & T PIPER – GOSSELIN V. QUEBEC
Second, although it appears that the majority did not explicitly ask for the
claimant to adduce data or social science evidence, this palliative does not withstand
closer inspection. The majority complained that Ms. Gosselin had not adduced
sufficient evidence of the problems faced by other members of the class and implied
that she might not be representative of the class.” This begs the rhetorical question of
how many claimants would be required to prove that Ms. Gosselin is representative
of a class. We think it unlikely that the testimony of four (0.005 percent of
participants), fifteen (0.02 percent) or even one hundred (0.1 percent) participants in a
program of 75,000 participants would have been adequate or rigorous enough to meet
this elusive standard of representativeness. Ultimately the evidentiary requirements to
demonstrate discrimination against even a minute fraction of participants could
become an extensive exercise of social science data collection. As pointed out by
Bastarache J. in dissent, this burden of proof seemed particularly onerous since Ms.
Gosselin’s claim of “the existence of a group of persons harmed by facts deriving
from a common origin”20 had already been proved in the authorization as a class
action. As the authorization was not a live issue in the appeal, there was no legal
requirement that Ms. Gosselin provide extra proof that she represented the class.
Third, a claimant’s decision to organize the claim as a class action should not
jeopardize her case. Had Ms. Gosselin presented her claim without the class action,
and had the Court found in her favour, it would have undoubtedly considered the fact
that 75,000 other potential plaintiffs existed and that Ms. Gosselin was a member of a
“class”. The Court could then have walked the “fine line between fualflling [the]
judicial role of protecting rights and intruding on the legislature’s role”‘ and
mitigated the plaintiff’s demands through remedial provisions. Applicable remedial
provisions were outlined by Bastarache J. in dissent22 and remedial strategies for
similar situations have been suggested or used in other constitutional cases, such as
Schachter v. Canada23 and Egan v. Canada.24
Fourth, the inadequacy of Ms. Gosselin’s evidence, as held by the majority,
seemed to have affected the finding of discrimination
in her particular case.
Discrimination against one claimant, however, should be sufficient to found a
Charter violation. As Bastarache J. accurately pointed out in his dissent, “it would be
a departure from past jurisprudence for this Court to refuse to find a Canadian
Charter breach on the basis that the claimant had not proven disadvantage to enough
others.”’25 Stringent evidentiary requirements on plaintiffs under the first stage of the
Moge v. Moge, [1992] 3 S.C.R. 813, 99 D.L.R. (4th) 456, L’Heureux-Dub6 J. [Moge]; Mossop v.
Canada, [1993] 1 S.C.R. 554, 100 D.L.R. (4th) 658, L’Heureux-Dube I [Mossop].
” Gosselin, supra note 1 at para. 33.
20Ibid. at para. 249.
2 Ibid. at para. 292, Bastarache J., dissenting. See also Sauvg v. Canada (Chief Electoral Oficer),
[2002] S.C.R 98, C.R.R. (2d) 1, 2002 S.C.C. 68.
22 Gosselin, supra note I at paras. 291-99.
23 [1992] 2 S.C.R. 679, 93 D.L.R. (4th) 1.
24 [1995] 2 S.C.R. 513, 124 D.L.R. (4th) 609, lacobucci J., dissenting [Egan cited to S.C.R.].
25 Gosselin, supra note 1 at para. 249.
758
MCGILL LAW JOURNAL / REVUE DE DROIT DE McGILL
[Vol. 48
section 15 analysis, such as those applied by the majority in this case, force plaintiffs
to shoulder a large part of the evidentiary burden that should properly rest on the
government in the section 1 analysis.
The fifth major problem is the barrier this creates for socio-economically
deprived claimants who may wish to challenge the allocation of benefits by the
government. Placing such a high evidentiary standard on claimants may put these
challenges out of reach of such parties, both in terms of the investment required to
generate the data for these cases as well as the difficulty of contacting people who
insecure
live in poverty. Those
accommodations and employment, potentially with limited access to technology and
other resources. 6 Further, such individuals may be reluctant to respond to any
demands for information to avoid jeopardizing the benefits they currently receive.
living in poverty are often transient due to
Finally, the seemingly heightened burden on this particular group of claimants,
and in particular on Ms. Gosselin as the representative of the group, could lead to the
inference that the evidence provided by someone in poverty is somehow less worthy
of being believed. As we will see in the following section, the idea that the credibility
of those receiving social assistance is somehow impaired, especially in the context of
claiming benefits, derives its force in part from discriminatory assumptions and
prevalent stereotypes.
2. The Use of Stereotypes
The presumed lack of evidence may lead courts to rely more heavily on
assumptions or stereotypes about the classes of claimants before them. Although
McLachlin C.J. rejected this approach, it was arguably central to the majority’s
analysis. First, McLachlin C.J. repeatedly rested arguments in her judgment on the
stereotype of the enhanced employability of younger people. She stated, for example,
that “young adults as a class do not seem especially vulnerable or undervalued.”’27 She
continued by stating that to believe that young adults may be subject to “negative
preconceptions” would be a “counter-intuitive” proposition,2″ adding that “[i]f
anything, people under 30 appear to be advantaged over older people in finding
26 See e.g. Statistics Canada, “Internet use rates, by location of access and household income”,
online: Statistics Canada
27 Gosselin, supra note I at para. 33.
28 Ibid. at para. 32:
Concerns about age-based discrimination typically relate to discrimination against
people of advanced age who are presumed to lack abilities that they may in fact
possess. Young people do not have a similar history of being undervalued. This is by no
means dispositive of the discrimination issue, but it may be relevant, as it was in Law.
2003]
N. KIM & T PIPER – GOSSELIN V. QUEBEC
employment.” ’29 These comments were based on a belief that youth are more flexible
and have more modem skills than older people.30
A second stereotype underpinning the majority judgment (and workfare programs
generally) was that youth must be forced through financial desperation to pursue
work training opportunities.3′ Further stereotyping in the case posited that younger
people do not respond as well as older people to the incentive programs created by
the government.32
Not only did the Court base its decisions on stereotypical assertions regarding
youth employment, it also denied that it was engaging in this exercise. Bastarache J.
pointed out that even though the legislature might have had positive intentions in
differentiating between the over and under thirty groups, doing so was based on the
“unverifiable presumption that people under 30 had better chances of employment
and lower needs.”’33 McLachlin C.J. refuted this argument by saying that Bastarache
J.’s point seemed “to place on the legislator the duty to verify all its assumptions
empirically, even when these assumptions are reasonably grounded in everyday
experience and common sense.”’34 She further held that “the idea that younger people
may have an easier time finding employment than older people” was not an “arbitrary
and demeaning stereotype” and it was, therefore, unproblematic.35
the
through
Although not touched on by the Court, many more considerations underpin youth
lens of common
employability. These are often understood
misperceptions that can be refuted by empirical fact.36 The problems with the “fact”
or stereotype of youth employability upon which the majority relies are both
theoretical and empirical. Theoretically, youth are not necessarily at an advantage.
Youth are burdened by the presumption that they can find jobs easily if they look for
them and that they are not “family breadwinners” with whom older employers may
identify. Both these presumptions may make it easier for an employer to terminate a
young person’s employment. Further, youth may be thought to be unreliable,
transient, rebellious (particularly if they have piercings or body art), and resourceful
such that they will find a way to survive with less money. By contrast, older claimants
may be particularly advantaged by the fact that they have more job and life
experience, greater awareness of available training programs and opportunities,
longer track records, greater knowledge of the system, and larger networks of
29 Ibid. at para. 34.
30 Ibid., citing McKinney v. University of Guelph, [1990] 3 S.C.R. 229, 76 D.L.R. (4th) 545 at para.
92 [McKinney cited to S.C.R.].
31 See e.g., Gosselin, ibid. at para. 60. See also ibid. at para. 250, Bastarache J. This will be
discussed more fully in Part U.D, below.
32 Ibid. at para. 250, Bastarache J.
33 Ibid. at pam. 248.
34 Ibid. at para. 56 [emphasis added].
35 Ibid.
36 Ibid at para. 250, Bastarache J.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 48
contacts. Also, older claimants may inspire greater commitment from employers as
they share certain contextual commonalities.
These common misperceptions about younger claimants, and potentially more
favourable conditions for older claimants, are supported by the empirical evidence.
The most recent Statistics Canada data for 2001-easily accessible and accurate data
available through the Intemet–demonstrate that the fact that youth may be more able
to find and maintain employment is incorrect.37 The group facing the highest rates of
unemployment is that of fifteen to nineteen year olds (16.6 percent); the lowest rate is
held by forty-five to fifty-four year olds (5.4 percent). Even between groups with
relatively similar participation rates, younger people fare worse than older people:
twenty-three to thirty-four year olds have a 6.9 percent unemployment rate whereas
forty-five to fifty-four year olds have a 5.4 percent unemployment rate. These
statistics are paralleled by the data available for Quebec in the census years 1981,
1986, 1991, and 1996, which show twenty to twenty-four year olds as having
unemployment rates of 16.3 percent, 18.4 percent, 17.2 percent and 17.1 percent in
those years and
to fifty-four year olds (with relatively similar
participation rates in the labour market) as having unemployment rates of 8.0 percent,
10.8 percent, 10.8 percent and 10.6 percent.8 Unemployment rates for younger
people in Quebec have consistently been over five percent higher than those of older
people.
twenty-five
3. The Problem of Judicial Notice
A lack of evidence may lead courts to rely more heavily on, as Iacobucci J.
foreshadowed in Law, non-evidentiary tools; in particular, “[a] court may often,
where appropriate, determine on the basis of judicial notice and logical reasoning
alone whether the impugned legislation infringes s. 15(1).””9 In Gosselin the majority
37 Statistics Canada, “Labour Force Characteristics by Age and Sex”, online: Statistics Canada
analysis of the statistics adduced in Gosselin, supra note I at para. 235 [emphasis added]:
The purpose of undertaking a contextual discrimination analysis is to try to determine
whether the dignity of the claimant was actually threatened. In this case, we are not
dealing with a general age distinction but with one applicable within a particular social
group, welfare recipients. Within that group, the record makes clear that it was not, in
fact, easier for persons under 30 to get jobs as opposed to their elders … Thus, the
stereotypical view upon which the distinction was based, that the young social welfare
recipients suffer no special economic disadvantages, was not grounded in fact; it was
based on old assumptions regarding the employability of young people. The creation of
the assistance programs themselves demonstrates that the government itself was aware
of this disadvantage.
38 See Statistics Canada, “1981-1996 Census – Labour Force Activity by Sex and Age Groups,
Quebec, Both Sexes”, online: Statistics Canada
39Law, supra note 5 at para. 77.
2003]
N. KIM & T PIPER – GOSSELIN V. QUEBEC
relied on Iacobucci J.’s finding in Law that “the increasing difficulty with which one
can find and maintain employment as one grows older is a matter of which a court
may appropriately take judicial notice,” particularly since the Court has “often
recognized age as a factor
in the context of labour force attachment and
detachment.”4
This assertion and its adoption by the majority in Gosselin poses three major
problems. First, to qualify as a fact worthy of judicial notice, the fact must be either
“(a) so notorious as not to be the subject of dispute among reasonable persons, or (b)
capable of immediate and accurate demonstration by resorting to readily accessible
sources of indisputable accuracy.”‘ Judicial notice of the fact that younger people
more easily find and maintain work than older people violates both these conditions:
it is the subject of dispute among reasonable persons, as Justice Bastarache’s dissent
clearly establishes,42 and as noted above, it cannot even be confirmed immediately
and accurately by Canada’s readily accessible and reliable source of statistical data,
Statistics Canada. As a result, an unverified hypothesis that is not the subject of
agreement amongst reasonable people is transformed into an indisputable fact
supported by the weight of precedent that can determine the just allocation of
government benefits.
Second, taking judicial notice in this case directly contradicts lacobucci J.’s
warning in Law that caution must be exercised in the equality analysis such that “one
should not unwittingly or otherwise use judicial notice to invent stereotypes or other
social phenomena which may not or do not truly exist.”’43
Lastly, taking judicial notice of the proposition that there is increasing difficulty
in finding and maintaining employment as one grows older is problematic because of
the imprecise nature of this statement. This imprecision has huge implications for its
application by both the Supreme Court of Canada and lower courts in the future. The
terms “older” and “younger” are so relative and vague as to be meaningless,
especially when one asks who is older and younger and in what job market. This can
be demonstrated by the fact that the Court has applied this same reasoning to a
shifting age standard that depends on the case at hand-in McKinney v. University of
Guelph” the Court drew the line between young and older persons at forty-five years
40 Ibid. at para. 34. laccobucci J. cited the following judgments as supportive: McKinney (supra note
30 at para. 92, La Forest J.: “[b]arring specific skills, it is generally known that persons over 45 have
more difficulty finding work than others. They do not have the flexibility of the young, a disadvantage
often accentuated by the fact that the latter are frequently more recently trained in the more modem
skills”); Machtinger v. HOJ Industries Ltd. ([1992] 1 S.C.R. 986, 91 D.L.R. (4th) 491, lacobucci and
McLachlin JJ.); Moge (supra note 18, McLachlin J.).
41 John Sopinka, Sidney N. Lederman & Alan W. Bryant, The Law of Evidence in Canada (Toronto:
Butterworths, 1999) at 1055.
42 Gosselin, supra note I at paras. 150-304.
43 Law, supra note 5 at para. 79.
44Supra note 30.
MCGLL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 48
of age, in Law it was thirty-five years, and in Gosselin it was thirty years.45 If
employability based on age is to be used as a fact, it ought to be precisely defined or
delimited to prevent its abuse.
The notion of “finding employment ‘
offers another demonstration of the
vagueness of the facts that were given judicial notice. Must employment be
maintained for a given period of time? Does “finding” mean “securing” or merely
“discovering” employment? How can one be sure that the underlying concept will not
be altered and misrepresented in subsequent judicial decisions? Legislative facts
should be proven by the testimony of experts in the relevant field of knowledge,
precisely delimited, and adapted to the contextual facts of a case.47 It is admirable that
the impetus in taking judicial notice of facts and using logical reasoning was to limit
the evidentiary burden of data and social science evidence on claimants and
respondents. However, it serves neither party’s interests if judicial notice is taken of
to perpetuate
incorrect facts or of imprecise concepts
disadvantage and misinformation.
that ultimately serve
B. Contrast with Past Equality Jurisprudence
Throughout its judgment in Gosselin, the majority used Law as justification for
many of its pronouncements due to the “striking similarity”‘ of the two cases. At
least three major differences between the cases demonstrate, however, that heavy
reliance on Law may in fact have been a misapplication of Law’s holdings. Similarly,
the majority judgment took jurisprudential licence with previous section 15 cases of
the Court. In particular, this includes the revitalization of the idea of “relevant”
distinctions from Miron v. Trudel”9 and Egan, and the consideration of “ameliorative
purposes” from Lovelace v. Ontario.5 As a consequence, the judgment appears to be
making regressive steps in the equality record that the Supreme Court of Canada had
previously set, arguably undermining and contradicting its recent pronouncements on
the substantive equality standard.”
45 While giving some margin to the special nature of the ground of “age” as compared to other
immutable grounds like race (see e.g. Gosselin, supra note I at para. 225; Peter Hogg, Constitutional
Law of Canada, looseleaf (Toronto: Carswell, 1999) at 52-54), to take an example at the extremes,
there is a big difference in the employability of a teenager and a 44 year old, though arguably judicial
notice may have been taken of the fact that their employability is the same. This highlights the
meaninglessness and malleability of this assumed fact to suit the purposes of judicial reasoning, if not
justice.
46 Gosselin, ibid. at para. 56.
47 This echoes Bastarache J’s concern (ibid. at para. 235).
48 Ibid at para. 73.
41 [1995] 2 S.C.R. 418,23 O.R. (3d) 160 [Mirn].
‘0 [2000) 1 S.C.R. 950, 188 D.L.R. (4th) 193,2000 SCC 37 [Lovelace].
51 See e.g. Eldridge v. British Columbia (A.G.), [1997] 3 S.C.R. 624, 151 D.L.R. (4th) 577
[Eldridge]; Vriendv. Alberta, [1998] 1 S.C.R. 493, 156 D.L.R. (4th) 385.
2003]
N. KIM & T PIPER – GOSSELIN V. QUEBEC
1. Comparison between Gosselin and Law
In addition to providing judicial notice of convenient facts, the majority relied on
Law to support a number of its holdings, at one point explicitly stating that “[i]n
many respects, the case before us is strikingly similar to Law.”‘2 With respect, this
reliance may be misplaced, since the cases are distinctly different. As Bastarache J.
commented, “The fact that a certain legislative provision which limited the benefits to
those under a certain age was found to be constitutional in one case does not
necessarily lead to the same conclusion here.”’53 Overemphasis on the holdings in Law
not only produces the problematic results related to stereotyping and judicial notice,
discussed above, but also raises the following issues.
First, the Canada Pension Plan (“CPP”) benefits at issue in Law were related to
marital status and age and were provided regardless of financial need while in
Gosselin, social assistance was a program of last resort allocated to meet basic needs.
Unlike the beneficiaries in Law, the beneficiaries in Gosselin uniformly had no other
sources of income or employment and clearly required the payments to live. Drawing
a parallel between the two cases reflects an underlying assumption that social
assistance recipients choose to receive social assistance.54 As such, comparing the
government’s consideration of “long-term needs” in both cases, as the majority did in
Gosselin,” overlooks the urgency of the short-term needs of social assistance
recipients in Gosselin and generally misrepresents the different circumstances of the
two kinds of claimants.56
Second, the CPP payments in Law did not impose training or educational
requirements on the recipients, demonstrating that the allocation of those benefits
came with a much lighter burden on the recipients in Law than in Gosselin. As well,
the legislation in Law did not draw a bright line between those over and under thirty,
and in fact provided a scheme of gradually decreasing payments for widows between
thirty-five and forty-five.
Finally, “[i]n Law … lacobucci J. held that a piece of legislation might be less
harmful to a group’s dignity if its purpose or effect is to help a more disadvantaged
person or group in society.”’57 This argument worked in that case to justify the
government’s actions because the legislative purpose was to aid elderly widows (as
52 Gosselin, supra note 1 at para. 73.
3 Ibid. at para. 233.
54 See Part 1.D, below,
55 Gosselin, supra note I at para. 44.
56 Ibid. at para. 252. As Bastarache J. argues:
The difference in the nature and importance of the interest affected–provision for basic
needs immediately as opposed to over the long term-is one of the crucial distinctions
between the present case and Law. … A genuine contextual approach will appreciate
this distinction and will not find the result determined by the apparent similarities in
that both cases address an age distinction for a government benefit
57Ibid. at para. 250, Bastarache J.
764
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 48
opposed to the disadvantaged young widow claimants). Conversely, in Gosselin, the
legislation’s purpose was to help the very group it disadvantaged. As we will examine
in the next sections, the majority’s treatment of this purpose in the same way as the
ameliorative purpose in Law compromises its section 15 equality analysis.
2. Comparing Gosselin to Miron and Egan: Relevant Distinctions
and Legislative Purpose
Throughout the dissenting judgments in Gosselin, a recurring criticism of the
majority decision was the use of legislative purpose in the third step of the Law
analysis to incorporate section 1 justification concerns. The criticism was two-fold:
that the inclusion distorted the analysis of every previous Charter case by not keeping
the violation and justification inquiries distinct;” and that a positive legislative
purpose was not an appropriate consideration at the section 15 stage of the analysis.59
Both of these concerns can be traced to a deeper criticism that the methods of analysis
used by the majority employ concepts which, although introduced in past decisions
by the Court, have since been refuted by its own jurisprudence. The first such concept
is the consideration of relevant distinctions and the second is the blurring of the two
concepts of discriminatory intent (as opposed to discriminatory effect) and legislative
purpose. Inextricably linked, one derives analytical strength from the other. Both
were introduced in the twin cases of Miron and Egan.6’ We examine each concept
separately below.
a. Relevant Distinctions versus Irrelevant Personal Characteristics
The idea that “relevant” distinctions could be considered in the section 15 inquiry
was supported by the same four justices in both Miron and Egan.6 Finding against
the expansion of the definition of spouse to include either common-law or same-sex
spouses in those cases, these justices reformulated the section 15 analysis, as
originally stated in Andrews, to include a third step of “whether the distinction made
by Parliament is relevant [considering] ‘the nature of the personal characteristic and
its relevancy to the functional values underlying the law.”’62
This subtle reformulation transforms the accepted analytical step of determining
whether a distinction is based on an “irrelevant personal characteristic” to the very
different exercise of determining whether the distinction being made is relevant to the
“functional values” of the law.63 As becomes clear through the judgments, the term
“functional values” connotes a positively perceived legislative purpose. In Miron and
” Ibid. at para. 103, L’Heureux-Dubd J.
s9 Ibid at para. 243, Bastarache J.
60 Miron, supra note 49 at para. 13; Egan, supra note 24 at paras. 42-48.
61 Lamer CJ. and La Forest, Gonthier, and Major JJ.
62Egan, supra note 24 at para. 13. See also Miron, supra note 49 at paras. 14-15.
63Egan, ibid.
2003]
N. KiM & T PIPER – GOSSELIN V. QUEBEC
Egan, that purpose was the support of marriage as a heterosexual institution: “The
distinction adopted by Parliament is relevant, indeed essential, to describe
the
relationship in the way the statute does so as to differentiate the couples described in
the statute from all couples that do not serve the social purposes for which the
legislature has made the distinction. ‘ ‘”
to
Other cases that followed, including the Law restatement of section 15, have
affirmed that the balancing and policy considerations allowed by an examination of
relevant distinctions are not appropriate
the section 15 discrimination
determination, although these considerations may have force under section 1.6′ The
only so-called relevant distinction is the one being assessed; that is, the legislative
distinction defined in the first stage of the Law analysis is only relevant because it is
being challenged as discriminatory. In the same vein, the personal characteristic,
defined in the second part of the Law analysis will be irrelevant simply because the
legislative distinction is derived from stereotypes or assumptions about a person
based on the characteristic. In this way, enumerated and analogous grounds are
irrelevant personal characteristics when used as “an illegitimate proxy for merit … or
ability,”” although the distinction may not necessarily result in a finding of
discrimination under the third part of the section 15 test. The use of (ir)relevance to
reflect determinations
intrinsic to the analysis should not act as a door to
inappropriate balancing of interests or justifications of legislative purpose.
Unfortunately, the majority judgment in Gosselin has revived the inquiry into
relevant distinctions under the third part of the analysis. This is especially apparent in
its application of the second contextual factor enumerated in this part of the Law test:
the correspondence, or lack thereof, between the ground on which the claim is based
and the actual need, capacity, or circumstances of the claimant.67 The original
motivation for considering this factor is clear. Given our benchmark of substantive
over formal equality, we have recognized that in certain circumstances, true equality
may require that different people are treated differently. Therefore, when government
action responds to the need for substantive equality, this should mitigate against a
finding of discrimination. To the contrary, the majority in Gosselin used the analysis
of needs, capacities, and circumstances to embark on an inquiry reminiscent of the
proportionality analysis of section 1 to determine the reasonableness of the legislative
purpose:
” Egan, ibid at para. 27.
65 See e.g. Eldridge, supra note 51. Somewhat surprisingly, the Eldridge case was neither
mentioned by the majority nor were its principles incorporated. Considering that the challenges of
addressing underinclusiveness do not even arise in this case, the simple idea articulated by the full
Court in Eldridge (that once a benefit is given it should not be given in a discriminatory manner)
should be uncontroversial at this stage of our equality jurisprudence. For a good discussion, see also
Collins v. Canada, [2002] 3 F.C. 320, 285 N.R. 359 (C.A.) (Factum of the Appellant at paras. 72-77).
“Gosselin, supra note 1 at para. 226, Bastarache J.
67 See Law, supra note 5 at paras. 69-71.
MCGILL LAW JOURNAL/ REVUE DEDROITDE MCGILL
[Vol. 48
A final objection is that the selection of 30 years of age as a cut-off failed to
correspond to the actual situation of young adults requiring social assistance.
However, all age-based legislative distinctions have an element of this literal
kind of “arbitrariness”. That does not invalidate them. Provided that the age
chosen is reasonably related to the legislative goal, the fact that some might
prefer a different age-perhaps 29 for some, 31 for others-does not indicate a
lack of sufficient correlation between the distinction and actual needs and
circumstances. Here, moreover, there is no evidence that a different cut-off age
would have been preferable to the one selected.6
This blurring between the violation and justification analyses is closely related to the
second branch of our critique: the use of purpose as a justificatory proxy for
discriminatory intent.
b. Legislative Purpose versus Discriminatory Intent
Most of the foundational section 15 equality jurisprudence has centered around a
dichotomy influenced by rulings regarding statutory human rights codes: direct
discrimination (emanating from discriminatory intention) and indirect discrimination
(resulting from an adverse effect).69 Recent human rights codes jurisprudence has
moved away from this dichotomy in favour of a more holistic approach that better
targets the effects of discrimination inbred into entrenched standards, systems, and
institutions. This more nuanced approach to the statutory equality analysis was
explained by McLachlin J. (as she then was) in her judgment for the Court in British
Columbia (Public Service Employee Relations Commission) v. BCGSEU:70
Under the conventional analysis, if a standard is classified as being
“neutral” at the threshold stage of the
is never
questioned. The focus shifts to whether the individual claimant can be
accommodated, and the formal standard itself always remains intact. The
conventional analysis thus shifts attention away from the substantive norms
underlying the standard, to how “different”
individuals can fit into the
“mainstream”, represented by the standard.
inquiry, its legitimacy
Interpreting human rights legislation primarily in terms of formal equality
undermines its promise of substantive equality and prevents consideration of
the effects of systemic discrimination …7
We have seen very little of this sort of analysis transferred, however, into the section
15 arena, particularly with what appear to be regressive steps being taken by the
majority in the Gosselin case.
6 Gosselin, supra note I at para. 57 [emphasis added].
69 This distinction was ennunciated in the landmark human rights code case of Ontario (Human
Rights Commission) v. Simpson-Sears ([1985] 2 S.C.R. 536 at 551, 23 D.L.R. (4th) 321) and was
subsequently acknowledged in Andrews (supra note 18 at para. 37), the Court’s first section 15 case.
70 [19991 3 S.C.R. 3, 176 D.L.R. (4th) I [BCGSEUcited to S.C.R.].
71 Ibid. at paras. 40-41.
2003]
N. KIM & T PIPER – GOSSELIN V. QUEBEC
The section 15 analysis, above all, is aimed at determining whether, intentionally
or by adverse effect, discrimination has occurred. Granted, because legislation, rather
than the actions of individuals, is more often at issue, it can perhaps be difficult to
speak of a discriminatory “intent”.72 This may have contributed to McLachlin C.J.
translating this conceptual step of classifying the type of discrimination into an
evaluation of legislative purpose in Gosselin, losing the focus on the discriminatory
aspect of the purpose in the process. Although the majority still addressed effects, the
lack of an overtly malevolent legislative purpose appeared to counterbalance any
adverse effects rather than act as an alternative.” For example, in evaluating the
social assistance scheme at issue, the majority, despite interspersed comments to the
contrary, clearly evaluated purpose in a manner beyond determining whether the
scheme, by purpose, design, or intent, was discriminatory:
I emphasize
that a beneficent purpose will not shield an otherwise
discriminatory distinction from judicial scrutiny under s. 15(1). Legislative
purpose is relevant only insofar as it relates to whether or not a reasonable
person in the claimant’s position would feel that a challenged distinction
harmed her dignity. As a matter of common sense, if a law is designed to
promote the claimant’s long-term autonomy and self-sufficiency, a reasonable
person in the claimant’s position would be less likely to view it as an assault on
her inherent human dignity…. However, where the legislature is responding to
certain concerns, and where those concerns appear to be well founded, it is
legitimate to consider the legislature’s purpose as part of the overall contextual
evaluation of a challenged distinction from the claimant’s perspective, as called
for in Law. This is reflected in the questions Iacobucci J. asked in Law: “Do the
impugned CPP provisions, in purpose or effect, violate essential human dignity
and freedom through the imposition of disadvantage, stereotyping, or political
or social prejudice?”; “Does the law, in purpose or effect, perpetuate the view
that people under 45 are less capable or less worthy of recognition or value as
human beings or as members of society?” 74
Arguably, consistent with the legacy of the human rights jurisprudence of the Court,
what was meant in the test quoted from Law, is to determine whether the provisions,
in intent or effect, are a violation of substantive equality. This inquiry is qualitatively
different from the approach taken by the majority. The latter judgment instead applied
an altered criterion of “legislative purpose” and thus found that the government’s
purpose was beneficial to the claimant group and reasonable in the circumstances.
72 See also R. v. Big MDrug Mart Ltd., [1985] 1 S.C.R. 259, 18 D.L.R. (4th) 321.
71 Interestingly, this is a similar approach to the one taken by McLachlin J.A. (as she then was) in
her Court of Appeal judgment in Andrews v. Law Society of British Columbia ((1986) 2 B.C.L.R. (2d)
305, 27 D.L.R. (4th) 600), which was subsequently overturned (Andrews, supra note 18 at 181-82):
I would reject, as well, the approach adopted by McLachlin J.A. She seeks to define
discrimination under s. 15(1) as an unjustifiable or unreasonable distinction. In so doing
she avoids the mere distinction test but also makes a radical departure from the
analytical approach to the Charter which has been approved by this Court. In the result,
the determination would be made under s. 15(l) and virtually no role would be left for s. 1.
74 Gosselin, supra note I at para. 27 [emphasis added]
MCGILL LAW JOURNAL / REVUE DE DROITDE MCGILL
[Vol. 48
The majority’s comments at the section 15 stage were reminiscent of a section 1
proportionality analysis, notably imbued with considerable legislative deference:
Instead of turning a blind eye to these problems, the government sought to
tackle them at their roots, designing social assistance measures that might help
welfare recipients achieve long-term autonomy … Even if one does not agree
with the reasoning of the legislature or with its priorities, one cannot argue
based on this record that the legislature’s purpose lacked sufficient foundation
in reality and common sense to fall within the bounds of permissible discretion
in establishing and fire-tuning a complex social assistance scheme. Logic and
common sense support the legislature’s decision to structure its social assistance
programs to give young people, who have a greater potential for long-term
insertion into the work force than older people, the incentive to participate in
programs specifically designed to provide them with training and experience …
Aside from distorting the section 15 analysis, these conclusions are problematic
in two substantive ways: they ignore the majority’s own warning that paternalistic
purposes for a group’s “own good” will still be suspect,76 and they fail to incorporate
the lesson learned in BCGSEU that a blunt, bifurcated examination of intent and
effect can leave a discriminatory standard or norm unquestioned.
The majority, and to a certain extent Bastarache J.,7 failed to recognize that good
intentions, when based on presumed or stereotypical characteristics of a group, can
still be discriminatory. 7’ The idea that younger people, simply because they are
young, are more capable than those thirty and over of finding employment if they
only make the effort to do so, is such an unsupported assumption. 79 This assumption
effectively acts as an unquestioned standard that young people receiving social
assistance are required to meet, or, in other words, a stereotype incorporated into the
analysis to justify the finding of no discrimination. Although not demeaning in the
strict sense, it is still an arbitrary generalization. Despite having recognized earlier in
for
its
disproportionately high rates of youth unemployment, the majority used “everyday
experience and common sense” to validate the generalization that “younger people
may have an easier time finding employment than older people.”‘8 By using the
the
government’s non-malicious
these
discriminatory assumptions underlying
assumptions to be perpetuated, endorsed, and left unquestioned.
the market conditions were
intent as an analytical
tool
for masking
that intent,
the majority allows
judgment
that
largely
responsible
” Ibid. at para. 44.
76 Ibid. at para. 27.
77 Ibid. at para. 243 (only a “detrimental” purpose is to be considered).
78 See also ibid. at para. 112, L’Heureux-DubM J.
79 See ibid. at paras. 403-10, Lebel J.
‘o Ibid. at para. 56.
2003]
N. KIM & T PIPER- GOSSELIN V. QUEBEC
3. Lovelace: Affirmative Action and Ameliorative Purpose
The third contextual factor of the third part of the section 15 Law test is the
“ameliorative purpose or effects of impugned legislation or other state action upon a
more disadvantaged person or group in society.”‘” As alluded to in Law and implicitly
confirmed in Lovelace, the third factor incorporates into section 15(1) the concerns of
section 15(2) of the Charter that affirmative action initiatives, which have the
furtherance of substantive equality at their foundation, should not be undermined by
claims of discrimination by more advantaged groups.82 Without precluding the
in Lovelace clarified that
independent operation of section 15(2), the Court
ameliorative programs or legislation should be “confirmatory and supplementary” to
section 15(1) and should not act as a defence or exemption to a violation of equality
rights.83
In Gosselin, the majority found the third factor in the dignity analysis to be
neutral since the legislation was not designed to improve the position of another less
advantaged group. At the same time, however, it considered the incentive scheme for
younger recipients to be aimed at ameliorating their situation and thus a factor for
assessing whether the scheme affected their human dignity. ”
There are three errors in this reasoning. First, as pointed out by L’Heureux-Dub6
J. and apparent from the Law and Lovelace precedents, “the ameliorative purpose
must be for the benefit of a group less advantaged than the one targeted by the
impugned distinction.”” Second, by acknowledging that the legislative purpose was
ameliorative or by portraying the education programs as some sort of affirmative
action for younger recipients, the majority must have implicitly recognized that the
group suffered from a pre-existing disadvantage. Yet, the majority found that the
recipients in the younger group did not suffer from any pre-existing disadvantage
under the first contextual factor of the Law dignity test.
Lastly, this reasoning reflects a recurring discrepancy in the majority’s treatment
of the legislative scheme. The legislative distinction is portrayed in the judgment as
the requirement that those under thirty participate in programs in order to raise their
level of assistance. They were distinguished from those thirty and over who did not
have to participate to raise their levels of assistance. It is arguable, however, that the
distinction at issue, quite apart from the programs available, was simply that of the
differential in rates: social assistance recipients under thirty received approximately
one-third the amount as those over thirty (170 dollars as compared with almost 500
in the construction of the Regulation
dollars). This distinction was reflected
respecting social aid, where the amounts for ordinary needs were established in
81 Law, supra note 5 at 539.
82Lovelace, supra note 50 at paras. 84-85, 93-108.
s31bid at para. 105.
4 Gosselin, supra note 1 at paras. 59-62, 65-66.
85 bid at para. 136.
MCGILL LAW JOURNAL/ REVUE DEDROITDE MCGILL
[Vol. 48
section 23 as a basic scale and the amounts for single individuals under thirty
specified in section 29 as an exception to the general rule. Similarly, these amounts
were set out in Section III of the regulation, which determined “Ordinary Needs”,
while the provisions with regard to the programs were provided for separately in
Section IV under “Special Needs”.86 In addition, as noted in the dissenting judgments,
(1) participation in programs (and the concomitant increase in assistance) was not
limited to those under thirty, (2) access to programs was not universal to the under
thirty group, and (3) there was no possible way for participation to consistently raise
the level of assistance for those under thirty to that of the thirty and over group. Thus,
participation in the programs should have been seen as an extra hurdle or burden for
the younger recipients and, if considered in the section 15(1) rather than the section 1
analysis at all, should have been regarded as a disadvantage rather than an advantage.
legislative distinction as the creation of an
“incentive-based” system 7 for one group as opposed to the other, the majority was
able to place an ameliorative spin on what was essentially an asymmetrical workfare
system. Perhaps this is because the majority, and the dissents, proceeded directly to
the third part of the Law test without a directed inquiry at the first stage into what
distinction was at issue. Unfortunately, this shortcut allowed an inappropriately
defined distinction to taint the whole analysis.
By portraying
the
impugned
C. Intersectionality
As mentioned, all the judges in Gosselin proceeded directly to the third step of
the Law section 15 test. We argue above that the failure to accurately define the
impugned distinction at the first step of the test thus distorted the majority’s analysis
in the third step. In this section, we inquire as to whether the lack of directed inquiry
into the second step, “whether one or more enumerated or analogous grounds of
discrimination are the basis for the differential treatment,”88 similarly resulted in a
tainted third step. We conclude that it did by failing to incorporate the level of
contextuality that recognizes the manner in which intersections of identity markers
can contribute to discrimination.
1.
Intersectionality in Past Supreme Court Judgments
Since the inaugural section 15 case of Andrews, an “enumerated or analogous
grounds” approach has been the analysis favoured over other formulations for
determining discrimination, such as the formalistic “similarly situated” or “likes
86 Regulation respecting social aid, supra note 2, ss. 23, 29.
87 Interestingly, there were already other “incentives” in the legislative scheme directly related to the
rates which appear to be designed to encourage employability in the manner conceived of by the
majority. This included a 50 dollar reduction in assistance for six months for those who refused or
abandoned employment. See Regulation respecting social aid, ibid., s. 14.
88 Law, supra note 5 at para. 88.
2003]
N. KM & T PIPER – GOSSELIN V. QUEBEC
alike” tests for equality. 9 Unfortunately, cases that followed Andrews have illustrated
that the grounds approach is as susceptible to being reduced to a game of
categorization as the similarly situated test, especially when courts are faced with
addressing intersecting grounds of discrimination or multiple discrimination. In
Mossop v. Canada,”‘ for example, the majority could not conceive of family status as
incorporating a same-sex family and encouraged Mr. Mossop to frame his claim
under sexual orientation only.9
In Symes v. Canada, the majority could not
accommodate the idea of inequalities between women suffering from different types
or levels of disadvantage; rather, a claim based on sex required the same disadvantage
to all women equally. 2 A formalistically applied grounds approach has thus resulted
in claimants whose identities may traverse more than one of the discrete boxes of
identity vying for recognition between watertight judicially defined compartments.93
In the Law synthesis, the Court appeared to try to address this rigidity by
acknowledging a more holistic approach:
[I]t is open to a claimant to articulate a discrimination claim under more than
one of the enumerated and analogous grounds. … If the court determines that
recognition of a ground or confluence of grounds as analogous would serve to
advance the fundamental purpose of s. 15(1), the ground or grounds will then
be so recognized.
There is no reason in principle, therefore, why a discrimination claim
positing an intersection of grounds cannot be understood as analogous to, or as
a synthesis of, the grounds listed in s. 15(1). 94
This statement was applied in Corbiere v. Canada (Minister of Indian and Northern
Affairs)” where the Court recognized “Aboriginality-residence” ‘ or “off-reserve
band member status”’97 as an analogous ground, conforming with the “central concept
of immutable or constructively immutable personal characteristics, which too often
have served as
illegitimate and demeaning proxies for merit-based decision
making.” 5 L’Heureux-Dub J., in her reasons, noted that the characteristics that may
comprise analogous grounds should be considered in a fluid and contextual manner
‘9 Andrews, supra note 18 at 168 [references omitted].
90 Mossop, supra note 18.
91 See Nitya lyer, “Categorical Denials: Equality Rights and the Shaping of Social Identity” (1993)
19 Queen’s LJ. 179, discussing the problems of falling through the cracks or pushing others through
the cracks in the context of Mossop and Symes v. Canada ([1993] 4 S.C.R. 695, 110 D.L.R. (4th) 470
[Symes cited to S.C.R.]). See also Nitya Duclos, “Disappearing Women: Racial Minority Women in
Human Rights Cases” (1993) 6 CJ.W.L. 25.
92 Ibid. See also Audrey Macklin, “Symes v. MN.R.: Where Sex Meets Class” (1992) 5 CJ.W.L.
498.
93 Iyer, supra note 91 at 193-203.
” Law, supra note 5 at paras. 93-94 [notes omitted].
9’ [199932 S.C.R 203, 173 D.L.R. (4th) 1 [Corbiere].
96Ibid. at para. 6, McLachlin CJ. and Bastarache J. for five justices, concurring in the result.
97 Ibid at para. 62, L’Heureux-Dub J. for four justices.
9’Ibid. at para. 13, McLachlin C.J. and Bastarache J.
MCGILL LAW JOURNAL / REVUE DE DROITDE MCGILL
[Vol. 48
so that concerns of overlapping and intersecting discrimination can be addressed as
they arise:
I should also note that if indicia of an analogous ground are not present in
general, or among a certain group in Canadian society, they may nevertheless
be present in another social or legislative context, within a different group in
Canadian society, or in a given geographic area, to give only a few examples….
The second stage must therefore be flexible enough to adapt to stereotyping,
prejudice, or denials of human dignity and worth that might occur in specific
ways for specific groups of people, to recognize that personal characteristics
may overlap or intersect (such as race, band membership, and place of
residence in this case), and to reflect changing social phenomena or new or
different forms of stereotyping or prejudice.99
McLachlimi C.J. and Bastarche J. for the other five justices took issue with this
approach, arguing that it endorsed shifting analogous grounds.” It appears, however,
that L’Heureux-Dube J. was simply stating that we should not foreclose on new
analogous grounds that might encompass previously rejected grounds arising in new
contexts and intersecting with other characteristics. Although McLachlin CI. and
Bastarache J.’s point that grounds should be permanent suspect markers of
discrimination is not contrary to this, their opposition and insistence that it was only
the third part of the test-whether discrimination exists-that would respond to
differences in claims reflects an analytical rigidity which is reproduced in Gosselin.’0
2.
Intersectionality in Gosselin
The judicial drive to categorize, and more specifically, to work within
traditionally recognized categories, is apparent in Gosselin, where every judgment
unquestioningly accepted that the impugned distinction was based on the enumerated
ground of age. To be fair, this was how the claimant framed her section 15 claim. We
would argue, however, that this is, if not wrong, then inaccurate at the least and
creates four problematic issues for the analysis in this case.
First, recall that the first part of the Law test requires two alternative, but
complementary inquiries: “Does the impugned law (a) draw a formal distinction
between the claimant and others on the basis of one or more personal characteristics,
or (b) fail to take into account the claimant’s already disadvantaged position within
in substantively differential treatment between the
Canadian society resulting
claimant and others on the basis of one or more personal characteristics?”‘ 2 Clearly,
under the Regulation respecting social aid, the formal distinction was based on age:
distinguishing between those under thirty and those over thirty. However, the second
part of the inquiry-regarding a failure to account for “the claimant’s already
99Ibid at para. 61.
’00 Ibid at para. 10.
101 Ibid. at paras. 9, 108.
102 Law, supra note 5 at para. 88.
2003]
N. KIM & T PIPER – GOSSELIN V. QUEBEC
773
disadvantaged position … resulting in substantively different treatment … “03
is
equally applicable to this case. In particular, the claimant’s status as either a poor
person, an unemployed person, a recipient of social assistance, or all three, places her
in a position of disadvantage under the regulation. This disadvantage is exacerbated
by the substantive effects of the difference in social assistance rates. When we then
proceed to the second part of the Law test-whether the differential treatment is
based on an analogous ground-can a contextual and purposive analysis really
exclude consideration of the claimant’s socio-economic status? Although the test is
defined as determining the ground of the distinction, there can be little doubt that
in section 15(1) are prohibited grounds of
what is sought to be enumerated
discrimination, and not merely any formal ground on which distinctions are made.”
regulation of personal
invasions of privacy,”0 5
In substantive terms, although the legislative distinction in Gosselin was formally
one of age, it was the claimant’s socio-economic status and her dependence on social
assistance that made this distinction possible. It has been recognized that the poor,
and especially those on social assistance, are disproportionately susceptible to state-
and
sanctioned
discrimination.” By summarily deciding, as the majority did in Gosselin, that the
prohibited ground is one of age alone, there can only be a fragmented and partial
understanding of how age and socio-economic status interact at the heart of the
differential treatment in this case. The granting and withholding of resources for basic
human necessities should be the distinction at issue and such recognition of the role
of socio-economic status in a legislative age distinction would allow for a more
accurate, realistic, and contextual approach to a claim of discrimination. The
following comment made by former Justice La Forest as chair of the Canadian
Human Rights Act Review Panel, which recommended the inclusion of “social
condition” as a prohibited ground of discrimination to address the claims of
discrimination of those living in poverty, is apposite:
lifestyle,”
103 Ibid.
104 Dianne Pothier, “Connecting Grounds of Discrimination to Real People’s Real Experiences”
(2001) 13 C.J.W.L 37 (advocating a strengthened emphasis on grounds in examining section 15
claims).
(4th) 279,2001 SCC 89.
‘c5 See e.g. Glasgow v. Nova Scotia (Minister of Community Services) (1999), 178 N.S.R. (2d) 115,
178 D.L.R. (4th) 181. A similar reading could be inferred from the result in Re Privacy Act (Can.),
[2001] 3 S.C.R. 905,210 D.L.R1
106 Falkiner v. Ontario (Minister of Community and Social Services) (2002), 59 O.R. (3d) 481, 212
D.L.R. (4th) 633 (C.A.) and other cases where the “spouse in the house” rule affecting social
assistance has been found unconstitutional.
107 See e.g. Dartmouth/Halifax County Regional Housing Authority v. Sparks (1993), 119 N.S.R.
(2d) 91, 101 D.L.R. (4th) 224 (S.C.); Sheilagh Turkington, “A Proposal to Amend the Ontario Human
Rights Code: Recognizing Povertyism” (1993) 9 J.L. & Social Pol’y 134; Martha Jackman,
“Constitutional Contact with the Disparities in the World: Poverty as a Prohibited Ground of
Discrimination Under the Canadian Charter and Human Rights Law” (1994) 2 Rev. Const Stud. 76;
Canadian Human Rights Act (“CHRA”) Review Panel, Promoting Equality: A New Vision (Ottawa:
CHRA Review Panel, 2000) (Chair: Gfrard V. La Forest) [CHRA Review Panel].
MCGILL LAW JOURNAL / REVUE DE DROITDE MCGILL
[Vol. 48
Some barriers related to poverty could be challenged on one or more of the
existing grounds. However, these cases have rarely been successful. They are
difficult to prove because they do not challenge the discrimination directly …
[If a policy or practice adversely affects all poor people or all people with a
low level of education, a ground-by-ground consideration of the issue can be
seen as a piecemeal solution that fails to take into account the cumulative effect
of the problem….
Second, while both McLachlin C.J. and Bastarache J. focussed on case-specific
differences in the third (rather than the second) part of the Law analysis, as they
in Corbiere, the judgments reached diametrically different
jointly advocated
conclusions. The difference can be attributed partly to McLachlin C.J. who, having
assumed age to be the enumerated and applicable ground, turned the discrimination
analysis into a marginally elevated similarly-situated test:
Given the lack of pre-existing disadvantage experienced by young adults, Ms.
Gosselin attempts to shift the focus from age to welfare, arguing that all welfare
recipients suffer from stereotyping and vulnerability. However, this argument
does not assist her claim. The ground of discrimination upon which she founds
her claim is age. … Re-defining the group as welfare recipients aged 18 to 30
does not help us answer that question, in particular because the 30-and-over
group that Ms. Gosselin asks us to use as a basis of comparison also consists
entirely of welfare recipients.l19
Although the majority was willing to compare those over thirty to those under thirty
at the level of generality in determining pre-existing disadvantage, it refused to
evaluate differences between social assistance recipients because they may have
constituted a disadvantaged group as a whole. This “minus one” approach to
evaluating discrimination is rigid and unrealistic. It only allows single deviations
from the societal norm: young versus old; affluent versus poor; employed versus
unemployed. This sort of dichotomous thinking can (and should) be avoided if a more
substantive analysis is undertaken in the earlier parts of the section 15 test.
too assumed
that age was
Though he came to a very different result, Bastarache J.’s judgment is similarly
problematic. He
the only applicable ground of
discrimination and thus considered receipt of social assistance as a contextual factor
under the third part of the Law test in determining pre-existing disadvantage:
The fact that their status as beneficiaries of social assistance was not argued as
constituting a new analogous ground should not be a matter of concern at this
stage of the analysis, since it has already been determined at the second stage of
the Law test that the differentiation has been made on the basis of an
enumerated ground. The issue, at this stage, is to determine whether, in the
context of this case, a differentiation based an enumerated ground is threatening
to the appellant’s human dignity. If the vulnerability of the appellant’s group as
108 CHRA Review Panel, ibid. at 108 [emphasis added).
109 Gosselin, supra note I at para. 35.
2003]
N. KIM & T PIPER – GOSSELIN V. QUEBEC
welfare recipients cannot be recognized at this stage, can we really be said to be
undertaking a contextual analysis?” 0
While the majority characterized poverty as the product of the lack of individual
effort to become employed, Bastarache J. situated poverty as a precursor of
discrimination, to be considered merely as another contextual element. In both cases,
poverty was understood as an externality, and neither of the justices was willing to
consider poverty as a possible ground, in itself or in combination with another
ground, so long as the second stage of the test could be based on an enumerated
ground.
to
to encourage a “race
This leads us to the third sign of analytical rigidity: while the Court forewarned
the bottom” for competing
in Lovelace not
us
disadvantages,”‘ the decision in Gosselin to focus on the single enumerated ground
of age only highlights the lacuna in the jurisprudence regarding intersectionality and
equality rights-particularly as they relate to discrimination claims based on socio-
economic status. There was evidence before the Court in Gosselin that women in
poverty were more susceptible to abuse, harassment, and sexual exploitation.”‘
Additionally, persons with disabilities, racialized persons, including Aboriginal
persons, and single parents disproportionately live in poverty.” 3 As we have seen, age
is also a marker of poverty for both youth and seniors. Similarly, geographical region
can also be a factor: a prime example being the particular situation in Quebec at issue
“n Ibid. at para. 238.
“‘ Lovelace, supra note 50 at para. 69. Bastarache J. reiterates this stance in Gosselin (ibid. at para.
237). Similarly, a contextual and sensitive inquiry at the grounds-definition stage should not mean a
substantially heightened burden for the claimant. The test, and the judges applying it, should be
receptive to new and complex claims without insisting on strict standards of causation between the
ground(s) and discrimination nor perfect proof as to the discreteness or insularity of the group.
We would also warn against using the analogous grounds analysis to create a proliferation of
detailed grounds in the name of responding to multiple discrimination when the grounds, or
components thereof, are encompassing facts rather than identity markers. We are currently witnessing
this in refugee law where proof of distinct elements of the refugee definition is required in order to
establish that the claimant forms a member of a “particular social group” (see e.g. Chantal Tie, “Sex,
Gender, and Refugee Protection in Canada under Bill C-11: Are Additional Protections Required in
Light of In re RA.?” Refuge 19:6 (August 2001) 54, online:
individual
discrimination analysis
classifications. Intersectionality analyses should allow for a nuanced and holistic examination, not a
requirement that the claimant repeatedly satisfy the test for a number of grounds in succession.
to prevent splintering analogous grounds
into smaller,
112 See Gosselin, supra note 1 (Factum of the Intervenor, National Association of Women and the
(“NAWL”), at paras. 5-9, online: PovNet
“1 See CHRA Review Panel, supra note 107 at 108; NAWL Factum, ibid. See also A. Wayne
MacKay, Tina Piper & Natasha Kim, “Social Condition as a Prohibited Ground of Discrimination
under the Canadian Human Rights Act” (Canadian Human Rights Act Review, 2000), online: Canada,
Department of Justice, Canadian Human Rights Act Review
MCGILL LAW JOURNAL/REVUE DEDROITDE MCGILL
[Vol. 48
in Gosselin. Clearly, a claim based on numerous characteristics should not
delegitimize or preclude the claims of those who suffer discrimination on fewer
grounds or on a single ground. At the same time, an additive or compounding
approach to the intersectionality of discrimination does not further the cause of
substantive equality but rather devolves it into a formalistic calculus.” 4 Unfortunately,
all of the judgments in Gosselin, for the most part, glossed over the intersectionality
aspects of the claim and opted instead for the more simplistic, but necessarily
incomplete approach of focusing on the single enumerated ground of age.
Finally, it should be noted that whereas flexibility is required for categorizing and
acknowledging intersectional grounds of discrimination, precision should be required
when limiting rights through judicial notice. This accords with the underlying
philosophy of equality analysis which posits a broad interpretation of rights and a
narrow interpretation of limitations to those rights.” 5 As two sides of a liberal and
purposive approach to equality, both serve the object of section 15 and were
undermined in this case. What is apparent is that a holistic analysis of multiple
discrimination in cases where socio-economic status is at issue will continue to face
challenges so long as stereotypes of those living in poverty persevere.
D. Stereotypes
People living in poverty or of low socio-economic status face a host of barriers
and discrimination. We will attempt to highlight some of the underpinning attitudes
and stereotypes that have slipped into the decision in Gosselin. The following
comment of the majority is one example:
Simply handing over a bigger welfare cheque would have done nothing to help
welfare recipients under 30 escape from unemployment and its potentially
devastating social and psychological consequences above and beyond the short-
term loss of income … A young person who relies on welfare during this
crucial initial period is denied those formative experiences which, for those
who successfully undertake the transition into the productive work force, lay
the foundation for economic self-sufficiency and autonomy, not to mention
self-esteem. The longer a young person stays on welfare, the more difficult it
becomes to integrate into the work force at a later time. In this way, reliance on
welfare can contribute to a vicious circle of inability to find work, despair, and
increasingly dismal prospects.” 6
This kind of statement fails to recognize that dependence on social assistance offers
neither a liveable existence nor a valued status in our society. It ignores that the effort
involved in simply surviving on only 170 dollars per month could be an all-
consuming job in itself. Daily trials would include finding enough food when access
114 See Iyer, supra note 91.
115 See e.g. McKinney, supra note 30 at 382-84, Wilson J., dissenting. See also Andrews, supra note
18; Hunter v. Southam, [1984] 2 S.C.R. 145, 11 D.L.R. (4th) 641.
116 Gosselin, supra note 1 at para. 43 [emphasis added].
2003]
N. KiM & T PIPER- GOSSELIN V. QUEBEC
to food banks is limited and restricted; finding reasonable accommodation when rents
are high, when landlords are unwilling to rent to social assistance recipients, and
when public housing is scarce; finding employment without the expected attire and
tools for job interviews; and maintaining employment, or even accepting promotion,
when the amount of any extra revenue or cost-saving measure is “clawed-back” by
social assistance as an offset to the deemed amount of needs.”17 A greater amount of
assistance or “a bigger welfare cheque” could, in fact, be more conducive to
employability”‘ because it would enable people to have a small measure of security
and time to assess their options and opportunities.” 9
In addition to failing to account for the simple realities of those living in poverty,
the paternalistic undertones of the passage above would seem to be based on
underlying stereotypes of the poor and the young as being unemployed by choice,
lack of motivation, or laziness. Clearly the majority did not intend to invoke
stereotypes, but its subtle assumptions (combined with the lack of proof or discussion
of their veracity) are reflective of the insidious discrimination faced by the poor in
society generally. Jean Swanson provides an evocative account of such
discrimination:
Somewhat surprisingly, moral explanatory accounts of poverty were more
common and powerfiully perceived causes of poverty: lack of responsibility,
effort or family skills were universally cited explanations … Most secure
participants [in a political focus testing study] see children as deserving and
their parents as less so [possibly unwitting agents of their children’s misfortune]
… Welfare recipients are seen in unremittingly negative terms by
the
economically secure. Vivid stereotypes [bingo, booze, etc.] reveal a range of
images of SARs [Social Assistance Recipients] from indolent and feeble to
instrumental abusers of the system. Few seem to reconcile these hostile images
of SARs as authors of their own misfortune with a parallel consensus that
endemic structural unemployment will be a fixed feature of the new
economy. 20
Such blatant contradictions between group characteristics and societal realities
are recurring indications that stereotypes are at play. As discussed above, this is
the majority judgment where evidence mitigating a finding of
reflected in
discrimination was cited (as to the unemployment rates of youth in Quebec at the
time), but subsequently disregarded in favour of “common sense” assumptions that
youth, if they had just tried hard enough, could have become “productive” members
“7 SocialAidAct, supra note 2, ss. 3, 12; Section VIII of the Regulation respecting social aid, supra
note 2.
” We hesitate, however, to endorse the amount given to those over-thirty as sufficient.
119 See e.g. NAWL Factum, supra note 112; Gosselin, supra note 1 (Factum of the Intervenor,
Charter Committee on Poverty Issues).
120 Part of a submission by Jean Swanson of End Legislated Poverty obtained from Human
Resources Development Canada through an Access to Information request and cited in CHRA
Review Panel, supra note 107 at 110 [all notes in square brackets except the first are in the CHRA
report].
MCGILL LAW JOURNAL/ REVUE DEDROITDE MCGILL
[Vol. 48
of society. There was no discussion that the scheme itself may have created or
perpetuated barriers to employment. For instance, the social assistance claw-back,
which is still a strong aspect of our current social aid schemes, was completely
ignored. Thus, it was open to the majority to freely assume that social assistance was
simply an income supplement:
[There is no] evidence of the actual income of under-30s who did not
participate; clearly “aid received” is not necessarily equivalent to “total
income”.12’
Even though receiving other income (in addition to the deemed amount of need to
which one may be entitled) would be contrary to the Regulation respecting social aid,
possibly even criminal, 2 and could fulfill the stereotype of “welfare cheats” abusing
the system, it was the assumption that it occured that was the basis for finding that no
discrimination existed.
Underlying all of these stereotypes-dishonesty, irresponsibility, and laziness, for
example-is the latent and lurking conception of social assistance as charity rather
than as a societal duty or individual right. The legacy of the “poor laws” of the
nineteenth century lingers on today.’23 So long as social assistance is conceived of as,
at best, the benevolent generosity of the majority, and at worst, stealing from the rich
to give to the poor, then the human dignity of those living in poverty or those
receiving social assistance will always be impaired.
The majority used the term “dignity” freely when supporting its judgment. The
concept of dignity, however, is inherently malleable and can be a vessel to be filled by
many different concepts, as has been discovered by many common law courts around
the world.’24 The majority’s, conception of dignity in Gosselin is particularly
challenging. References to the dignity of work and long-term self-sufficiency
regardless of whether it means living at home or being unable to survive demonstrate
a lack of consideration for the realities of the class before them: there is no discussion
of the “dignity” of being compelled to perform the work no one else wants for
minimum wage. There is little dignity in the stereotypical assumption that social
assistance recipients will not participate in work or training opportunities unless
forced through financial deprivation. Fundamentally, the workfare nature of the
121 Gosselin, supra note I at para. 51.
122 See the case of Kimberley Rogers who died while under house arrest for social assistance fraud.
The results of a coroner’s inquest into her death included a number of recommendations for changing
the operation of the social assistance scheme. Dawn Ontario: Disabled Women’s Network Ontario,
“Justice With Dignity: Committee to Remember Kimberly Rogers”, online: Dawn Ontario
12 See Turkington, supra note 107 at 157-64.
12 See e.g. National Coalition for Gay and Lesbian Equality v. Minister of Home Affairs, [2000] 2
S. AfM. L.R. 1, [2000] 1 B. Const. L.R. 39 (S. Affi. Const. Ct.). See further Errol P. Mendes, “Taking
Equality into the 21st Century: Establishing the Concept of Equal Human Dignity” (2000-2001) 12
N.J.C.L. 3. Compare Roger Gibbins, “How in the World Can You Contest Equal Human Dignity?”
(2000-2001) 12 N.J.C.L. 25.
2003]
N. KIM & T PIPER – GOSSELIN V. QUEBEC
Quebec legislation removes the choice to work and the right to be free from coercion
that should be central to human dignity’ 2
Quite apart from a question of whether a minimum level of assistance should be a
governmental obligation, discriminatory treatment within a social assistance scheme
is particularly egregious because the purported purpose underlying the scheme claims
to be highly complementary to that of equality provisions: to promote the equal
participation in our society of groups that may be particularly vulnerable to systemic,
attitudinal, and other barriers to the realization of their potential or goals as
individuals; to promote “a society in which all are secure in the knowledge that they
are recognized at law as equal human beings, equally capable, and equally
deserving.”126
Conclusion
Our discussion has traversed four major problems with the Supreme Court of
Canada’s section 15 analysis in Gosselin. Throughout we have noted the importance
of understanding the purpose of the section 15 equality guarantee and remembering
the rationale and precedents that underpin various legal tests, particularly the Law
test. We have noted as well the importance of strictly limiting measures that constrain
rights, broadly interpreting provisions that expand rights guarantees, and the need to
avoid stereotypes. It is from these problems and principles that we derive the
following recommendations for future equality cases.
First, we suggest the promulgation of and adherence to clear evidentiary
standards and guidelines for the evidentiary responsibilities of claimants in class
actions under section 15. We have considered how the evidentiary burden imposed on
Ms. Gosselin, despite prior judicial direction to the contrary, could prevent low-
income and other claimants
laws by requiring an
unnecessarily high standard. This barrier to access to justice is even more egregious
when the standard is hidden, shifting, and based on unsubstantiated stereotypes of the
claimant group.
from challenging unjust
125 Interestingly, the majority noted that one of the motivations for the implementation of a
“conditional” scheme was that section 15(3)(a) of the Canada Assistance Plan (R.S.C., c. C-i, as rep.
by Budget Implementation Act, 1995, S.C. 1995, c. 17, s. 32) did not make workfare to be compulsory.
Portrayed as “one of the major cornerstones of the social security system in Canada”, this portion of
the Canada Assistance Plan reflected the principle set forth in the first paragraph of article 6 of the
International Covenant on Economic, Social and Cultural Rights, 16 December 1966, 993 U.N.T.S. 3,
art. 6, Can. T.S. 1976 No. 46 (entered into force 3 January 1976, accession by Canada 19 August
1976) [emphasis added]: “The State Parties to the present Covenant recognize the right to work, which
includes the right of everyone to the opportunity to gain his living by work which he freely chooses or
accepts … “See Gosselin, supra note 1 at para. 44.
126 Egan, supra note 24. See also Candian Human Rights Act, R.S.C. 1985, c. H-6, s. 2; Act to
combat poverty andsocial exclusion, S.Q. 2002, c. 61, Preamble, s. 1.
780
MCGiLL LAW JOURNAL/ REVUE DE DROIT DE MCGLL
[Vol. 48
Second, precise and conservative use of judicial notice should be maintained for
facts in section 15(1) cases that operate to the detriment of the claimant. We
highlighted that the pernicious use of stereotypes and arguably improper use of
judicial notice may impede goals of substantive equality. It would be consistent with
a liberal and purposive interpretation of section 15(1) to approach with caution
unproven assumptions that could be founded upon stereotypes of the claimant group.
Furthermore, a flexible approach should be applied to those facts to be proven by a
claimant (showing a section 15(1) violation) when they are consistent with the bases
for judicial notice and logical reasoning.
Third, the section 15(1) test should be applied thoroughly, but flexibly and
contextually, to avoid fallacies in reasoning and dangerous analytical shortcuts. In
comparing Gosselin to past section 15(1) jurisprudence, we considered the major
differences between Gosselin and Law that mitigate against applying the Law analysis
wholesale to Gosselin. We also exposed the regressive steps taken by the majority in
its understanding and application of the concept of equality, particularly through the
use of “relevant” distinctions and “ameliorative purpose” to distort the purposive
application of section 15(1). Lastly, we noted that the failure to consider both the first
and second step of the Law test for discrimination resulted in a skewed analysis at the
third step. Although we would stop short of suggesting that detailed reasons should
be required for every step in every case, we would caution against the complete
omission of an inquiry at any of the stages of the test.
Our fourth recommendation concerns avoiding rigid categorization of claimants.
We critiqued the inadequacy of the majority and dissenting analyses in accounting for
the intersections of various identity markers which formed part of the claim. In
particular, the intersection of socio-economic status with other identity markers raises
challenges to the categorical traditions of the law, which have been (and continue to
be) barriers to the realization of substantive equality. A major step towards addressing
this problem would be to recognize an analogous ground such as social condition or
socio-economic status under the Charter. This ground would have to be defined
fluidly enough so as to capture the intersectionality of many claims which may
traverse narrow slots of identities.
Finally, we highlighted some of the underpinning assumptions about social
assistance recipients and people living in poverty that perpetuate stereotypes and act
as a barrier to the attainment of substantive equality. Clearly, we cannot impugn the
intent or good faith of the majority judges of Gosselin in this regard, however, their
evaluation of the effect of the Regulation respecting social aid on Ms. Gosselin’s
human dignity and the interspersed and implied stereotypes, which appear to be relied
on throughout the judgment, indicate that serious education may be required. One of
the hallmarks of systemic discrimination is the ability to cloak itself in “common
sense” and to erase the realities of those suffering from discrimination. Appointments
of judges from diverse backgrounds and judicial education exposing the barriers to
participation faced by those living in poverty are some of the effective measures that
may be taken in this regard.
2003]
N. KIM & T PIPER – GOSSELIN V. QUEBEC
781
Throughout this piece we have worked within the legal framework in which
Gosselin arose: the Law test, the section 15(1) jurisprudence, and the general rules of
interpretation. Our recommendations are not so bold as to challenge the existing law
or judicial tests for establishing the constitutional right to equality under the Charter.
After all, the majority’s reasoning -should be able to withstand the limits of its own
boundaries if it is to stand at all.
