The Women’s Court of Canada: Law v. Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497

TheCourt.ca is very pleased to reproduce the decisions of the Women’s Court of Canada. In 2004 this group of feminist/equality Charter activists, lawyers, and academics, decided to do something about what they saw as the sorry state of equality jurisprudence under s. 15. Their solution – rewrite the key decisions of the Supreme Court of Canada in this important area. The first six judgments of the Women’s Court of Canada have now been published in Volume 18 of the Canadian Journal of Women and the Law.

Over the coming months TheCourt.ca will have the honour of reproducing these judgments and providing them with a permanent online home. If you scroll down, you will find the logo of the Women’s Court of Canada on the right of your screen. It is hyperlinked. In future, by clicking on that link, readers will be able to view each of the Women’s Court judgments previously published at TheCourt.ca.

Author’s Note

Initially, I saw the exercise of writing a judgment in Law v. Canada (Minister of Employment and Immigration) merely as an opportunity to convert some previous academic work into a more ‘‘practical’’ form, partly to test how well my theoretical ideas about equality worked in dealing with the intricacies of a concrete case. Unlike many others involved in this enterprise, I have no prior involvement with the case, and I had no strong views about the correct outcome when I started. If anything, I thought that the Supreme Court of Canada had probably gotten the right answer. I was mainly concerned that the abstract test devised was not particularly illuminating and, for this reason, was open to abuse and obfuscation. I was also concerned that the widespread condemnation of the dignity element of the test laid out in Law has been too hasty and leaves us with no constructive avenue for engagement with the jurisprudence. I wanted to take a stab at giving more content to the idea of human dignity in a manner consistent with the art of judicial decision making. (I am not sure I was entirely successful at this latter objective—I fear that you may be able to take the girl out of the academy, but you cannot entirely take the academy out of the girl.)

In a word, I thought this was going to be a breeze. It was anything but, and I have both gained greater appreciation for the judicial enterprise and, I think, made my own views about equality more precise in the effort to grapple with Law. Writing the judgment has been a profound learning experience—both my understanding of the issues in the case and aspects of my general account of equality changed in the process. Bringing together theory and application deepened my understanding of each. Above all, writing this judgment has illustrated to me once again the truth of the feminist claim that taking gender into account makes a difference in the analysis of legal issues. I initially did not have much sympathy for Nancy Law’s claim. Yet, this being the Women’s Court of Canada, it seemed appropriate to consider the equality claim not only through the lens of age but also in light of the gender dynamics operating behind the scenes. For me, this transformed almost everything about the analysis of the concrete facts. I hope the demonstration of this transformation is evident to others.

The desire to deal with the gender implications of the age limitation on eligibility for the Canada Pension Plan survivor benefit introduces what some might think is an element of artificiality into the judgment. It means dealing with an argument that was never put to the lower courts and requires introducing new empirical material that was not before the courts. I have tried to mitigate the artificiality by drawing only on material that is publicly available through sources such as Statistics Canada. This raises a more profound issue, however. I do not know why counsel decided to argue the case exclusively on age grounds—perhaps way back in the early nineties they thought that the explicit use of an enumerated ground of discrimination was bound to be struck down. As it turned out, equality doctrine was a moving target all through the period during which the case moved through the system. Under these circumstances, it strikes me as unfair and unwise to stick closely to the original arguments when so much has changed in the interim. It also struck me as something of an abdication of responsibility not to go into the sex discrimination dimension of the case simply because the claim had not been seen in that light when the design of the survivor pension is so much bound up with the gendered conditions that govern women’s financial security. So I have taken some license to allow the Women’s Court of Canada to look more broadly at the issues raised by the case.

Given the symbolic status that Law has, both because the Supreme Court of Canada was intending to impose order on the doctrine and because of the influence of ‘‘the Law test’’ since 1999, many may read this Women’s Court of Canada decision looking for either a critique or endorsement of this test. I did not approach the case either as an opportunity to lay down the law or as a vehicle for addressing all of the concerns and criticisms that have been voiced about the Supreme Court of Canada’s decision. While I acknowledge the value of predictability in the law, I doubt that we are yet in a position to lay out a test for equality violations that is capable of dealing adequately with the range and complexity of the issues likely to arise. If this is true, the attempt to formulate a test risks unduly constraining the development of the law. So I have not framed my analysis as a direct response to the Supreme Court of Canada’s approach but, rather, “decided” the case as I would have done in the Supreme Court of Canada’s place.

Le jugement du Tribunal des Femmes du Canada dans l’arrê t Law c. Canada (Ministre de l’emploi et de l’immigration) retrace l’émergence de la notion de dignité humaine comme pierre angulaire de l’article 15 et passe en revue les facteurs pertinents pour conclure à une violation de cette dignité, énonceés dans la décision Law de 1999 de la Cour suprême du Canada. Ces facteurs sont réinterpreétés pour indiquer trois formes d’indignités que peuvent infliger une loi ou une politique: d’abord, fonder la loi ou la politique sur des préjugés; ensuite, utiliser des stéréotypes et enfin, exclure de l’accès à des bénéfices constitutifs de dignité. En analysant la disposition du Regime de pensions du Canada (RPC) limitant aux personnes a ˆgées de plus de 35 ans, l’admissibilité à la pension de la survivante ou du survivant à la suite du décè s d’un conjoint ayant contribué à ce régime, le Tribunal des Femmes est d’avis que les tribunaux inférieurs ont erré en traitant exclusivement des différences relatives à l’âge et que leurs jugements peignent, en conséquence, un tableau incomplet du programme et de sa raison d’être. En effet, la vaste majorité des récipiendaires de ces pensions sont des femmes. Cette partie du RPC ne peut se comprendre ni s’évaluer en occultant cette réalité de genre. Refuser toute aide aux personnes qui ont moins que l’âge prescrit a un impact disproportionné à l’égard des femmes de ce groupe d’âge.Law v. Canada (Minister of Employment and Immigration) traces the emergence of dignity as the touchstone of section 15 equality rights, and reviews the factors relevant to finding a violation of dignity from the Supreme Court of Canada’s 1999 decision in Law. These factors are reinterpreted as pointing to three forms of indignity that may be inflicted by law or policy: grounding law or policy in prejudice, the use of, or reliance on, stereotype, and exclusion from dignity-constituting benefits. In analyzing the Canada Pension Plan (CPP) limitation on eligibility for a spousal survivor’s pension to those over age thirty-five, the Women’s Court finds that the exclusive focus on the age-related differential treatment in the courts below provides an incomplete picture of the program and its rationale.CPP cannot be properly understood or evaluated absent this gendered context. Denying any assistance to those under a certain age differentially affects mainly women in that age group. Further, the assumption that losing a spouse causes no financial hardship to younger survivors, even in the short term, adopts a male norm by treating as typical of younger spouses those who are employed and fully self-sufficient. Given the roadblocks women typically encounter in their efforts to be self-sufficient, the financial dislocation due to a spouse’s death may easily be severe enough to impair women’s ability to live lives of dignity and be full participants in society. Accordingly, the Women’s Court reverses the Supreme Court decision, and holds that the CPP survivor pension scheme is discriminatory on the combined basis of age and sex. Based on the arguments put forward in the courts below, the legislation cannot be saved under section 1.

De plus, présumer que la perte d’un-e conjoint-e par des personnes plus jeunes ne crée aucune difficulté financie `re, me ˆme à court terme, c’est entériner une norme masculine, en vertu de laquelle la conjointe typique plus jeune est une personne employée et entièrement autonome. Étant donné les obstacles que rencontrent typiquement les femmes dans leurs efforts pour assurer leur autosuffisance, le bouleversement financier provoqué par le décès du conjoint peut facilement êtresérieux au point de porter atteinte à leur aptitude à vivre avec dignité et à participer pleinement à la société. C’est pourquoi le Tribunal des Femmes du Canada juge que les dispositions relatives aux pensions pour conjoint-e survivant-e sont discriminatoires quant au sexe et quant à l’âge. La loi ne peut être sauvegardée par l’article premier, eu égard aux arguments présentés devant les tribunaux inférieurs.

The judgment of the Women’s Court of Canada in

The vast majority of survivor pension recipients are women, and this part of the

Reconsideration of Law v. Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497 (judgment of the Supreme Court of Canada reversed).

The decision of the Women’s Court of Canada was delivered by:
DENISE REAUME

I. Introduction

1. This case concerns the constitutionality of sections 44(1)(d) and 58 of the Canada Pension Plan, R.S.C., 1985, c. C-8 (CPP), which uses age as a criterion of entitlement to the survivor’s pension. The issue is whether the provisions infringe section 15(1) of the Canadian Charter of Rights and Freedoms, Part 1 of the Constitution Act 1982, being Schedule B to the Canada Act, 1982 (U.K.), 1982, c. 11, and, if so, whether the infringement is justified under section 1 of the Charter.

2. This case came before the Supreme Court of Canada at a time of great turmoil in the development of section 15 jurisprudence. In three important cases in 1995, the Court had failed to achieve consensus about the central ingredients of the section 15 analysis. (Miron v. Trudel, [1995] 2 S.C.R. 418, Egan v. A-G. Canada, [1995] 2 S.C.R. 513, and Thibaudeau v. Canada, [1995] 2 S.C.R. 627.) In dealing with Ms. Law’s challenge to the CPP, the Supreme Court of Canada sought to rectify this situation and apparently succeeded in uniting the Court around a general approach to section 15 cases. Central to this endeavour is the identification of human dignity as an underlying value of section 15. An inquiry into whether a distinction violates human dignity, whether explicit or de facto, forms the third step of the test outlined by the Supreme Court of Canada.

3. While it has become increasingly evident in the case law that some substantive foundation is necessary for section 15, the concept of human dignity as invoked by the Supreme Court of Canada in its decision in Law lacks precision. It is in danger of being used as a cover for judicial timidity or incomprehension of the social and legal mechanisms that impose and reinforce disadvantage. When this happens, the result is to reinforce and entrench the derogatory and demeaning attitudes or effects imposed by the challenged legislation or policy instead of eradicating them. For this reason, the concept itself has attracted much criticism (see, for example, Sheilah Martin, “Balancing Individual Rights to Equality and Social Goods” (2001) 80 Canadian Bar Review 299 at 328-30; June Ross, “A Flawed Synthesis of the Law” (2000) 11 Constitutional Forum 74 at 83; and R. Gibbins, “How in the World Can You Contest Equal Human Dignity” (2000) 12 National Journal of Constitutional Law 25).

4. This reconsideration provides the Women’s Court of Canada an opportunity to re-frame the role of human dignity in the section 15 analysis so that it may help, rather than hinder, the articulation of a meaningful conception of substantive equality. The Supreme Court of Canada has attempted to organize a range of factors that play a role in finding discrimination in past cases and has loosely tied them to the concept of dignity. It has failed, however, to really grapple with the underlying value of human dignity.

5. In particular, the Supreme Court of Canada has not fully grasped the range of ways in which legislation or policy is capable of violating human dignity. For this reason, its test comes across as a mechanical checklist of factors to be looked for in any given case without an adequate understanding of why and when they are significant. This makes it easy for courts to pick and choose among the factors and deny equality claims without providing a clear and satisfactory justification. In the process, the constitutional guarantee of equality is in danger of being watered down.

6. Despite the gendered social circumstances to which the survivor benefit responds, this case was not argued as a sex discrimination case but rather was argued exclusively on the basis of age. A very important dimension of the constitutionality of these provisions was therefore not examined. Given the claimant’s failure to raise sex discrimination, and the absence of intervenors before the Court, the Supreme Court of Canada would have been wise to have appointed amicus curiae to present arguments about the gender implications of the progressive denial of benefits to those below age forty-five.

7. This decision seeks to make up for that oversight, drawing on publicly available data about women’s position in the workforce and the family. Indeed, attention to gender transforms the understanding of the equality issue in this case — the focus should not be primarily on the effects on surviving spouses under forty-five but also on the effects on younger women who survive their spouses. The grounds of age and sex are intricately intertwined, but the significance of the negative effects of exclusion from the survivor benefit flow largely from women’s subordinate status in the labour market. The failure of the legislature to take account of the specific conditions that women face in the workforce amounts to imposing a male norm on women under the age of forty-five. Expecting younger women to “bounce back” after the death of a spouse just as a man could, despite the greater obstacles that women face in the workforce, can have serious, long-term consequences for women’s ultimate financial security. The interests threatened are sufficiently vital to implicate human dignity.

II. Factual Context

8. The CPP is a compulsory social insurance scheme enacted in 1965 to provide contributors and their families with reasonable minimum levels of income upon the retirement, disability, or death of the wage earner (see House of Commons Debates, 26th Parl., 2nd Sess., vol. 6, 10 August 1964, at p. 6636). One aspect of the scheme is the survivor’s pension. This monthly benefit is paid to a surviving spouse whose deceased partner has made sufficient contributions to the CPP and who meets the eligibility criteria specified in section 44(1)(d), namely an age threshold, responsibility for dependent children, or disability.

9. A claimant who is over the age of forty-five at the time of the contributor’s death, or is maintaining dependent children of the deceased contributor, or is (or becomes) disabled is entitled to receive a full survivor’s pension. However, section 58 gradually reduces this pension for able-bodied surviving spouses without dependent children who are between the ages of thirty-five and forty-five. Able-bodied surviving spouses without dependent children who are under thirty-five at the time of the death of the contributor are precluded from receiving a survivor’s pension until they reach the age of sixty-five.

10. The appellant, Nancy Law, married Jason Law in 1980. Mr. Law died in 1990 at the age of fifty, having contributed to the CPP for twenty-two years. Ms. Law was only thirty years old. The couple had no children. They worked together in a family business. According to the Pension Appeals Board’s rendition of the facts, Ms. Law carried on the business after her husband’s death, but it was not as successful as it had been.

11. The appellant’s application to receive survivor’s benefits was denied because she was under thirty-five, not disabled, and had no dependent children. She unsuccessfully appealed to the Minister of National Health and Welfare, then took her case to the Pension Plan Review Tribunal, arguing that the age criterion violates section 15. The tribunal agreed, but a majority upheld the distinction under section 1. It is the appeal of this decision to the Pension Appeals Board, thence to the Federal Court, and ultimately to the Supreme Court of Canada that is before us for reconsideration.

III. Relevant Statutory and Constitutional Provisions

12. Canada Pension Plan, R.S.C., 1985, c. C-8

44. (1) Subject to this Part,
. . .
(d) a survivor’s pension shall be paid to the surviving spouse, as determined pursuant to this Act, of a deceased contributor who hasmade contributions for not less than the minimum qualifying period, if the surviving spouse

(i) has reached sixty-five years of age, or
(ii) in the case of a surviving spouse who has not reached sixty-five years of age,
(A) had at the time of the death of the contributor reached thirty-five years of age,
(B) was at the time of the death of the contributor a surviving spouse with dependent children, or
(C) is disabled;

…

58. (1) Subject to this section, a survivor’s pension payable to the surviving spouse of a contributor is a basic monthly amount as follows:
(a) in the case of a surviving spouse who has not reached sixty-five years of age and to whom no retirement pension is payable under this Act or a provincial pension plan, a basic monthly amount consisting of
(i) a flat rate benefit, calculated as provided in subsection (1.1), and
(ii) 37.5 per cent of the amount of the contributor’s retirement pension, calculated as provided in
subsection (3), reduced, unless the surviving spouse was at the time of the death of the contributor a surviving spouse with dependent children or unless he is disabled, by 1/120 for each month by which the age of the surviving spouse at the time of the death of the contributor is less than forty-five years, and reduced, if at any time after the death of the contributor the surviving spouse ceases to be
(iii) a surviving spouse with dependent children and is not at that time disabled, or
(iv) disabled and is not at that time a surviving spouse with dependent children, by 1/120 for
each month by which the age of the surviving spouse at that time is less than forty-five
years…

13. Canadian Charter of Rights and Freedoms

1. The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.

15. (1) Every individual is equal before and under the law and has the right to the equal protection and equal benefit of the law without discrimination and, in particular, without discrimination based on race, national or ethnic origin, colour, religion, sex, age or mental or physical disability.

IV. Judicial History

A. Pension Appeals Board (1995), C.E.B. & P.G.R. 8574
(1) Justice Douglas Rutherford (Justice Armand Dureault concurring)

14. Rutherford J. held that the denial of the survivor’s pension did not constitute discrimination on the basis of age under the Charter. Although many laws create legal distinctions, not all amount to discrimination within the meaning of section 15(1). He began by noting that age is not the sole criterion in determining eligibility for survivor’s benefits under the CPP, but, rather, is only one factor. It is a combination of age, healthful employability, and freedom from responsibility for dependent children that may lead to the exclusion of benefits. He held that, to the extent that age is a factor in the denial of benefits, sections 44(1)(d)(ii)(A) and 58 do not create the kind of distinction that has been characterized as “discrimination” in the constitutional sense.

15. Rutherford J. noted section 15(1)’s use as a means of protecting discrete and insular minorities and of shielding vulnerable groups against stigmatization, stereotyping, and prejudice. He found that none of the evidence suggested the appellant is a member of a group that suffers “discrimination” in these terms. Nor did he find that those excluded are treated differently from those who do receive survivor’s pension benefits on the basis of an irrelevant personal characteristic. Rather, Rutherford J. found that age is a relevant characteristic in determining relative need for survivor’s benefits. Accordingly, he concluded that, even though the impugned provisions draw a distinction based on age, this does not constitute discrimination within the meaning of section 15(1) of the Charter.

16. Although unnecessary in order to dispose of the appeal, Rutherford J. went on to find that if there was an infringement of section 15(1) of the Charter, it would be justified under  section 1. He acknowledged that the extension of benefits to widowers and the elimination of remarriage as a bar to continuing survivor’s benefits had diluted the original legislative objective, making it difficult for sections 44(1)(d)(ii)(A) and 58 to pass the justificatory test under section 1 of the Charter without being found vulnerable on one point or another. However, in his view, the complexity of the CPP, its status as an over-arching federal-provincial benefits system and its onerous amendment requirements justify deference to Parliament’s choice of measures.

(2) Justice of Appeal Francois Angers

17. Angers J.A. agreed with his colleagues’ reasons regarding discrimination on the basis of age but preferred not to comment on the effect of section 1 of the Charter.

B. Federal Court of Appeal (1996), 135 D.L.R. (4th) 293

18. Chief Justice Julius Isaac, delivering judgment on behalf of a unanimous court, was unconvinced that the Pension Appeals Board had committed a reviewable error. The court substantially agreed with the reasons of the board that neither section 44(1)(d) nor section 58 of the CPP infringes upon the appellant’s equality rights guaranteed by section 15(1) of the Charter. The Court of Appeal was also in substantial agreement with the majority opinion that, even if those provisions do infringe section 15(1) of the Charter, they constitute a reasonable limit under section 1. Accordingly, the appeal was dismissed.

C. Supreme Court of Canada, [1999] 1 S.C.R. 497

19. Justice Frank Iacobucci, delivering judgment for a unanimous court, dismissed the appeal and used the present case to provide a set of guidelines for courts in analyzing a discrimination claim under the Charter. He warned against confining analysis under section 15(1) to a fixed and rigid formula. A purposive and contextual approach to discrimination is preferable in order to realize the strong remedial purpose of the equality guarantee. He outlined three broad inquiries that a court should make in determining a discrimination claim under section 15(1), the first focusing on the search for differential treatment between the claimant and others in purpose or effect and the second looking for a connection to one or more enumerated or analogous grounds. The third inquiry asks whether the differential treatment discriminates by imposing a burden upon or withholding a benefit from the claimant in a manner that reflects the stereotypical application of presumed group or personal characteristics.

20. In discussing the purpose of section 15, Iacobucci J. held that the general purpose is to prevent the violation of essential human dignity and freedom through the imposition of disadvantage, stereotyping, or political or social prejudice and to promote a society in which all persons enjoy equal recognition at law as human beings or as members of Canadian society, equally capable and equally deserving of concern, respect, and consideration. He noted that the equality guarantee is a comparative concept, which ultimately requires a court to establish one or more relevant comparators. A variety of factors may be relevant in order to demonstrate that legislation demeans a claimant’s dignity. Iacobucci J. outlined four such contextual factors: (1) pre-existing disadvantage, stereotyping, prejudice, or vulnerability experienced by the individual or group at issue; (2) 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; (3) the ameliorative purpose or effects of the impugned law upon a more disadvantaged person or group in society; and (4) the nature and scope of the interest affected by the impugned law.

21. Applying these factors, Iacobucci J. found that although a clear distinction is drawn between the appellant and others on the basis of age, adults under the age of forty-five have not been consistently and routinely subjected to the sorts of discrimination faced by some of Canada’s discrete and insular minorities. Consequently, this distinction does not violate the human dignity of the appellant. The purpose of the impugned CPP provisions is not to remedy the immediate financial need experienced by widows and widowers but, rather, to enable older surviving spouses to meet their basic needs during the longer term. Although the law imposes a disadvantage on younger spouses, it is unlikely to be a substantive disadvantage, viewed in the longer term. Iacobucci J. also stated that the clear ameliorative purpose of the pension scheme for older surviving spouses is another factor supporting the view that the impugned CPP provisions do not violate essential human dignity.

22. Finally, Iacobucci J. held that legislation need not always correspond perfectly with social reality in order to comply with section 15(1) of the Charter. The fact that legislation is premised upon informed statistical generalizations that may not correspond perfectly with the long-term financial need of all surviving spouses does not affect the ultimate conclusion that the legislation is consonant with the human dignity and freedom of the appellant. Parliament is entitled, under these limited circumstances at least, to premise remedial legislation upon informed generalizations without running afoul of section 15(1) of the Charter and being required to justify its position under section 1.

23. Finding no violation of section 15, Iacobucci J. held that it was unnecessary to consider whether the provisions could be justified under section 1.

V. Analysis

A. Constitutional Equality

24. Canadian equality jurisprudence in the Charter era has been marked from the beginning by its rejection of a formal equality approach such as that of Albert Venn Dicey (Introduction to the Study of the Law of the Constitution, 8th edition (Holmes Beach, FL: Gaunt, 1996)), in favour of the pursuit of substantive equality (Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143).

25. Equality rights are the constitutional means of challenging the existing allocation of some benefit or burden, understood in the everyday sense of those terms. The challenge may be to the explicit basis for allocation or the de facto result of legislation or policy. Every allocation requires criteria that govern it. The criteria in a rule or policy provide a purely formal basis on which equality and inequality can be assessed—anyone who has not received the benefit but fulfils the criteria has not been treated equally. However, this conception of equality cannot exhaust the constitutional right to equal treatment and equal benefit of the law. To stop here would mean automatically accepting the criteria provided by the legislation or policy under challenge. This would treat equality as simply a matter of treating likes alike, while allowing the legislation’s own terms to determine what counts as alike. This narrow conception of equality would also provide no basis for going behind the formal criteria to evaluate the actual effects of a rule or policy. We agree with the Supreme Court of Canada that a formal conception of equality does not go far enough.

26. The jurisprudence under the Canadian Bill of Rights, S.C. 1960, c. 44, amply illustrates the shortcomings of a Dicean conception of constitutional equality. Treating like cases alike as defined by the legislature enabled the Supreme Court of Canada to find no inequality in depriving Indian women of their status upon marriage to a non-Indian provided that all Indian women who married out were equally deprived (A.G. Canada v. Lavell, [1974] S.C.R. 1349). Similarly, this approach justified restricting women whose paid employment was interrupted by childbirth to inferior unemployment benefits or none at all because all women in similar circumstances were treated the same (Bliss v. A.G. Canada, [1979] 1 S.C.R. 183).

27. At its most formal, the Dicean approach gave members of disadvantaged groups no protection against legislation that manufactured, reinforced, or exacerbated that very disadvantage. The desire to transcend this past was a major rallying point in the debates surrounding the introduction of
the Charter. We must not betray the hope with which Canadians in general, and disadvantaged groups in particular, embraced section 15.

28. A claim that substantive equality has been denied is a claim that some principle of entitlement wider in some respect than that used in the challenged law or policy is the appropriate criterion for allocation of the benefit in issue. In response, courts must determine, in a principled way, when state policy has adopted the wrong criteria of entitlement for some benefit and if and why the criteria should be widened and by how much. Although some caution may be wise in assessing existing criteria for distribution of benefits, as individual cases arise the courts must develop principles according to which legislative allocations can be assessed.

29. The pursuit of substantive equality requires attention to the actual conditions of life of the disadvantaged. Rules or policies creating or exacerbating inequalities, or perhaps simply not correcting background inequalities, should be changed. Allocative criteria should be altered when necessary to address these realities (see Kathy Lahey, “Feminist Theories of (In)Equality,” in Sheilah L. Martin and Kathleen Mahoney, eds., Equality and Judicial Neutrality (Toronto: Carswell, 1987), 83).

30. The equality jurisprudence of the Supreme Court of Canada to date has forced the realization that if legislative criteria cannot be accepted at face value—as they cannot—there must be some substantive value telling us which criteria are illegitimate and why. This will in turn guide the search for alternatives. This Court has settled on the interest in dignity as this underlying value. The process of naming dignity as the touchstone of equality has been laborious. The challenge of giving the concept meaningful content stands as perhaps the most significant one facing the courts in the coming years. Since some substantive interest or value must underpin section 15 if it is to do any work, the task of articulating that interest must be tackled.

31. The concept of dignity reminds us that all human beings have inherent worth deserving of respect. Yet our history has frequently included laws and practices denying the equal worth of some. Women, for example, have traditionally been treated as less valuable than men in virtually every domain of life—their needs neglected, their aspirations denied, their interests subordinated to those of men. These laws and practices have often been inflected with racism, ableism, and heterosexism to create exacerbated or new forms of hardship and disadvantage for some women. As a result, women as a group labour under inferior standards of living by most significant measures of quality of life (for the most recent survey of women’s position in Canadian society, see Women in Canada: A Gender-based Statistical Report, Catalogue no. 89-503-XPE, 5th edition (Ottawa: Statistics Canada, 2006)). Section 15 calls us to root out such practices and ensure that new forms of subordination do not arise. Respect for human dignity requires it.

32. The affirmation of human dignity grounds a universal entitlement to respect for dignity. The distribution of concrete benefits is therefore to be judged according to whether they are consistent with the respect each is owed equally. A concept of dignity must develop organically, as a product of grappling with the specific issues raised by the cases in accordance with our best critically reflective judgments about what is most important to people, as individuals, members of communities, and participants in the larger society. If protection of these fundamental interests is unequally distributed, human dignity is not equally respected. This process of developing the concept of dignity must attend to the lived experience of women and other disadvantaged groups to understand how their dignity has been, and continues to be, undermined and must be oriented to the policy initiatives necessary to uphold their equal worth.

B. The Emergence of Protecting Dignity as the Purpose of Section 15

33. The search for a substantive foundation for section 15 goes back to Andrews, supra, in which Justice William McIntyre pointed out (at 168) that not every legislative distinction is discriminatory. This insight gestures towards the need for criteria to distinguish discriminatory distinctions from non-discriminatory ones. He recognized the importance of the impact of legislation on those affected in identifying discrimination. However, since every piece of legislation has some impact that leaves some better off than others (just as all legislation distinguishes between classes of persons), we must still ask what kind of impact discriminates or inflicts a “real” disadvantage, and what kind does not?

34. Decisions in the section 15 analysis turn on two points — which personal characteristics are illegitimate bases for legislative distinctions and what kinds of disadvantaging effects constitute discrimination. The former requires an account of why the grounds enumerated in section 15 are potentially problematic bases for legislative distinctions or effects, which, in turn, will guide the addition of new analogous grounds. The latter requires an analysis of what harm we think discrimination does beyond mere differentiation itself. A dignity-based analysis has emerged from the case law to answer these questions.

Vol. 18 2006 15735. Justice Bertha Wilson first treated the protection of human dignity as part of the purpose of section 15 in McKinney v. The University of Guelph, [1990] 3 S.C.R. 229. However, the centrality of the concept of dignity began to emerge only in Miron v. Trudel [1995] 2 S.C.R. 418, and Egan v. A-G. Canada, [1995] 2 S.C.R. 513.

36. In Miron, the majority decision of Justice Beverley McLachlin illustrates how the type of distinction implicates dignity. The grounds listed in section 15 are often connected to stereotypes that falsely attribute negative attributes to groups. This violates dignity. To expand the list, we should look for other personal characteristics that are used to stereotype (at para. 147). McLachlin J. linked four factors identified in the case law as bases for recognizing a new ground to the “unifying principle” of preventing the violation of dignity: the historically disadvantaged status of a group, its minority status, the personal nature of the characteristic relied on by the legislature, and its immutable nature (at para. 149).

37. Justice Peter Cory, writing for the majority in Egan, invokes dignity differently, using it to identify a form of harm attracting section 15 attention. Denying recognition to same-sex relationships, he held, ‘‘may have a serious detrimental impact upon the sense of self-worth and dignity of members of a group because it stigmatizes them’’ (at paras 160-1). The corollary of this insight, drawn out in the Supreme Court of Canada decision before us, is that differentiation tied to an enumerated or analogous ground may not be discriminatory if it does not cause the kind of harm that implicates dignity.

38. Justice Claire L’Heureux-Dube has made most explicit the connection between dignity and equality (Egan, at para. 36). Her suggested two-part framework for section 15 examines both the nature of the group affected by the legislation and the particular interests affected. The various factors flagged throughout the case law—minority status, history of bias against a group, vulnerability, constructive immutability—are relevant to deciding whether a group was at risk of having its dignity violated. The nature of the concrete disadvantage imposed by the legislation determines whether it is significant enough to constitute a violation of dignity.

39. In Vriend v. Alberta, [1998] 1 S.C.R. 493, Cory J. identified section 15 as “essential to achieving the magnificent goal of equal dignity for all” (at para. 67). Cory J. treats dignity itself as the “good” or “benefit” that must be protected equally to satisfy section 15. This explains why formal equality, in the sense of faithfully following the rules as prescribed or applying uniform criteria to everyone, is insufficient. The criteria within the rule or policy may themselves fail to respect the dignity of all, and a uniform rule may exclude the needs or interests of some groups in a way that denies equal dignity. In either case, we must identify some deeper respect in which people must be treated equally in order to test the criteria for allocation. In the Supreme Court of Canada decision under reconsideration here, Iacobucci J. collected these previous efforts to connect equality and dignity, and attempted to elaborate.

[T]he purpose of s. 15(1) is to prevent the violation of essential human dignity and freedom through the imposition of disadvantage, stereotyping, or political or social prejudice, and to promote a society in which all persons enjoy equal recognition at law as human beings or as members of Canadian society, equally capable and equally deserving of concern, respect and consideration,,,

(Law, at para. 51).

On the meaning of dignity, Iacobucci J. had this to say:

There can be different conceptions of what human dignity means … [T]he equality guarantee in s. 15(1) is concerned with the realization of personal autonomy and self-determination. Human dignity means that an individual or group feels self-respect and self-worth. It is concerned with physical and psychological integrity and empowerment. Human dignity is harmed by unfair treatment premised upon personal traits or circumstances which do not relate to individual needs, capacities, or merits. It is enhanced by laws which are sensitive to the needs, capacities, and merits of different individuals, taking into account the context underlying their differences. Human dignity is harmed when individuals and groups are marginalized, ignored, or devalued, and is enhanced when laws recognize the full place of all individuals and groups within Canadian society. Human dignity within the meaning of the equality guarantee does not relate to the status or position of an individual in society per se, but rather concerns the manner in which a person legitimately feels when confronted with a particular law. Does the law treat him or her unfairly, taking into account all of the circumstances regarding the individuals affected and excluded by the law?

(at para. 53).

41. Iacobucci J. then reformulated the section 15 test into three steps: first, whether the law draws a formal distinction or fails to take into account the claimant’s already disadvantaged position; second, whether the differential treatment identified in the first step is tied to one or more enumerated or analogous grounds; third, whether that differential treatment constitutes discrimination “by imposing a burden upon or withholding a benefit from the claimant in a manner which reflects 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” (at para. 88).

42. Iacobucci J. collected from the past cases four categories of “contextual factors” helpful in identifying discrimination. These are said to demonstrate whether a legislative distinction or impact has the effect of demeaning dignity. The first of these is “pre-existing disadvantage, vulnerability, stereotyping, or prejudice experienced by the group” (at para. 63). Pre-existing disadvantage is linked to stereotyping — the use of inaccurate assumptions about the merits, capabilities, and worth of an individual or group that stigmatize members of a group (at para. 64). The second contextual factor capable of indicating a violation of dignity asks whether the egislative distinction properly reflects the actual needs, merits, and circumstances of the members of the group affected by it (at para. 69). The third factor identified by Iacobucci J. is that of whether the legislation seeks to ameliorate the situation of a more disadvantaged group. In such circumstances, the distinction used is unlikely to be found to violate dignity (at para. 72). Finally, Iacobucci J. adopted L’Heureux-Dube J.’s suggestion in Egan, supra, that a crucial factor in finding harm to dignity is the nature of the interest affected by the legislation. It is not the actual, concrete effect of the legislation that matters—the dollars and cents or the specific opportunity, benefit, or service denied — but rather whether imposing such a cost on the individual or group implicates dignity (at para. 75). The social and constitutional significance of the effects on the individual must be taken into account. Does the law or policy affect membership in a basic way, deny participation, or constitute a complete failure to recognize members of a particular group?
(at para. 74).

43. These factors are all presented as indicators of whether a legislative distinction violates human dignity, yet just how each is related to dignity remains unclear. Similarly, Iacobucci J. denies that the four factors are either necessary or sufficient conditions for the violation of dignity, yet there is a danger of their being used exactly this way. Without the relationship between the factors and dignity being more fully explained, it is the factors themselves that courts may tend to rely upon to decide cases.

44. Without a vibrant conception of dignity at their core, these factors are in danger of being used to reinvent old mistakes. For example, although the second factor in the Law test reaffirms the need to take difference into account in the design of rules, and thereby counsels against “one size fits all” rules that unfairly exclude or disadvantage vulnerable groups, it can collapse into the discredited “relevance” argument offered by Justices Gerard La Forest and Charles Gonthier in Egan, supra, and Miron, supra, which is, in turn, traceable back to Bliss, supra, and, ultimately, to Dicey. Similarly, the fourth factor requires that we develop a conception of what sorts of interests are sufficiently important that their impairment implicates dignity.

45. It is time to take this analysis a step further by articulating more carefully what dignity is as a constitutional value and how it has informed past cases and should inform the future development of equality law.

C. Defining Dignity

(1) Human Dignity as a Value

46. The concept of dignity has been part of the foundation of modern political and moral theories since the Enlightenment (see Herbert Spielberg, “Human Dignity: A Challenge to Contemporary Philosophy” (1971) 9 Philosophy Forum 39). Martha Nussbaum, for example, claims for dignity “broad cultural resonance and intuitive power” and universal relevance (Women and Human Development: The Capabilities Approach (Cambridge: Cambridge University Press, 2000) at 72).

47. Dignity refers to a somewhat ineffable quality that we ascribe equally to all human beings in virtue of which we accord them a special kind of worth. Nussbaum speaks of the “awe-inspiring something” that human beings have (at 73). As Nussbaum says, “[t]he core idea is that of the human being as a dignified free being who shapes his or her own life in cooperation and reciprocity with others, rather than being passively shaped or pushed around by the world in the manner of a ‘flock’ or ‘herd’ animal” (at 72). Human beings have intrinsic, incomparable, and indelible worth, simply as human beings (see Immanuel Kant, Groundwork of the Metaphysics of Moral, translated by H.L. Paton (New York: Harper and Row, 1964) at 96; Thomas E. Hill, Jr., Dignity and Practical Reason in Kant’s Moral Theory, Chapter 2, “Humanity as an End in Itself” (Ithaca: Cornell University Press, 1992) at 38-57; and Nussbaum, supra, at 73-4). This sense of moral worth is simply part of our conception of the person.

48. What grounds the attribution of inherent worth associated with the concept of dignity is, of course, an issue of enormous philosophical dispute. We can, however, identify some themes in the common currency of thought about dignity to develop a concept that finds significant resonance in the key elements of the section 15 jurisprudence and which can help push this jurisprudence forward.

49. Two aspects of human personality in virtue of which human beings are valued and worthy of respect should be highlighted because they illuminate the equality jurisprudence so far and form the starting point for further reflection. First, human beings are capable of a conception of the self. Thus, respect for identity is crucial to respect for dignity, and demeaning the identity of members of particular groups is a key form of indignity. Second, humans are engaged, in large and small ways, in a continuous enterprise of crafting a path or direction for their life. This can sometimes be an intensely personal undertaking, sometimes a matter of participation with others in collective endeavours. Either way, the importance to people of controlling the shape of their lives makes respect for the ability to participate in this shaping, and support for its flourishing, relevant to protecting dignity. These two aspects of personality are connected since one develops an identity partly through the life one participates in creating and a conception of the good partly in the context of one’s sense of the kind of person one is. Similarly, one’s sense of worth flows from others’ recognition of how significant aspects of identity and particular plans and ambitions combine to create a life.

50. The attribution of dignity calls for a particular moral response — respect for the intrinsic worth it signifies. The dignity inherent in human personality is dishonoured through a failure to show respect, through treating others as less than creatures of inherent worth. To enjoy the sense of worth connoted by the concept of dignity requires that a person be secure in her identity as an individual, including as a member of those communities with which she identifies. Dignity may be an inherent quality, but a subjective sense of self-worth is an empirical psychological and emotional state that has enormous implications for the quality of life. A person’s subjective sense of self as someone of worth is crucially tied to how she is treated by others (see John Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1972) at 62; and Charles Taylor, “The Politics of Recognition,” in Amy Gutman, ed., Multiculturalism and “The Politics of Recognition” (Princeton, NJ: Princeton University Press: 1992) at 25-6).

51. The law cannot guarantee that people actually feel the sense of worth to which they are entitled, but we can aspire to a state in which empirical realities strive to match inherent moral entitlements, in which at least state policy is not an obstacle to enjoying the subjective sense of self that is characteristic of those whose dignity is respected. At the same time, harm to dignity does not depend on the causing of actual psychological harm. Internalizing others’ lack of respect
is a common, but contingent, consequence of the violation of dignity. The core violation lies in the want of respect for the intrinsic worth of others, even if it fails to bring its targets down in their own estimation.

52. Security in one’s identity includes a sense of belonging in one’s society and of entitlement to participate in its institutions, opportunities, and endeavours (see Donna Greschner, “The Purpose of Canadian Equality Rights” (2002) 6 Review of Constitutional Studies 291). Exclusion from the kinds of practices and activities that constitute a social concept of citizenship in a given society marks the excluded as outsiders, as unworthy to share the forms of life that make our lives meaningful. Such exclusion also impairs people’s ability to craft the life of their choosing.

53. Exclusion can happen directly through criteria for participation in certain activities or institutions, but it can also happen through the denial of forms of support that make participation possible. Participation in social life in a complex modern society often has material and social pre-conditions. Without a decent education, or stable housing, or adequate income, other forms of participation are beyond imagining let alone realizing. For this reason, as Nussbaum argues, internal capabilities for distinctively human forms of functioning must be combined with “suitable external conditions for the exercise of the function” (Nussbaum, supra, at 84-5).

54. For most of our history, it has been those men who enjoy privilege along a variety of dimensions (race, religion, sexual orientation, able-bodiedness, for example) who have been the most able to aspire to the ideal of a dignified life. Their self-worth has been affirmed; their dreams and mbitions fostered by legal structures and social practices; their interests made the focal point of social planning; and their participation in social institutions facilitated. They shape and have been supported by the society to which they belong. Yet the concept of human dignity claims equal worth for all, calling legislatures to examine old attitudes and practices for partiality and requiring vigilance by courts to ensure that legal policy is remade in the image of all members of society.

55. By contrast with the ways in which privileged men’s dignity has been cherished, women have been subjected to derogatory stereotypes (and the gendered barriers they create), constrained opportunity, gendered violence, control over their reproduction, and disrespect for their role as mothers — all violations of human dignity because they impinge upon women’s self-respect and self-realization. Respect for the human dignity of women means state policies must not only refuse to tolerate but also work to counteract the kinds of conditions that have resulted in women’s relegation to inferior status in the past.

56. The confining and demoralizing effects of policies controlling women’s opportunities, relationships, and bodies are now widely acknowledged. It does not follow that all vestiges of these attitudes, and, in particular, their long-term effects, have been swept away. Wherever they linger, they undermine women’s dignity.

57. Equality claims deal with an unending array of goods and benefits to which there may be no specific entitlement as a matter of human rights. The task ahead is that of showing how the concept of dignity can help bridge the gap between a concrete benefit claimed and the human right to equality.

(2) Three Forms of Indignity

58. As noted earlier, the Supreme Court of Canada has elaborated a list of apparently independent, though overlapping, contextual factors said to reveal violations of dignity. These factors suggest three forms of indignity or denial of worth, which, if inflicted by a statute or state policy, violate section 15. Further forms may be identified over time. We can have no confidence so early in our constitutional history of rights adjudication that we understand all facets of equality and all forms of inequality. Two of these forms of indignity explain most of the cases to date: the first consists in grounding legislation or policy in prejudice and the second in the use of, or reliance on, stereotype.

59. A third form of indignity has been less well articulated by the Supreme Court of Canada to date. It involves exclusion from benefits, institutions, or opportunities access to which is integral to our notion of personhood or constitutes part of the conditions for a life with dignity. In grasping this form of indignity, it is the nature and importance of the benefit itself that makes its denial a violation of dignity rather than the attitudes, mistaken beliefs, or motivations that prompt the denial. When either prejudice or stereotype operates as well, the indignity of such denial is exacerbated.

60. Any of the three forms of indignity can be inflicted through the explicit use of a characteristic important to personal and group identity or through facially neutral standards that fail to take account of the diverse circumstances of various groups. These are simply alternate formal means of achieving the same effect. For this reason, a formal conception of equality must be rejected. To discover whether the legislation or policy in issue conveys a message of lesser worth, the various factors commonly invoked operate as diagnostic tools in discerning whether indignity, in whatever form, is present.

61. Since the courts have dealt most often with the first two forms of denial of worth, the factors identified as markers of the violation of dignity have tended to be ones that alert us to the presence of prejudice or stereotype or help us to contextualize their effects. They are relevant not in their own right but rather because and insofar as they help us detect prejudice and stereotype. This explains why, as the Supreme Court of Canada has often said, the various relevant factors are neither necessary nor sufficient conditions of a finding of discrimination. The presence of these features of context does not constitute discrimination. They simply help us “read” the implicit meaning of legislative distinctions imposing burdens or denying benefits to see whether it connotes the inferiority of a group.

(a) Prejudice

62. The wrong of prejudice is easily understood as a violation of dignity. A distinction based on prejudice connotes a belief in the inferiority of those targeted. One that reinforces the prejudices of others, even if unintended by the policy maker, has the same effect. This not only deprives those targeted of the concrete benefit at issue but, through doing so, it also treats them as being unworthy of basic human respect. Distinctions that target an important aspect of human identity for stigmatization, humiliation, and the excuse for deprivation violate dignity in this way. Prejudice works through the categorical and false attribution of negative worth to personal characteristics that are important aspects of identity. Such an assault on the sense of self of its victims constitutes the violation of dignity.

63. Personal identity has both an individual and a social dimension. The kinds of characteristics that people regard as important to their sense of self tend to be, at the same time, characteristics by which they define themselves as individuals and through which they identify as members of a group (see Greschner, supra, at 298). It should be remembered, of course, that individuals can identify with more than one such group. We know from our social and political history that it has tended to be precisely this aspect of identity that has often been targeted for contempt — individuals have been denied respect through the use of a characteristic identifying them as part of a group that is categorically devalued.

64. The enumerated grounds mark aspects of personal identity that are socially understood as aspects of self that are important to self-understanding, but nevertheless are often used to classify groups as unworthy. Our knowledge of how this has been done in the past can guide our reading of current behaviour to see if past practices and attitudes are being reinforced. History can also enable us to detect other, unenumerated, ways of categorizing that subject members of a group to contempt in a similar way. The sorts of factors often relied upon to recognize a new ground as analogous to those listed in section 15 operate as more concrete signposts.

65. For example, the characteristics that are important to one’s sense of self are typically, actually, or constructively immutable (see Andrews, supra, at 195; Miron, supra, at paras. 158-59 and 163; Benner v. Canada (Secretary of State), [1997] 1 S.C.R. 358 at para. 84; and Corbiere v. Canada (Minister of Indian and Northern Affairs), [1999] 2 S.C.R. 203 at para. 13). However, too much emphasis should not be placed on actual immutability. Prejudice is rationalized by the lie that it is some difference within others that justifies their derogatory treatment even though difference is often socially constructed. Actual human attributes are less important in understanding the dynamic of prejudice than what society makes of those attributes. If an attribute has been socially constructed as being inferior, there should be no onus on those so characterized to change themselves, even if they could.

66. Historical disadvantage has also often been referred to in the cases but without a very precise analysis of its relevance (see Andrews, supra, at 180; R. v. Turpin, [1989] 1 S.C.R. 1296 at 1333 (per Wilson J.); Miron, supra, at para. 158; and Egan, supra, at para. 176). One function that historical disadvantage serves is as an indicator, and result, of past prejudices previously translated into legal disabilities. The fact that a group is historically disadvantaged may indicate a long-standing failure by the legal system or social mores to extend equal concern and respect. The challenged law may perpetuate or further promote this failure. Likewise, status as a chronic minority is a warning signal because it renders a group vulnerable to such prejudice. It may also be the result of a history of prejudice.

67. However, these contextual factors are all merely guides meant to make us confront the typical features of social prejudice. They should not be mistaken for a formal test. They remind us to think about how prejudice operates as a social phenomenon and to use this knowledge in interpreting the rule or policy in issue.

68. Entrenched prejudice can unleash social forces that devalue members of particular groups even when those acting within the practices shaped by those social forces have no subjective desire to show contempt. The sorts of distinctions and denials that constitute an infringement of dignity are a matter of social meaning rather than subjective attitudes. To understand the violation of dignity in its entirety, we must look to the meaning of government action, and this must always be read in social context rather than merely treated as a matter of plumbing the content of an actor’s mind.

(b) Stereotype

69. Stereotypes are inaccurate generalizations, normally negative, about the characteristics or attributes of members of a group. They can usually be traced back to a time when social relations were based more overtly on contempt for the moral worth of the group. Positive stereotyping can exist too, although normally these form the binary opposite of a negative stereotype about another group. “Men are courageous” is the positive correlate of the negative attribution of timidity to women, for example. Positive stereotypes are usually problematic only when they help feed their negative counterpart.

70. Stereotypes partly stem from backhanded recognition that acting on prejudice is a violation of human dignity. Negative characteristics, such as low intelligence, laziness, suitability for menial pursuits, predisposition to criminality, avarice, vice, and so on, which are in fact distributed throughout the human race, are falsely attributed predominantly to members of a particular group. It is then the negative characteristic that becomes the focus of contempt so that the actor can rationalize his or her conduct by claiming that he or she is not prejudiced against members of a group per se but simply properly disapproving of certain negative traits.

71. Through such practices of rationalization, inaccurate stereotypes about the capacities, needs, or desires of members of a particular group can operate even if the legislators did not understand them as being grounded in animosity, and, when they do, they can carry forward ancient connotations of second-class status. The overt hostility may have come to be washed out of the picture with the passage of time or the ‘‘normalization’’ of such attitudes and the practices based on them, but the implication that those to whom the stereotype applies are less worthy than others remains.

72. Once this construction of a group has set in, others are likely to treat members disadvantageously out of an honest belief that this merely reflects their just desserts or even just because that is simply how everyone treats them, without ever reflecting on the insult involved. They may even understand their conduct, as with certain traditional sexist practices, as a generous effort to accommodate the “natural weaknesses” of the stereotyped group. However, neither the absence of contempt as a subjective matter nor well-meaning paternalism prevents the use of stereotype from violating dignity.

73. To be denied access to benefits or opportunities that are available to others on the basis of the false view that certain attributes render one’s group less worthy of those benefits or less capable of taking up those opportunities can scarcely fail to be experienced as demeaning. The message that such legislation sends is that members of this group are inferior or less capable, and such a message is likely, in turn, to reinforce social attitudes attributing inferiority to the group, especially if such exclusion has been a systemic feature of their experience. Their lives may reflect a lower level of accomplishment and worth than they are capable and desirous of, and this shortcoming is likely to be falsely attributed to them rather than to the conditions to which they are subjected. This creates a feedback loop leading to future treatment reinforcing a status of lesser worth.

74. Because of the connection between prejudice and stereotype, many warning signals of prejudice operate to flag possible stereotyping as well. In particular, historical disadvantage suffered by a group is a reason to look for lingering prejudice converted into stereotype. Since stereotypes can often construct a pervasively negative image of a particular group, they ground systemic patterns of disadvantage likely to leave an historical trail.

75. Inaccurate assumptions about the attributes or capacities of a group are the stuff of stereotype and may render use of a particular characteristic irrelevant to the legislative objective at hand. This explains why so many cases have turned on an assessment, within the section 15 analysis, of the relationship between the criterion used and the purported legislative objective (see Andrews, supra; Corbiere, supra; and Egan, supra). If the proffered objective is not itself discriminatory, but the criteria used to distribute the statute’s benefit do not serve it very well, it is a signal that the use of those criteria may indicate the implicit acceptance of derogatory stereotypes. There is a danger though. If the courts are not careful to scrutinize the proffered objectives of law or policy for traces of stereotype, looking merely to see whether the criteria used for allocation are relevant, the objectives will not only not catch the dignity violation but will also reinforce it.

c) Harm to Fundamental Interests/Denial of Dignity-Constituting Benefits

76. Prejudice and stereotype are core cases of impugning the moral worth of others. Their eradication is a necessary part of the landscape of equality. However, the absence of prejudice and stereotype does not exhaust the notion of respect for human dignity. A link has been made in the case law from the beginning between stereotyping and its tendency to undermine its victims’ “personal development” (Andrews, at 197). This insight connects dignity and autonomy.

77. Respect for dignity includes respect for agency as being fundamental to our notion of personhood. Prejudice and stereotyping can indeed compromise autonomy, but they are not the only means of doing so. It is the concern for autonomy that should be at the core of the analysis. If there are other means by which state policy can interfere with self-realization by denying benefits to some that are afforded to others, these too should be matters of concern for section 15.

78. Human beings are choosers and planners. We have projects and dreams, make commitments and attachments, and at least partly measure our sense of worth according to our ability to exercise our capacities and realize our dreams. Some goods, institutions, and opportunities are so important to
agency and its exercise that their denial fundamentally undermines autonomy. Likewise, in any given society, membership is partly defined by access to certain benefits. These can be seen as dignity-constituting benefits—benefits that have a social meaning and significance that partly define a dignified existence in our society.

79. The dignity-constituting nature of a benefit is the corollary of the fundamental nature of the interest at stake in its denial. Concern with whether legislation or policy “affects a basic aspect of full membership in Canadian society” or “restricts access to a fundamental social institution” suggests that these forms of participation are benefits crucial to a life with dignity (Law, supra, at para. 74 (quoting L’Heureux-Dube J. in Egan, supra, at paras. 63-4)). This can only be because of their importance to agency, namely to exercise control over the character of one’s life. Respect for autonomy is part of respect for the inherent worth of persons.

80. A reasonable level of physical security, education, housing, adequate health care, and opportunity to participate in the workforce are all obvious examples of such benefits. These are important to everyone since they are the preconditions for the pursuit of most other projects and plans. The significance of these benefits may be accentuated for those who are multiply disadvantaged. For example, access to social housing may be especially crucial to poor women because without the security of adequate housing women are especially vulnerable to sexual violence and exploitation, which further violate dignity. Just as importantly, the measures needed to secure these goods may vary depending on the social circumstances of different groups. For example, assuring women a reasonable level of physical security requires dealing with gender-specific threats such as domestic violence.

81. The violation of section 15 in Miron, supra, and Egan, supra, though partly arising out of the use of stereotypes, can also be understood in light of the importance to dignity of a symbolic benefit — the institution of marriage. The benefits that define marriage and its status itself are of fundamental importance in our society as it has developed. To exclude some couples arbitrarily from those benefits treats their relationships as less worthy. The harm to dignity is worse if the denial is tinged with prejudice or stereotype, but part of why these benefits affect human dignity, even when they are of small economic value, as in Egan, is the importance of the institution of marriage in our society.

82. Eldridge v. British Columbia (Attorney General), [1997] 3 S.C.R. 624, provides another example of a dignity-constituting benefit. It also makes clear that the fact of exclusion from certain benefits enjoyed by others, even in the absence of prejudice or stereotype, can violate dignity. In declaring the failure of hospitals to provide sign language interpreters for the deaf as being a violation of section 15, La Forest J. proclaimed: “A legal distinction need not be motivated by a desire to disadvantage an individual or group in order to violate section 15(1). It is sufficient if the effect of the legislation is to deny someone the equal protection or benefit of the law” (at para. 62). It is the failure to take account of the particular needs of persons with disabilities in the design of practices and its consequences for participation in key institutions or goods that constitute the discrimination because this very tendency to discount the needs of persons with disabilities conveys the implication that their interests do not count (see also Eaton v. Brant County Board of Education, [1997] 1 S.C.R. 241 at para. 67).

83. La Forest J.’s argument recognizes that some benefits or opportunities, some institutions or enterprises, are so fundamental that denying participation in them itself implies the lesser worth of those excluded (Eldridge, at para. 69-71). Medical care is a vitally important pre-condition of a life with dignity. Treatable illness or disease, left untreated, can impair in innumerable ways the quality of life and the ability to pursue projects. To put some at risk of receiving inadequate care is to imply their inferior worth — interpreters are necessary to give deaf patients the same quality of care that hearing patients receive.

84. In addition, without an interpreter a deaf patient is denied one of the core rights of personhood—the right to decide for herself what will be done to her person. To deny deaf patients the means of understanding the treatment proposed for them denies the very possibility of meaningful consent. It turns decisions about care over to doctors or family, which is a situation that no one who routinely enjoys autonomy over such decisions would tolerate. The insult to dignity is profound. The exercise of the power to give or refuse consent is itself an important social institution to which access is denied by the failure to provide sign interpreters.

85. One word of warning, however, about making concrete the connection between dignity and autonomy: the protection of dignity requires attention to the conditions under which choice is exercised. When members of a particular group are operating under constraints that deny meaningful choice, these conditions themselves violate dignity. It is the role of the legal system to counteract such constraint and the role of section 15 to ensure that it does. The mere fact that some choice is possible and agency is exercised does not necessarily mean that autonomy has been respected.

86. Working out which benefits and social institutions have the importance required to be designated a dignity constituting benefit will have to take place on a case-by-case basis, developing the few small steps already taken by the Supreme Court of Canada. The point here is not to provide a comprehensive account but merely to recognize the denial of such benefits as a distinct form of indignity that section 15 must guard against.

D. Dignity and Feelings

87. Seeing violations of dignity as a matter of the social meaning of certain distinctions reveals the error of treating the harm as a matter of harm to the feelings of those affected. Prejudice and stereotype do stigmatize and often humiliate. This connection makes it easy to slip into treating the harm to be protected against as a kind of emotional harm. However, this “subjectivizes” the nature of the interest in issue. Instead, the harm should be understood to inhere in the denial of respect per se. Harm to dignity is an independent objective harm, not a matter of hurt feelings. Feelings of worthlessness may be a common symptom of disrespectful treatment and relevant diagnostically, but the evil to be prevented or remedied is the attribution of unworthiness.

88. The viewpoint of those affected must be taken into account. A court must try to put itself in the shoes of members of this group rather than simply adopting the perspective of the legislator. At the same time, more than the affected group’s say-so is required. The court must be satisfied that the claimant’s interpretation captures the legislation’s import, whether intended or a side-effect of the contested provisions.

89. In other words, it is the responsibility of the court to determine whether the import of legislation or other governmental action is to attribute lesser worth to some group. While it is crucial for judges to make every effort to understand the perspective and life experience of the claimant(s), the appropriate standard for liability is not whether the legislation might reasonably make members of a group “feel” demeaned or devalued. We must not reduce the question of the relevant impact of legislation to its psychological effects and, instead, keep the focus on the meaning of dignity and its impairment.

90. At the same time, it bears repeating that a court must always be sensitive to the meaning that legislation has for those negatively affected by it in order to have any hope of not simply ratifying dominant, potentially oppressive understandings of social relations. The project of constructing the concept of dignity is an active normative one, requiring critical reflection on existing social relations. It cannot be reduced either to a question of the legislature’s intentions or to a sampling of popular opinion.

E. The Importance of Context

91. To parse the social meaning of legislation for violations of dignity requires an interpretive exercise that takes account of the entire social context within which the challenged legislation operates. This is true whether the indignity complained of is grounded in prejudice or stereotype or in the denial of a dignity-constituting benefit.

92. Neither prejudice nor stereotype typically operates in discrete, isolated circumstances. Instead, the wholesale classification of a group as unworthy of full moral status operates to subordinate members of the group in ways that stand to affect every corner of their lives and infect every attitude and predisposition towards them. The resulting web of restrictions and exclusions subjects members of the group to the pervasive message that the social meaning of the most intimate aspects of their personality is one of inferiority. The social relations that ensue give birth to the stereotypes that feed the next round of putative justifications for continued exclusion.

93. The effects of such pervasive and long-term crimping of the lives of the members of such a group are not easily undone. Even once the most egregious exclusions have been remedied, the fact of having been associated with a wide array of negative meanings in the past makes the group more vulnerable to continued devaluation even as a result of fairly minor exclusions. Pockets of negative meaning may remain here and there in the law and in practices long after the central institutions of discrimination have been removed. This explains La Forest J.’s effort in Andrews to explore how immigrants have been systematically excluded from the best employment opportunities throughout much of our past (Andrews, supra, at paras. 68-9). The apparently isolated use of citizenship as a criterion for admission to the legal profession must be read against this backdrop, and, in this context, it inevitably attributes lesser worth to the non-citizen.

94. Context is just as important when interpreting the meaning of a denial of fundamental benefits and opportunities. Understanding the short- and long-term effect of the denial, how it is related to other forms of participation in society, and how it is likely to affect others’ attitudes towards those excluded requires keen attention to broader social realities, not merely the immediate consequences of the exclusion at issue.

F. Application to the Case at Bar

(1) Introduction

95. It follows from our analysis that determining whether legislation or state policy constitutes discrimination and violates equality rights requires a holistic analysis. The differential treatment complained of, whether explicit or the result of facially neutral rules, must be examined to see whether its effect on the claimant is connected to a personal characteristic so as to evoke dignity concerns or whether it disadvantages some in respect of a fundamental interest or opportunity so as to impair dignity. There is no rigid or formulaic test that fully captures this inquiry. Attention to context is essential in every case. Openness to critical re-examination of social meanings is just as important.

96. In this case, the differential treatment at issue is the use of age to condition receipt of a survivor pension under the CPP. The CPP grants benefits to surviving spouses over the age of thirty-five immediately following the death of the contributor. However, these benefits are not available to able-bodied spouses without dependent children when the claimant is less than thirty-five years of age at the time of the death of the contributor, until she reaches the age of sixty-five or becomes disabled in the interim. In addition, while those over age forty-five are entitled to receive benefits at the full rate, those between the ages of forty-five and thirty-five receive a progressively reduced sum.

97. Ms. Law was widowed after eleven years of marriage at the age of thirty. Her husband had contributed to the CPP for twenty-two years. Had she been disabled or in charge of dependent children, she would have received a full survivor’s pension. Had she been forty-five years old at the time of her husband’s death she would have received a full pension. Had she been between thirty-five and forty-five she would have received a partial pension. Because she was under thirty-five, able-bodied, and without dependents her entitlement to the survivor’s benefit was deferred until her own retirement age. The question at hand is whether this demeans those denied either through relying on prejudice or stereotype or through denying a benefit important to participation in social, political, or economic life.

98. The scheme involves explicit differential treatment tied to one of the enumerated characteristics in section 15, namely age. This is sufficient to satisfy the first two steps of the test laid out by the Supreme Court of Canada in this case. But does this distinction violate human dignity? The Supreme Court of Canada thought not. The Women’s Court of Canada finds this analysis incomplete because of the failure to take account of the gender implications of the scheme. Indeed, this legislation is better analyzed as an age-based distinction that has a differential effect on women. Although men under forty-five are also affected, women are more likely to find themselves in this situation and to suffer more from exclusion. The real constitutional question, therefore, is whether the effects of excluding young women from CPP income security protection are such as to violate dignity.

(2) Rationale

99. Existing case law holds that proof of discrimination requires more than mere differential negative effect connected to an enumerated or analogous ground. It follows that the section 15 inquiry must delve into the rationale for the law or policy at issue because this sets the stage for ascertaining whether it imposes the kind of harm that discrimination is. To begin with, the claim may be that the objective itself is discriminatory — that is, it violates human dignity by design or effect through being grounded ultimately in prejudice, invoking stereotype, or impairing a fundamental interest of persons identified by reference to an enumerated or analogous ground. In response, the defending party is likely to describe the objective of the provisions in a non-discriminatory fashion. The court can only resolve such a dispute by deciding for itself what the rationale of the rule or policy is and whether it is indeed discriminatory.

100. Further, given the importance that the case law has attached to the question of stereotype in identifying discrimination, questions of whether the generalizations built into policy are accurate and well matched to the ends envisioned are inevitable. This focuses on the means used by the legislature — the classifications employed. Yet only in light of the ultimate objectives can we assess the accuracy of the means, so as to detect the stereotypes at play. Indeed, the means used may themselves impugn dignity in any of the same three ways we have identified. It may not always be clear at the outset whether a challenge is best levelled at the objective as a whole or at the means used to achieve an independent end. It is important not to jump to an assessment of means without first having tested the objective.

101. The analysis of each case must then begin with a clear investigation into the objectives of the law or policy. Without getting this right, we can have no confidence that we have properly understood the implications for human dignity of the impugned law or policy. It can be argued that it was a mistake for the Supreme Court of Canada to have made dignity, or anything else, an additional ingredient in an equality claim because the consideration of rationale that comes in its wake moves section 1 considerations into the section 15 analysis. Equality jurisprudence would be very different now if the initial approach had been different. Analysis under section 15 would have been much simplified, but the section 1 analysis would bear much more of the substantive weight. Rather than unravel all of the strands that have already been woven into equality jurisprudence, this judgment accepts the need to consider legislative objective within the section 15 analysis, but it aims to demonstrate how this inquiry must be conducted to avoid unfairly prejudicing claimants and allowing government to evade careful scrutiny of its aims and techniques.

102. It is crucial that the investigation into the point of the legislation or policy be full and rigorous. A court cannot simply take the word of government actors. When it does, it slides back perilously close to the formalism of Dicey. It reduces discrimination to subjective intentions since it is rare that a government will explain its own legislation as designed to penalize or disadvantage members of a group identified by an enumerated or analogous ground. The court itself has the responsibility to determine what the legislation’s real import is. Of course, the government’s stated objective is relevant, but so are the underlying assumptions behind those objectives as well as the actual design and impact of the scheme.

103. The fact that the government may have had some worthwhile objective in putting into place a particular rule or policy does not end the inquiry. Complex schemes can have multiple objectives. The pursuit of even a worthwhile objective may differentiate in a way that inflicts an indignity on some. When it does, it violates section 15, and we must pass on to section 1 to determine whether the harm done is nevertheless justified. Likewise, the conclusion that legislation employs rational means to an acceptable end does not end the inquiry. The use of means that are not well suited to the government end may help diagnose a dignity violation, but the rationality of the means is not a guarantee of constitutionality. It is possible for even rational means to have discriminatory effects. Again, when this happens, there is a violation of section 15, which must be tested against section 1.

104. To begin with, then, we must identify the rationale for the provisions at issue — in this case, the survivor benefit under the CPP, including its restrictions—and the relationship of the means used—age restrictions — to the rationale behind the benefit.

(3) Canada Pension Plan Survivor Benefits

(a) Background

105. The CPP is one of several instruments aimed at providing an adequate standard of living for retired Canadians. For participants in the paid labour market, the CPP supplements the Old Age Security plan by providing pension benefits based on income earned. In effect, the CPP requires contributors to save for retirement. Both the employee and employer contribute based on the employee’s income up to certain maximums. The contribution rate is supposed to ensure a large enough fund to pay a specified pension in retirement. The pension is meant to be replacement for the income lost because of retirement. The CPP does not provide total income replacement. It covers only a percentage of the contributor’s earnings.

106. In addition to the contributor’s pension entitlement, the CPP provides survivor benefits for the spouses of contributors. Although sex was not claimed as a ground of discrimination by Ms. Law, the meaning and consequence of the benefit can scarcely be understood or evaluated without taking into account the differences between men and women with respect to their participation in the paid workforce, their earning capacity, and their likelihood of experiencing the death of a spouse. The survivor benefit is now available to both sexes, but it was originally designed with women in mind, and it is still paid mostly to women simply because women are more likely to survive a male spouse than vice versa. Ninety per cent of the recipients are women (Human Resources Development Canada (HRDC), Evaluation of Survivor Benefits and Other Features of Canada Pension Plan, Final Report, May 1997, at 18). Although the social realities of women’s exclusion from the
workforce were taken into account in the original design of the policy, Parliament has not been as attuned to the hurdles that women have faced and still face even as we have increasingly moved into paid work.

(b) Rationale for Survivor Benefits

107. In the litigation, the respondent presented the survivor benefit as being designed to help meet the long-term need of a surviving spouse and the eligibility age as a proxy for such need. The Supreme Court of Canada accepted this characterization. Iacobucci J. treated the human dignity issue as partly a function of “the aim and effects of the legislation in providing long-term financial security for Canadians who lose a spouse, coupled with the greater flexibility and opportunity of younger people without dependent children or disabilities to achieve long-term security absent their spouse” (Law, at para. 102 [emphasis in original]). Put simply, the argument is that the younger the surviving spouse is at the time of the contributor’s death, the less in need of a permanent monthly pension she or he is likely to be. Since the proffered rationale seems laudable on its face, the argument centred on whether the use of age as a proxy for need was acceptable.

108. This analysis is flawed in two respects, both connected to the failure to consider the gendered context and effects of the survivor benefit. First, the respondent’s argument seems to trade on an old-fashioned model of “wifely” dependency that is out of touch with contemporary understandings of spousal inter-dependency. Second, and more importantly, the respondent focuses exclusively on long-term need, making the use of age seem more defensible. Yet this choice must itself be tested against section 15. It is easy to treat as insignificant other forms of need in those under forty-five if one imagines that claimants are as likely to be male as female. The flaws in this image become clear when the fact that most recipients are female is put together with well-known data about gendered wage differences. This exercise reveals the male norm that is implicit in the scheme.

109. The respondent has also implicitly treated the relevant need as acute financial need. This both fits with the idea of wifely dependency and strengthens the empirical generalization that younger survivors are less likely to suffer need in the long term. However, this rationale fits uneasily with the actual eligibility requirements, which imply a relative conception of need — a concern about a survivor’s drop in standard of living from whatever level was previously enjoyed, whatever the size of the drop. When we correct the assumption that only acute need is targeted, it becomes harder to justify taking care of gender-related long-term relative need while ignoring short-term relative need that is equally gender-related.

110. The respondent’s argument misleadingly treats the survivor pension as a kind of social assistance program, which seems to hearken back to an outdated attitude towards “dependent wives.” It may once have been common to treat dependant widows as a central case of the deserving poor, worthy of society’s charity, but we now know better. In fact, in most cases a wife — whether in waged work or working in the home — makes an important contribution to her husband’s ability to earn income. Indeed, both spouses typically make an economic contribution towards their joint welfare. The survivor benefit is better understood as recognition of this contribution, as reflecting the economic partnership that spousal relationships embody. The idea of a pension as a family asset is explicitly recognized in the provisions allowing for the splitting of pension credits upon separation (CPP, s. 55.1). This reading is not only more accurate, it also dignifies women’s contribution
to their family’s economic well-being, including their unpaid labour in the home, rather than treating it as a source of women’s dependency.

111. However, the CPP is a pension plan, after all. As such, it is primarily designed to provide replacement income post-retirement. Adhering strictly to this objective would produce a survivor benefit that vested only once the recipient reached retirement age. And yet the benefit is provided for some at a younger age. Just as a contributor who becomes disabled is able to access his pension before he turns sixty-five, a surviving spouse who is herself disabled or caring for dependent children is eligible for the survivor pension early. This early vesting has been extended to all survivors between the ages of forty-five and sixty-five. Social patterns when the CPP was introduced explain this extension. The original assumption was that recipients would be traditional stay-at-home wives and homemakers. Having relied on her husband as her sole source of income, such a woman would likely find it extremely difficult to enter the labour force after a certain age. These pre-retirement age eligibility rules seem to recognize that some survivors below the age of sixty-five will need a pension because of likely barriers to their participation in the workforce.

112. The ineligibility or reduced eligibility of those under forty-five must then be fitted into this rationale — younger people, the respondent says, are able to support themselves, and are therefore not in need. This argument trades on the assumption that the benefit aims at remedying acute need and combines with the “alleviation of long-term need” rationale to lend credibility to the government’s claim that age is a sound proxy for need. Yet, the benefit is extended to some spouses who do not experience such dire need. Those over the age of forty-five are eligible whatever their alternative sources of income. The surviving wife with a career of her own is just as eligible as her non-wage-earning sister. Indeed, since 1975, men over the age of forty-five who survive their spouse are eligible, though few middle-aged men are financially dependent on their spouses. The age-related basis for eligibility may have initially had in mind the entirely dependent wife, but it has expanded considerably beyond this paradigm case. The construction of a rationale for the scheme must incorporate this expansion.

113. The extension of the scheme to all individuals over age forty-five seems to recognize that the loss of one income within a household is likely to occasion a drop in the survivor’s standard of living, creating hardship sufficient to justify an entitlement to offset it from the pension savings of the deceased spouse. A sudden drop in income is likely to cause trouble even in a dual-earner family and for people in most income brackets. A surviving spouse who is working may not be in acute need following the death of a spouse but is likely to have developed a lifestyle and range of financial commitments that are not easily adjusted to accommodate the loss of a spouse’s income and may therefore be in relative need of income support.

114. Indeed, HRDC acknowledges that the participation rate of women in all age groups in the paid workforce has changed significantly since the 1960s, and yet “the need for earnings replacement [following a spouse’s death] has not diminished” (Human Resources Development Canada, supra, at 10). The conception of need operating, then, must be relative to the survivor’s previous standard of living rather than being understood in absolute terms. The HRDC report also acknowledges that the CPP functions in part as an insurance policy for surviving spouses as reflected in the flat rate component of the benefit (at 11). This fact also counsels against construing the benefit as a species of welfare since payment of an insurance benefit is usually not tied to dire need.

115. Thus, a rationale for the survivor pension that treats it as being akin to a welfare benefit predicated upon acute need is inaccurate and untrue to the modern conception of the spousal relationship and some of the eligibility requirements of the scheme itself. These aspects of the rationale offered must be rejected. Instead, we find that the survivor pension is based on recognition of each spouse’s contribution to the couple’s joint wealth, such that the financial disruption flowing from a sudden drop in household income triggers entitlement to access the couple’s pension nest egg.

116. However, the government also presented the scheme as focused on long-term need for income support and relied heavily on this notion to support its use of age as a proxy for the relevant need. This aspect of the rationale fits with the fact that the benefit is structured as a monthly pension payable from the point of eligibility until the recipient’s independent pension benefits become available or her own death, as the case may be. In other words, since a long-term payment is envisioned, it must be meant to deal with long-term need, which is implicitly assumed to be acute need. The argument that follows assumes that confining the scheme to remedying long-term need is legitimate, so that the only question is whether the means used — age-based eligibility — is discriminatory. This initial assumption, however, itself requires examination.

117. The respondent’s claim that those under forty-five are less likely to suffer long-term (acute) need as a result of a spouse’s death seems plausible. The very fact that younger adults have more time to overcome or adjust to the financial effects of their loss before their working life ends must increase the accuracy of the generalization that they are better able to do so than an older person would be. If the generalization is accurate, a connection between youth and employability can be presented as not being demeaning — neither exhibiting prejudice or stereotype nor impairing fundamental interests for which others enjoy protection.

118. If this generalization’s accuracy were crucial to our decision, we would nevertheless say that the government should have been put to a fuller proof of the claim that those under the age of forty-five are better able to support themselves adequately, even in the long run. Since the respondent offers this rationale to deflect the claim of discrimination, and is in the best position to support it, it is the respondent who should prove it. A claimant cannot be expected to anticipate all of the possible rationales that the government might offer for legislation and to present evidence refuting the factual premises of each, especially when based on complex demographic data. If the government has done its homework in designing the policy in the first place, it should be a small matter to shoulder the evidentiary burden of supporting the factual claims that ground its work. The burden of persuading the court that the import of the law or policy is dignity violating may still reside with the claimant, but the job of proving that its factual assumptions are sound should lie on the government.

119. The fact that adequate empirical evidence of the link between age and long-term self-sufficiency was neither offered nor demanded indicates that the use of age is not merely based on an empirical generalization but rather on a normative expectation that younger adults who are able to work should do so rather than expecting a life-long annuity flowing from a spouse’s pension contributions. How else to explain the government’s failure to provide any opportunity for younger survivors to demonstrate that they share the same sort of need experienced by older survivors, despite their youth? This expectation may even be reasonable as long as we assume that the CPP may restrict itself to providing long-term income security for surviving spouses who need it. However, this assumption must be tested against the equality norm embodied in section 15.

(c) Is the Exclusive Provision for Long-Term Relative Need Discriminatory?

120. As reconstructed, the rationale for the survivor benefit has two dimensions. First, despite the assumption that acute need is targeted, it makes more sense of the actual eligibility criteria to treat the benefit as being designed to offset the drop in standard of living relative to that preceding the death of the survivor’s spouse. Second, the scheme is concerned with long-term need. The former is integrally connected to the survivor benefit’s role as an offshoot of the contributor’s CPP pension. There seems no basis for finding this aspect of the objective to be discriminatory. Quite the contrary, a concern to protect spouses from a decline in household income is a dignity-enhancing objective, not because money is important in itself but rather because life plans and quality of life are predicated on the level of income one has already achieved and can be severely, even permanently, disrupted by a sharp drop.

121. However, the exclusive focus on long-term need is another matter. Even if the government can make out the case that younger survivors do not suffer in the long term from the loss of a spouse’s income, this does not establish that they do not experience a short-term version of the need that the CPP aims to alleviate. Is the failure to consider the specific — that is, short-term—needs and interests of younger survivors discriminatory? Does it demonstrate prejudice, whether conscious or not, or reliance on stereotypes about the group affected, whether conscious or not, or does it deny a benefit available to others that is important to a life with dignity?

122. We could conduct this analysis, as the courts did, in a gender-neutral way, but such an attempt seems highly artificial and obscures a central question. We have noted that the vast majority of survivor pension recipients are women and there is every reason to think that younger survivors are also more likely to be women. Thus, the failure to provide for the short-term need of younger survivors is predominantly a failure to consider the specific needs of younger women who have lost a spouse. Indeed, it is perhaps because of a failure to take account of sex that the legislature could so easily conclude that survivors under the age of thirty-five have no income security needs at all and can be expected to fend for themselves. The discrimination question is whether the legislation has differential effects on young women that are discriminatory.

123. We may quickly eliminate any claim that the scheme violates dignity through the expression of prejudice. Although women, including young women, have been historically and systematically targeted for deprivation in some ways out of prejudice it would be hard to read the exclusion at issue in this case as an overt expression of the presumed inferiority of young women or as contributing to the persistence of such attitudes.

124. Is the legislation based on stereotype? Stereotypes are the general characteristics often falsely ascribed to subordinated groups that contribute to keeping them in a subordinate position. For this reason, stereotypes usually have a derogatory connotation even though they may not express conscious ill will. To say, for example, that non-citizens are unlikely to be knowledgeable about Canada and therefore would be less competent as lawyers or to say that those over the age of sixty-five are less competent and therefore should be forced to retire is to over-generalize and to assume that all members of a group have a quality that may actually be true only of some. The injustice is in depriving some of an opportunity based on the inaccurate generalization when they are, in fact, capable.

125. Much of the argument before the Supreme Court of Canada was bound up with whether the generalization that younger adults do not experience long-term need as a result of the death of a spouse is a stereotype. In many ways, it does not fit the stereotype paradigm, largely because it does not attribute a negative characteristic or incapacity to the target group. However, it is a mistake to try to shoehorn this case into the category of stereotype. The scheme may operate on the basis of inaccurate assumptions, but they are not the sort that characterizes stereotyping. Rather, the legislature has falsely assumed that younger survivors have no financial security needs at all—that is, it has failed to take account of the specific needs of younger spouses, even as it has provided for the comparable needs of older ones.

126. Whether this failure rises to the level of discrimination depends on the importance of the good or benefit at issue and on whether it is sufficiently important to fundamental interests that its denial to a particular class of persons when it is provided for others impairs human dignity. Here we are dealing with a benefit whose denial has a greater impact on women than on men, and the dignity inquiry must take account of this difference.

127. The exclusion of younger spouses does not convey a message that some intimate relationships are more valuable than others, as the exclusion of same-sex couples from similar benefits has done in the past. Rather, it is the economic support and its value that is at issue.

128. Providing such a pension for older survivors indicates a public recognition of the importance of income security in the face of the death of one’s spouse. This is obviously a time of great emotional vulnerability, and it should not be compounded by financial vulnerability. Young survivors may not need permanent assistance to get on their feet, but failure to create a bridge from (partial) reliance on a spouse’s income to full self-sufficiency can have serious, even permanent, repercussions, which are not very different from those facing many older survivors. Avoiding these kinds of consequences is the very point of the survivor benefit.

129. The loss of one spouse’s income is likely to occasion a significant drop in living standards for the young as well as the older, requiring at least some period of adjustment. The failure to take account of this short-term need has much more serious consequences for women than for men. The expectation that young widows can and should be instantly self-sufficient is clearly unsound. Women are certainly capable of employment and are participating in the workforce in ever increasing numbers. However, the lingering effects of systemic sex discrimination in the workforce continue to ensure that women’s earning power is considerably lower than that of men.

130. The most recent Statistics Canada data show little change from the data that would have been available when this case was being litigated. The employment rate among women is lower than men’s in all age categories except those under the age of twenty-four (Statistics Canada, CANSIMTable 282-0002: Labour Force Survey Estimates, by Sex and Detailed Age Group, Annual, 2005, version updated 1 July 2006, accessed 2 July 2007). Likewise, the percentage of women in paid work (the participation rate) is lower than that of men in all age groups except fifteen to nineteen year olds (ibid.). In the population as a whole, even in households without children, women’s workforce participation rate is lower than men’s (52.9 per cent versus 64.1 per cent) (Statistics Canada, 2001 Census of Population: Labour Force Activity, Class of Worker, Occupation, Industry, Place of Work, Mode of Transportation, Language of Work and Unpaid Work, database, version updated 31 August 2006, accessed 2 July 2007). Among those wanting full-time work, women’s unemployment rate is also higher than men’s in every age group except fifteen to twenty-four year olds (Statistics Canada, CANSIM Table 282-0086: Labour Force Survey Estimates, Supplementary Unemployment Rates by Sex and Age Group, Annual, 2005, version updated 1 July 2006, accessed 2 July 2007).

131. Women are still streamed into lower paying jobs and are still often paid less than those in comparable male jobs. Even unattached women, whose workforce attachment is likely to be most like men’s, do not earn as much as men. This is true even for younger women (women between the ages of thirty-five and forty-four and between the ages of twenty-four and thirty-four earn almost 88 per cent of what similarly situated men earn) (Women in Canada, supra, at 154). The statistics for the population as a whole (unattached women and women in families, with and without children) show that women aged fifteen to twenty-four earn 81 per cent of male earnings, while women aged twenty-five to thirty-four and thirty-five to forty-four earn close to 75 per cent (at 140).

132. Wives’ earnings represent only about 34 per cent of the income of dual-earner families (Women in Canada, at 140). Because these data are not broken down according to age or whether there are children in the household, they may understate the contribution of younger, childless women to household income. Nevertheless, they provide some reason to think that the loss of a spouse is likely to hit women harder than men. Most women continue to earn less than their male spouses, and many women, even those without children, continue to shoulder the lion’s share of household responsibilities to the detriment of their earning capacity (Statistics Canada, 2001 Census of Canada: Hours Spent Doing Unpaid Housework, 11 February 2003, Catalogue no. 97F0013XCB 2001001).

133. Most worryingly, if a widow has not previously been partially employed or fully employed, it will take time and resources to train or retrain for entry into, or repositioning in, the workforce — want of these resources may confine someone for many years, or even permanently, to low-skill, low-wage jobs. If it is unacceptable to inflict this on a forty-five-year-old woman when her spouse’s pension can help, it is hard to see why the needs of younger women should count for less.

134. Put bluntly, the popular adage that many women are just a divorce away from poverty could easily apply to the death of a spouse. To the extent that women earn less than men, women’s contribution to joint household income is likely to be less than their male spouse’s. Thus, loss of a male spouse will often precipitate a sharper drop in the surviving female spouse’s standard of living than a man’s loss of a female spouse and will be harder to make up. Thus, it is precisely the female survivor who is likely to face the most exigent circumstances following the death of her spouse — the loss of more than half, perhaps much more than half, of her previous household income and the need to make substantial investments in her own earning capacity in order to put herself on an even financial keel. Some young, well-paid professionals may be able to weather this adjustment period with ease, but many lower income survivors will undoubtedly be thrown into crisis. For those just making ends meet, a spouse’s death might well throw their life into complete financial chaos, leading to the loss of home or apartment and a downward spiral into poverty out of which it would be difficult to climb.

135. The degree of financial dislocation suffered will vary depending on the circumstances of the couple, but the same is true of older couples. The government has apparently decided that anyone over the age of forty-five legitimately needs some help. It has not made older survivors responsible for proving any particular level of need, with the attendant delay and additional stress. If this is the right judgment to make in respect of older survivors, it is difficult to see why the same consideration should not be extended to younger ones, adjusted to take account of the shorter-term nature of the need they are likely to experience. It may be legitimate for the government to revisit this decision and to focus more narrowly on those in the most serious need. We reserve judgment on this question. However, even in the event of such a policy change, it seems unlikely that age will be a good proxy for acute need. In other words, age may correlate with acute, long-term need, but it is not a good indicator of acute need more generally. Indeed, it might be more sensible to revisit the decision to extend the survivor benefit to men than to use age to determine eligibility.

136. It is legitimate to expect surviving spouses to adjust to a lower standard of living, if need be, or to take more responsibility for their own support, if need be, but it is harsh and dismissive of the needs of younger spouses, mostly women, to expect these adjustments to take place instantaneously. The forces that have combined and continue to combine to hamper women’s equal participation in the workforce and induce them to subordinate the development of their full economic potential to their family’s interests are well known to the government. These consequences and their further side-effects — economic vulnerability breeds other forms of disadvantage — are a major part of the story of women’s subordinate status in society. Under these circumstances, the creation of a pension scheme that seeks to counter these forces for some women, but not for others, abandons those excluded to the discriminatory forces operating in the workplace and the family. The concrete consequences for their lives may be profound, and the message is that their subjection to the consequences of these discriminatory forces does not matter.

137. The government seems not to have been attending to these gendered consequences. Extending eligibility for the survivor benefit to men has made it easy to lump men and women together when asking whether age is a good indicator of self-sufficiency. The result is to create a de-sexed younger surviving spouse. The danger is that the image created is really a male image to which women are expected to conform or bear the consequences. The complete failure to address whether the typical recipients of the survivor benefit — women — are as capable of achieving instant self-sufficiency following the death of a spouse as a comparable man would be suggests that a male norm guided both Parliament’s and the Supreme Court of Canada’s thinking. Perhaps it is true that relatively few younger men suffer financial dislocation to any significant degree from the loss of a female spouse. To impose the same expectations on women given their different experience of paid employment is to impose a male norm.

138. The tendency to adopt a male norm — in this case, the expectation that women under the age of forty-five support themselves like an able-bodied, childless man could — is powerful and pervasive. In feminist circles, it has a name: equality with a vengeance. Disadvantaged groups such as women tend either to be treated in accordance with derogatory or paternalistic stereotypes or assumed and required at their own peril to be like men whether they are or not, whether they have been given the chance to be or not, and whether it suits their own aspirations or not. This ignores the real work of achieving equality, namely a clear-sighted examination of the actual conditions of women’s lives to determine what they need in order to overcome socially imposed disadvantage, flourish on their own terms, and live lives of dignity and full participation in society.

139. Treating women the same as men when the circumstances of their lives are different tends to entrench existing disadvantage. This is especially so with an income support program that ignores the obstacles younger women face in achieving full and meaningful participation in the workforce. It may be asking too much that all legislation be perfectly calibrated to all of the differences in conditions affecting different groups in society, but refusing income support to younger women because younger men are unlikely to need it when this denial is likely to perpetuate background conditions of inequality fails to treat women with respect. The fact that only some women, younger ones, have been affected by the use of a male norm does not excuse the government. Put the other way around, the fact that substantive equality is achieved for older women, by providing a benefit that responds to needs more specific to women, does not excuse denying it to younger ones.

140. In some respects, society is like a game of snakes and ladders in which female players of all ages are pre-determined to land on more and longer snakes, although older players may be more likely to land on the very longest snakes. The government has done something to reduce the chances of a slide into poverty for older women but has treated younger women as though there were no more dangerous snakes in their path than in men’s. Doing this abandons them to life’s gendered traps.

141. The unwitting adoption of a dominant — in this case, male — norm in legal rules and policy to the detriment of women has been uncovered and rectified by courts in some contexts. The phenomenon and its remedy are not entirely foreign to the courts. For example, in R. v. Lavallee, [1990] 1 S.C.R. 852, the traditional interpretation of the self-defence provision of the Criminal Code, R.S.C. 1985, c. C-46, s. 264, requiring an imminent deadly threat, was modified to counteract the assumption that a woman can leave an abusive relationship as easily as a man can walk away from a violent confrontation. The Supreme Court of Canada has also recognized that policies designed with able-bodied persons in mind will rarely serve the interests of those with disabilities (Eaton, supra, and Eldridge, supra).

142. The second contextual factor the Supreme Court of Canada has identified as an indicator of discrimination — whether the law or policy takes account of the actual circumstances and needs of those affected by it — seems ideally formulated to prod courts to look for hidden male (or white or able-bodied) norms. Failure to use it in this way in this case serves to reinforce, rather than dismantle, the dominant male norm. Having accepted too quickly the government’s characterization of the survivor benefit as being designed to alleviate long-term need, the Supreme Court of Canada’s consideration of the “correspondence” factor focused on the connection between advancing age and long-term need. Yet deciding that the legislation does an admirable job of fulfilling the needs of those who are included does not tell us whether the needs of those who are excluded can be legitimately discounted. A factor that should have worked to focus on the specific age- and sex-related needs of younger, mostly female survivors ended up reinforcing their exclusion simply by pointing out that they do not share exactly the same needs that older, mostly female survivors have, even though the obstacles to full workforce participation they experience are very similar. This bespeaks an overly narrow conception of equality.

143. The failure to include some provision for short-term financial dislocation caused by a spouse’s death differentially affects young women because it is women who are more likely to suffer such a loss and the economic consequences are likely to be more severe for women than for men. Denying short-term income support to help women adjust to the drop in income and the need to maximize their economic potential stands to have potentially cascading, serious, and long-term effects. This must be held to violate younger women’s human dignity by exacerbating pre-existing economic vulnerability and creating obstacles to their ability to overcome it. It discriminates against younger persons who are predominantly women and excludes them from a survivor pension just because their income security need may be of shorter duration than that of older widows. The current provisions of the CPP therefore violate section 15 by discriminating on the combined grounds of age and sex.

G. Section 1 of the Charter

144. It remains to consider whether the scheme can be defended under section 1. The general outlines of the Oakes test are well known. First, “[t]he objective, which the measures responsible for a limit on a Charter right are designed to serve, must be of sufficient importance to warrant overriding a constitutionally protected right” (R. v. Oakes, [1986] 1 S.C.R. 103 at 138). The proportionality part of the test then assesses the rationality of the connection between the offending provision and the objective offered for it, and the reasonable necessity of violating rights to achieve that objective as a prelude to determining whether an objective that survives to the final stage of the analysis is more important than the right violated to achieve it.

145. Especially in section 15 cases, the objective relevant to the section 1 inquiry is the objective of the right-limiting provision — in this case, the decision to exclude from assistance those in short-term financial need following the death of a spouse even though it differentially and detrimentally affects younger women. Equality cases always involve the distribution of some benefit or burden. The government can therefore always say that its general objective was to benefit those who are included in the scheme. Conferring such a benefit will always be constitutionally valid and even of significant importance, but it will usually be beside the point. The question is whether the exclusion of others is defensible, and the objective behind leaving out those left out must therefore be assessed. The only circumstance in which it might be open to argue that the objective of the exclusion of some is the general objective of the legislation or policy to benefit others is when the only way to provide for some is to exclude others. In all other cases, the objective of the rights-violating provision must be to avoid whatever other difficulties or costs there would be in extending the benefit. These costs must then be measured against the value of protecting equality rights.

146. The objective offered for section 1 purposes must be one that is capable of competing with the importance of protecting equality. In this case, the government seems to have focused its energy on arguing that the interests of those needing short-term assistance to achieve financial security do not rise to the level of attracting protection under section 15. Given our finding that excluding short-term need differentially affects younger women in dignity-impairing ways, an objective for such exclusion that is capable of competing with the value of equality is hard to discern from the record from the courts that have considered this case.

147. Only the Pension Appeal Board went into the section 1 argument (Law v. Canada (Minister of Employment and Immigration) (1995), C.E.B. & P.G.R. 8574 (PAB)). Rutherford J., for the majority, put a great deal of weight on the complexity of the CPP and the level of multilateral support needed to amend it. Contribution levels are set according to the type and level of benefits payable. Major changes to the plan require the approval of the federal government and two-thirds of the provinces with at least two-thirds of the population (Law (PAB), at 6084). Against this backdrop, Rutherford J. advocated deference to Parliament in respect of decisions “which impact on the public purse” (at 6085).

148. Three threads seem to be woven together to support this conclusion, but none amounts to section 1 justification. In part, Rutherford J. relies on the complexity of the CPP scheme as a whole and the difficulty of achieving consensus between levels of government. These factors are more in the nature of general reasons for judicial deference to government decisions than the statement of a government objective capable of justifying an equality violation. There may be circumstances in which deference is justified, but they do not seem to be in evidence in this case. First, though the CPP scheme as a whole is complex, this should not weaken the courts’ resolve to scrutinize the justifications offered for violations of the right to equality. As our account of equality becomes more sophisticated, we should expect that cases raising more complex issues involving detailed benefit schemes will come before the courts. To defer to government in all of these cases will leave section 15 with very little role to play in guiding government behaviour. A plea of complexity seems like a smokescreen to relieve the government of the responsibility to demonstrate what it takes to be the compromises or sacrifices that would be necessary in order to uphold the right to equality. This is to subvert the function of section 1.

150. Second, we are not persuaded that the difficulty of achieving federal-provincial consensus on changes to the CPP is a reason to stint on Charter scrutiny of the program. Quite the contrary, political deadlock may mean that a rights violation continues indefinitely, absent judicial intervention. This case is a case in point. The Pension Appeal Board was assured by the respondent’s expert witness, Terry De March, chief of legislative development in the Policy and Legislation Division, Income Security Programs Branch of the HRDC, that the government “acknowledged that the current criteria may no longer be appropriate and stated its intention to work with the provinces to arrive at a consensus on this issue” (Law (PAB), at 6081). De March testified that work on reaching consensus and on alternative models was underway. This was in 1995. Eleven years later, no changes have been made to the survivor benefit. Equality rights cannot be left to the mercy of such political stasis. We prefer to believe that if the Supreme Court of Canada had clearly named the section 15 violation in the current scheme and held the federal government and the provinces to their responsibility to rework the scheme to cover the needs of younger spouses, the governments would have risen to the challenge.

151. Rutherford J.’s third argument concerns the implications for the public purse of changes to the CPP. He offered this as a further reason for deference, but it might also be formulated as an objective competing with the right to equality — that is, the government might argue that short-term assistance was excluded in order to save money. Either way, the argument is flawed. This is not a matter of the public management of scarce resources. Even assuming that cost may be called in aid to make out section 1 justification — an argument that is contentious in itself — the CPP is not in competition for public funds with other social programs mounted by the federal government and provinces. Pensions under the plan are paid entirely out of the contributions of employees and employers. No general government revenues go into the fund. If the costs of the plan go up, contributions may have to rise, but it does not follow that health care funding, or national
defense, or any other government responsibility must suffer.

152. No clear argument emerged from the litigation either that raising CPP contribution rates would be the only, or only feasible, way to include short-term assistance or that the damage that would be caused by such an increase is sufficiently important to justify a restriction on the section 15 rights of younger women. It follows that the government has not met its burden of establishing that the use of age as a criterion of eligibility is demonstrably necessary in a free and democratic society.

VI. Disposition

153. This reconsideration of Law v. Canada leads me to conclude that sections 22(1)(d) and 59 of the CPP infringe section 15 of the Charter on the ground that they discriminate on the basis of age and sex. The government has failed to meet the burden imposed by section 1 in order to save the provisions.

154. Sections 22(1)(d) and 59 are of no force and effect because the use of age as a criterion for eligibility is the means by which younger women’s need for short-term financial support is ignored. However, it does not follow that all survivors under the age of forty-five should be entitled to a full pension until their own retirement age. The fact that the government has overstated the differences between older and younger survivors does not mean that there are no differences. I would therefore suspend the effect of the declaration of invalidity for a period to be determined after hearing argument about the appropriate amount of time within which to expect the governments to revise the scheme. I would also invite representations about the appropriate level of compensation to be awarded to Ms. Law should the parties be unable to come to an agreement on this matter.