Date: Thu, 25 May 1995 18:53:45 -0400 From: ae913@freenet.carleton.ca (Timothy Ross Wilson) To: queerlaw@abacus.oxy.edu, queerplanet@abacus.oxy.edu Subject: Egan v. Canada Reply-To: ae913@freenet.carleton.ca - 1 - Document[5]51 egan v. canada James Egan and John Norris Nesbit Appellants v. Her Majesty The Queen in Right of Canada Respondent and Attorney General of Quebec, Canadian Human Rights Commission, Commission des droits de la personne du Quebec, Equality for Gays and Lesbians Everywhere, Metropolitan Community Church of Toronto, Inter-Faith Coalition on Marriage and the Family and Canadian Labour Congress Interveners Indexed as: Egan v. Canada File No.: 23636. 1994: November 1; 1995: May 25. Present: Lamer C.J. and La Forest, L'Heureux-Dube, Sopinka, Gonthier, Cory, McLachlin, Iacobucci and Major JJ. on appeal from the federal court of appeal Constitutional law -- Charter of Rights -- Equality rights -- Old age security legislation providing for allowance for spouse of pensioner -- Definition of "spouse" restricted to person of opposite sex -- Whether definition of "spouse" infringes s. 15(1) of Canadian Charter of Rights and Freedoms -- If so, whether infringement justifiable under s. 1 of Charter -- Old Age Security Act, R.S.C., 1985, c. O-9, ss. 2, 19(1). The appellants are homosexuals who have lived together since 1948 in a relationship marked by commitment and interdependence similar to that which one expects to find in a marriage. When E became 65 in 1986, he began to receive old age security and guaranteed income supplements under the Old Age Security Act. On reaching age 60, N applied for a spousal allowance under s. 19(1) of the Act, which is available to spouses between the ages of 60 and 65 whose combined income falls below a fixed level. His application was rejected on the basis that the relationship between N and E did not fall within the definition of "spouse" in s. 2, which includes "a person of the opposite sex who is living with that person, having lived with that person for at least one year, if the two persons have publicly represented themselves as husband and wife". The appellants brought an action in the Federal Court seeking a declaration that the definition contravenes s. 15(1) of the Canadian Charter of Rights and Freedoms on the ground that it discriminates on the basis of sexual orientation and a declaration that the definition should be extended to include "partners in same-sex relationships otherwise akin to a conjugal relationship". The Trial Division dismissed the action. The Federal Court of Appeal, in a majority decision, upheld the judgment. Held (L'Heureux-Dube, Cory, McLachlin and Iacobucci JJ. dissenting): The appeal should be dismissed. The definition of "spouse" in s. 2 of the Old Age Security Act is constitutional. Per Lamer C.J. and La Forest, Gonthier and Major JJ.: The analysis under s. 15 of the Charter involves three steps: the first looks to whether the law has drawn a distinction between the claimant and others; the second questions whether the distinction results in disadvantage, and examines whether the impugned legislation imposes a burden, obligation or disadvantage on a group of persons to which the claimant belongs which is not imposed on others, or does not provide them with a benefit which it grants others; the third step assesses whether the distinction is based on an irrelevant personal characteristic which is either enumerated in s. 15(1) or one analogous thereto. The first step is satisfied in this case, since Parliament has clearly made a distinction between the claimant and others. The second step is also satisfied: while it may be true that the appellants have suffered no prejudice because by being treated as individuals they have received considerably more in combined federal and provincial benefits than they would have received had they been treated as "spouses", there is nothing to show that this is generally the case with homosexual couples. Sexual orientation is a deeply personal characteristic that is either unchangeable or changeable only at unacceptable personal costs, and so falls within the ambit of s. 15 protection as being analogous to the enumerated grounds. All that remains to be considered under the third step is whether the distinction made by Parliament is relevant. In assessing relevancy for this purpose one must look at the nature of the personal characteristic and its relevancy to the functional values underlying the law. A form of comparative analysis must be undertaken to determine whether particular facts give rise to inequality. This comparative analysis must be linked to an examination of the larger context, and in particular with an understanding that the Charter was not enacted in a vacuum, but must be placed in its proper linguistic, philosophic and historical contexts. The singling out of legally married and common law couples as the recipients of benefits necessarily excludes all sorts of other couples living together, whatever reasons these other couples may have for doing so and whatever their sexual orientation. What Parliament clearly had in mind was to accord support to married couples who were aged and elderly, for the advancement of public policy central to society. Moreover, in recognition of changing social realities, s. 2 was amended so that whenever the term "spouse" was used in the Act it was to be construed to extend beyond legal married couples to couples in a common law marriage. Marriage has from time immemorial been firmly grounded in our legal tradition, one that is itself a reflection of long-standing philosophical and religious traditions. But its ultimate raison d'etre transcends all of these and is firmly anchored in the biological and social realities that heterosexual couples have the unique ability to procreate, that most children are the product of these relationships, and that they are generally cared for and nurtured by those who live in that relationship. In this sense, marriage is by nature heterosexual. It would be possible to legally define marriage to include homosexual couples, but this would not change the biological and social realities that underlie the traditional marriage. Many of the underlying concerns that justify Parliament's support and protection of legal marriage extend to heterosexual couples who are not legally married. Many of these couples live together indefinitely, bring forth children and care for them in response to familial instincts rooted in the human psyche. These couples have need for support just as legally married couples do in performing this critical task, which is of benefit to all society. Faced with the social reality that increasing numbers choose not to enter a legal marriage but live together in a common law relationship, Parliament has elected to support these relationships. Parliament is wholly justified in extending support to heterosexual couples like this, which is not to say, however, that it is obligated to do so and may not treat married and unmarried couples differently. Neither in its purpose nor in its effect does the legislation constitute an infringement of the fundamental values sought to be protected by the Charter. None of the couples excluded from benefits under the Act are capable of meeting the fundamental social objectives thereby sought to be promoted by Parliament. While these couples undoubtedly provide mutual support for one another, and may occasionally adopt or bring up children, this is exceptional and in no way affects the general picture. Homosexual couples differ from other excluded couples in that their relationships include a sexual aspect, but this sexual aspect has nothing to do with the social objectives for which Parliament affords a measure of support to married couples and those who live in a common law relationship. The distinction adopted by Parliament is relevant here to describe a fundamental social unit to which some measure of support is given. The impugned legislation, even had it infringed s. 15, would have been upheld for the reasons given in McKinney v. University of Guelph and for those mentioned in the discussion of discrimination in this case. Per Sopinka J.: The impugned legislation infringes s. 15(1) of the Charter, for the reasons given by Cory J. Such infringement, however, is saved under s. 1. Government must be accorded some flexibility in extending social benefits and does not have to be pro-active in recognizing new social relationships. It is not realistic for the Court to assume that there are unlimited funds to address the needs of all. A judicial approach on this basis would tend to make a government reluctant to create any new social benefit schemes because their limits would depend on an accurate prediction of the outcome of court proceedings under s. 15(1). This Court has recognized that it is legitimate for the government to make choices between disadvantaged groups and that it must be provided with some leeway to do so. When the definition of "spouse" in the Old Age Security Act is measured against overall objectives of alleviation of poverty of elderly spouses, it should not be judged on the basis that Parliament has made this choice for all time. The history of the legislation shows an evolving expansion of the definition of the intended recipients of the benefits. The Attorney General of Canada has taken the position that the means chosen does not have to be necessarily the solution for all time. Hence, since the impugned legislation can be regarded as a substantial step in an incremental approach to include all those who are shown to be in serious need of financial assistance due to the retirement or death of a supporting spouse, it is rationally connected to the objective. With respect to minimal impairment, the legislation represents the kind of socio-economic question in respect of which the government is required to mediate between competing groups rather than being the protagonist of an individual. In these circumstances, the Court will be more reluctant to second-guess the choice which Parliament has made. There is also proportionality between the effects of the legislation on the protected right and the legislative objective. The proper balance was struck by Parliament in providing financial assistance to those who were shown to be in the greatest need of assistance. Per Cory and Iacobucci JJ. (dissenting): In determining whether a s. 15(1) right to equality has been violated, the first step is to determine whether, owing to a distinction created by the questioned law, a claimant's right to equality has been denied. During this first step, the inquiry should focus upon whether the challenged law has drawn a distinction between the claimant and others, based on personal characteristics. The second step is to determine whether the distinction created by the law results in discrimination. In order to make this determination, it is necessary to consider first, whether the equality right was denied on the basis of a personal characteristic which is either enumerated in s. 15(1) or which is analogous to those enumerated, and second, whether that distinction has the effect on the claimant of imposing a burden, obligation or disadvantage not imposed upon others or of withholding or limiting access to benefits or advantages which are available to others. Any search for either equality or discrimination requires comparisons to be made between groups of people. Whether or not discrimination exists must be assessed in a larger social, political and legal context. The resolution of the question as to whether there is discrimination under s. 15(1) must be kept distinct from the determination as to whether or not there is justification for that discrimination under s. 1 of the Charter. This analytical separation between s. 15(1) and s. 1 is important since the Charter claimant must satisfy the onus of showing only that there exists in the legislation a distinction which is discriminatory. Only after the court finds a breach of s. 15(1) does the government bear the onus of justifying that discrimination. Since the law challenged draws a clear distinction between opposite-sex couples and same-sex couples, this case presents a situation of direct discrimination. As a result of the definition of a common law spouse as a "person of the opposite sex", homosexual common law couples are denied the benefit of the spousal allowance which is available to heterosexual common law couples. This distinction amounts to a clear denial of equal benefit of the law. In addition to being denied an economic benefit, homosexual couples are denied the opportunity to make a choice as to whether they wish to be publicly recognized as a common law couple because of the definition of "spouse" set out in the Old Age Security Act. The public recognition and acceptance of homosexuals as a couple may be of tremendous importance to them and to the society in which they live. To deny homosexual couples the right to make that choice deprives them of the equal benefit of the law. The distinction in the Act is based on a personal characteristic, namely sexual orientation. Sexual orientation is analogous to the grounds of discrimination enumerated in s. 15(1). The historic disadvantage suffered by homosexual persons has been widely recognized and documented. Sexual orientation is more than simply a "status" that an individual possesses: it is something that is demonstrated in an individual's conduct by the choice of a partner. Just as the Charter protects religious beliefs and religious practice as aspects of religious freedom, so too should it be recognized that sexual orientation encompasses aspects of "status" and "conduct" and that both should receive protection. The distinction drawn by s. 2 of the Old Age Security Act on the basis of sexual orientation constitutes discrimination. The legislation denies homosexual couples equal benefit of the law not on the basis of merit or need, but solely on the basis of sexual orientation. The definition of "spouse" as someone of the opposite sex reinforces the stereotype that homosexuals cannot and do not form lasting, caring, mutually supportive relationships with economic interdependence in the same manner as heterosexual couples. The appellants' relationship vividly demonstrates the error of that approach. The discriminatory impact cannot be deemed to be trivial when the legislation reinforces prejudicial attitudes based on such faulty stereotypes. The impugned legislation is not saved under s. 1 of the Charter. While the objective of the spousal allowance, which is geared toward the mitigation of poverty among "elderly households", is of pressing and substantial importance, the allowance in its present form is not rationally connected to its legislative goals. A program which included the appellants would better achieve the intended goal while respecting the Charter rights of gays and lesbians. Nor is the denial of the appellants' s. 15 rights through the ineligibility for receipt of the spousal allowance minimally impaired simply because the appellants' joint income would have roughly been the same because of N's receipt of provincial support supplementing his income for a completely unrelated reason. The provincial and federal programs are clearly not co-extensive, and even if they were part of the same overlapping legislative scheme, this is not sufficient to ground a s. 1 justification. Finally, the attainment of the legislative goal is outweighed by the abridgment of the right in this case. The importance of providing relief to some elderly couples does not justify an infringement of the equality rights of the elderly couples who do not benefit for constitutionally irrelevant reasons. The definition of "spouse" in s. 2 of the Act should be read down by deleting the words "of the opposite sex" and reading in the words "or as an analogous relationship" after the words "if the two persons publicly represent themselves as husband and wife". Per L'Heureux-Dube J. (dissenting): A return to the fundamental purpose of s. 15 of the Charter is necessary in order to reconcile the divergent approaches taken by this Court in recent jurisprudence, as well as in the present case and in Miron and Thibaudeau. At the heart of s. 15 is the protection of, and respect for, basic human dignity. "Discrimination" must therefore be at the forefront of the court's analysis. In order for discrimination to be addressed and identified in all of its varied contexts and forms, it is preferable to focus on impact (i.e. discriminatory effect) rather than on constituent elements (i.e. the grounds of the distinction). Discriminatory effects must, moreover, be evaluated from the point of view of the victim, rather than from that of the state. Considerations of relevance are more properly viewed as justifications under s. 1 than as factors integral to the identification of discrimination in the first place. The following factors must be established by a rights claimant before the impugned distinction will be found to be discriminatory within the meaning of s. 15 of the Charter: (1) there must be a legislative distinction; (2) this distinction must result in a denial of one of the four equality rights on the basis of the rights claimant's membership in an identifiable group; and (3) this distinction must be "discriminatory" within the meaning of s. 15. A distinction is discriminatory within the meaning of s. 15 where it is capable of either promoting or perpetuating the view that the individual adversely affected by this distinction is less capable, or less worthy of recognition or value as a human being or as a member of Canadian society, equally deserving of concern, respect, and consideration. The absence or presence of discriminatory impact should be assessed according to a subjective-objective standard -- the reasonably held view of one who is possessed of similar characteristics, under similar circumstances, and who is dispassionate and fully apprised of the circumstances. This determination is arrived at by considering two categories of factors: (1) the nature of the group adversely affected by the impugned distinction, and (2) the nature of the interest adversely affected by the impugned distinction. With respect to the first category, groups that are more socially vulnerable will experience the adverse effects of a legislative distinction more vividly than if the same distinction were directed at a group which is not similarly vulnerable. In evaluating the nature of the group affected by the impugned distinction, it is relevant to inquire into many of the criteria traditionally employed in the Andrews analysis, such as whether the impugned distinction is based upon fundamental attributes that are generally considered to be essential to our popular conception of `personhood' or `humanness', whether the adversely affected group is already a victim of historical disadvantage, whether this distinction is reasonably capable of aggravating or perpetuating that disadvantage, whether group members are currently vulnerable to stereotyping, social prejudice and/or marginalization, and whether this distinction exposes them to the reasonable possibility of future vulnerability of this kind. Membership in a "discrete and insular minority", lacking in political power and thus vulnerable to having its interests overlooked, is another consideration that may be taken into account. The absence or presence of some of these factors will not, however, be determinative of the analysis. However, awareness of, and sensitivity to, the realities of those experiencing the distinction is an important task that judges must undertake when evaluating the impact of the distinction on members of the affected group. Similarly, the more fundamental the interest affected or the more serious the consequences of the distinction, the more likely that the impugned distinction will have a discriminatory impact even with respect to groups that occupy a position of advantage in our society. While the Charter is not a document of economic rights and freedoms, the nature, quantum and context of an economic prejudice or denial of such a benefit are important factors in determining whether the distinction from which the differing economic consequences flow is one which is discriminatory. The discriminatory calibre of a particular distinction cannot, however, be fully appreciated without also evaluating the constitutional and societal significance of the interests adversely affected. Tangible economic consequences are but one manifestation of the more intangible and invidious harms flowing from discrimination, which the Charter seeks to root out. In other cases, the prejudice will be to an important individual interest rather than to one that is economic in nature. Both categories of factors emphasize that it is no longer the "grounds" of the distinction that are dispositive of the question of whether discrimination exists, but rather the social context of the distinction that matters. An effects-based approach to discrimination is the logical next step in the evolution of s. 15 jurisprudence since Andrews. Homosexual couples are denied the equal benefit of the law on the basis of the legislative distinction in s. 2 of the Old Age Security Act, which defines couples as relationships of "opposite sex". That the appellants are able to claim higher benefits as separate individuals does not alter the fact that they have been denied the benefits, both tangible and intangible, of filing for old age benefits as a couple. The impugned distinction excludes the rights claimants because they are homosexual. Consideration of both the nature of the group and the interest affected leads us to conclude that the distinction is discriminatory. Same-sex couples are a highly socially vulnerable group, in that they have suffered considerable historical disadvantage, stereotyping, marginalization and stigmatization within Canadian society. The distinction relates to a fundamental aspect of personhood and affects individuals who, in addition to being homosexuals, are also elderly and poor. Turning to the interest affected, the impugned legislation is a cornerstone in Canada's social security net, which is, in turn, a cherished and fundamental institution in our society. The violation of s. 15(1) of the Charter cannot be salvaged by s. 1, as it is not relevant to a proportionate extent to a pressing and substantial objective. While the objective of the legislation is pressing and substantial, the means chosen to achieve this objective fails all three branches of the proportionality test. The legislation excludes couples who would fill all of the other criteria in the Act except the requirement that they are of the opposite sex. To find that this distinction is rationally connected to the objective of the legislation requires us to conclude that same-sex couples are so different from married couples that it would be unreasonable to make the same benefits available to both. At best, the government has only demonstrated that this is its assumption. The presumption that same-sex relationships are somehow less interdependent then opposite-sex relationships is, itself, a fruit of stereotype rather than one of demonstrable, empirical reality. Nor is s. 15 minimally impaired. A reasonable alternative remedy is available: the discriminatory effect would be eliminated without prejudice to the rights or interests of any other group by extending coverage to same-sex couples who otherwise fulfil all of the other non-discriminatory criteria required in the Act. Deference under this branch of the s. 1 test is not appropriate when there is a reasonable alternative that is readily available, that is not the subject of conflicting social science views, and that could not result in a concomitant prejudice to another group. Finally, the deleterious effects of the impugned distinction outweigh its salutary effects. Per McLachlin J. (dissenting): The reasons of Cory and Iacobucci JJ. were substantially agreed with. On the basis of the principles outlined in Miron v. Trudel, released concurrently, the impugned legislation infringes s. 15(1) of the Charter and the infringement is not saved under s. 1. Cases Cited By La Forest J. Referred to: Canada (Attorney General) v. Ward, [1993] 2 S.C.R. 689; Miron v. Trudel, S.C.C., No. 22744, May 25, 1995; Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143; Symes v. Canada, [1993] 4 S.C.R. 695; R. v. Turpin, [1989] 1 S.C.R. 1296; R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295; United States of America v. Cotroni, [1989] 1 S.C.R. 1469; McKinney v. University of Guelph, [1990] 3 S.C.R. 229; Mahe v. Alta. (Gov't) (1987), 54 Alta. L.R. (2d) 212. By Sopinka J. Referred to: McKinney v. University of Guelph, [1990] 3 S.C.R. 229; Irwin Toy Ltd. v. Quebec (Attorney General), [1989] 1 S.C.R. 927. By Cory and Iacobucci JJ. (dissenting) R. v. Oakes, [1986] 1 S.C.R. 103; Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143; R. v. Turpin, [1989] 1 S.C.R. 1296; Symes v. Canada, [1993] 4 S.C.R. 695; Ontario Human Rights Commission v. Simpsons-Sears Ltd., [1985] 2 S.C.R. 536; Tetreault-Gadoury v. Canada (Employment and Immigration Commission), [1991] 2 S.C.R. 22; Brown v. Board of Education of Topeka, 347 U.S. 483 (1954); Brooks v. Canada Safeway Ltd., [1989] 1 S.C.R. 1219; Schachter v. Canada, [1992] 2 S.C.R. 679; Janzen v. Platy Enterprises Ltd., [1989] 1 S.C.R. 1252; R. v. Swain, [1991] 1 S.C.R. 933; Douglas v. Canada (1992), 58 F.T.R. 147; Haig v. Canada (1991), 5 O.R. (3d) 245 (Gen. Div.), aff'd (1992), 9 O.R. (3d) 495 (C.A.); Vriend v. Alberta (1994), 152 A.R. 1; Veysey v. Canada (Correctional Service) (1989), 44 C.R.R. 364; Brown v. British Columbia (Minister of Health) (1990), 42 B.C.L.R. (2d) 294; Knodel v. British Columbia (Medical Services Commission) (1991), 58 B.C.L.R. (2d) 356; Leshner v. Ontario (1992), 16 C.H.R.R. D/184; Weatherall v. Canada (Attorney General), [1993] 2 S.C.R. 872; R. v. Hess, [1990] 2 S.C.R. 906; Canada (Attorney General) v. Mossop, [1993] 1 S.C.R. 554; R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295. By L'Heureux-Dube J. (dissenting) R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295; Miron v. Trudel, S.C.C., No. 22744, May 25, 1995; Thibaudeau v. Canada, S.C.C., No. 24154, May 25, 1995; Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143; McKinney v. University of Guelph, [1990] 3 S.C.R. 229; R. v. Turpin, [1989] 1 S.C.R. 1296; Ontario Human Rights Commission v. Simpsons-Sears Ltd., [1985] 2 S.C.R. 536; R. v. Genereux, [1992] 1 S.C.R. 259; Weatherall v. Canada (Attorney General), [1993] 2 S.C.R. 872; R. v. Hess, [1990] 2 S.C.R. 906; Schachtschneider v. Canada, [1994] 1 F.C. 40; R. v. Oakes, [1986] 1 S.C.R. 103; Dagenais v. Canadian Broadcasting Corp., [1994] 3 S.C.R. 835; Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), [1990] 1 S.C.R. 1123; Symes v. Canada, [1993] 4 S.C.R. 695; Canada (Attorney General) v. Mossop, [1993] 1 S.C.R. 554; Bliss v. Attorney General of Canada, [1979] 1 S.C.R. 183; Brooks v. Canada Safeway Ltd., [1989] 1 S.C.R. 1219; Schachter v. Canada, [1992] 2 S.C.R. 679. By McLachlin J. (dissenting) Miron v. Trudel, S.C.C., No. 22744, May 25, 1995. Statutes and Regulations Cited B.C. Reg. 479/76, ss. 3(2), 10. Canadian Charter of Rights and Freedoms, ss. 1, 7, 15. Constitution Act, 1982, s. 52(1). Guaranteed Available Income for Need Act, R.S.B.C. 1979, c. 158, s. 8. N.B. Reg. 82-227. Old Age Security Act, R.S.C. 1970, c. O-6. Old Age Security Act, R.S.C., 1985, c. O-9, ss. 2, 19. Resolution on Equal Rights for Homosexuals and Lesbians in the European Community, European Parliament. A3-0028/94. Saskatchewan Assistance Act, R.S.S. 1978, c. S-8. Sask. Reg. 78/66. Social Welfare Act, R.S.N.B. 1973, c. S-11. Welfare Assistance Act, R.S.P.E.I. 1988, c. W-3. Welfare Assistance Act Regulations, EC746/84. Authors Cited Abella, Rosalie Silberman. Report of the Commission on Equality in Employment. Ottawa: Supply and Services Canada, 1984. Bayefsky, Anne F. "A Case Comment on the First Three Equality Rights Cases Under the Canadian Charter of Rights and Freedoms: Andrews, Workers' Compensation Reference, Turpin" (1990), 1 Supreme Court L.R. (2d) 503. Canada. House of Commons. Standing Committee on Health, Welfare and Social Affairs. Minutes of Proceedings and Evidence, June 12, 1975, p. 25:7. Canada. House of Commons. Sub-Committee on Equality Rights. Equality For All: Report of the Parliamentary Committee on Equality Rights. Ottawa: Queen's Printer, 1985. Duclos, Nitya. "An Argument for Legal Recognition of Same Sex Marriage". In Carol Rogerson, Family Law Cases and Materials 1991-92, vol. I. Faculty of Law, University of Toronto. Eichler, Margrit. Families in Canada Today, Recent Changes and their Policy Consequences, 2nd ed. Toronto: Gage, 1988. Ginson, Dale. "Analogous Grounds of Discrimination Under the Canadian Charter: Too Much Ado About Next to Nothing" (1991), 29 Alta. L. Rev. 772. Gibson, Dale. "Equality for Some" (1991), 40 U.N.B.L.J. 2. Gibson, Dale. The Law of the Charter: Equality Rights. Toronto: Carswell, 1990. Hogg, Peter W. Constitutional Law of Canada, 3rd ed. Scarborough, Ont.: Carswell, 1992. House of Commons Debates, vol. I, 1st Sess., 31st Parl., October 22, 1979, p. 476. Iyer, N. "Categorical Denials: Equality Rights and the Shaping of Social Identity" (1993), 19 Queen's L.J. 179. Lepofsky, M. David. "The Canadian Judicial Approach to Equality Rights: Freedom Ride or Rollercoaster?" (1992), 1 N.J.C.L. 315. Petersen, Cynthia. "A Queer Response to Bashing: Legislating Against Hate" (1991), 16 Queen's L.J. 237. Quebec. Commission des droits de la personne. Comite de consultation. De l'illegalite a l'egalite: Rapport de la consultation publique sur la violence et la discrimination envers les gais et lesbiennes. Montreal: La Commission, 1994. Smith, C. Lynn. "Judicial Interpretation of Equality Rights under the Canadian Charter of Rights and Freedoms: Some Clear and Present Dangers" (1988), 23 U.B.C. L. Rev. 65. Smith, Carolyn Gibson. "Proud but Cautious": Homophobic Abuse and Discrimination in Nova Scotia. Nova Scotia Public Interest Research Group, July 1994. Vanier Institute of the Family. Profiling Canada's Families. Ottawa: Vanier Institute of the Family, 1994. APPEAL from a judgment of the Federal Court of Appeal, [1993] 3 F.C. 401, 103 D.L.R. (4th) 336, 15 C.R.R. (2d) 310, 153 N.R. 161, affirming a judgment of the Federal Court, Trial Division, [1992] 1 F.C. 687, 87 D.L.R. (4th) 320, 38 R.F.L. (3d) 294, 47 F.T.R. 305, dismissing the appellants' claim. Appeal dismissed, L'Heureux-Dube, Cory, McLachlin and Iacobucci JJ. dissenting. Joseph J. Arvay, Q.C., and Leah Greathead, for the appellants. H. J. Wruck, Q.C., F. E. Campbell, Q.C., and L. M. Hitch, for the respondent. Madeleine Aube, for the intervener the Attorney General of Quebec. William F. Pentney and J. Helen Beck, for the intervener the Canadian Human Rights Commission. Helene Tessier, for the intervener the Commission des droits de la personne du Quebec. Cynthia Petersen, for the intervener Equality for Gays and Lesbians Everywhere. Charles M. Campbell and Susan Ursel, for the intervener Metropolitan Community Church of Toronto. Peter R. Jervis and Iain T. Benson, for the intervener Inter-Faith Coalition on Marriage and the Family. Steven Barrett and Vanessa Payne, for the intervener the Canadian Labour Congress. Solicitors for the appellants: Arvay Finlay, Victoria. Solicitor for the respondent: The Department of Justice, Ottawa. Solicitor for the intervener the Attorney General of Quebec: Madeleine Aube, Sainte-Foy. Solicitor for the intervener the Canadian Human Rights Commission: William F. Pentney, Ottawa. Solicitor for the intervener the Commission des droits de la personne du Quebec: Helene Tessier, Montreal. Solicitor for the intervener Equality for Gays and Lesbians Everywhere: Cynthia Petersen, Ottawa. Solicitors for the intervener Metropolitan Community Church of Toronto: Iler, Campbell, Toronto. Solicitors for the intervener Inter-faith Coalition on Marriage and the Family: Lerner & Associates, Toronto. Solicitors for the intervener the Canadian Labour Congress: Sack Goldblatt Mitchell, Toronto. PARAGRAPH NUMBERING The reasons are published in the SCR by order of precedence and the paragraph numbering included in those reasons has been prepared accordingly. For this case, the reasons will be published in the following order: La Forest J. (and Lamer C.J. and Gonthier and Major JJ. (paras. 1 to 30) L'Heureux-Dube J. (paras. 31 to 101) Sopinka J. (paras. 102 to 111) Iacobucci and Cory JJ. (paras. 112 to 230) McLachlin J. (para. 231) N.B.: The paragraph numbering will now appear in the SCR. LA FOREST J.: 1 This appeal concerns the constitutionality of ss. 2 and 19(1) of the Old Age Security Act, R.S.C., 1985, c. O-9, which accord to spouses of pensioners under the Act whose income falls below a stipulated amount, an allowance when they reach the age of 60, payable until they themselves become pensioners at age 65. The appellants maintain these provisions violate s. 15 of the Canadian Charter of Rights and Freedoms as discriminating against persons living in a homosexual relationship because the effect of the definition of "spouse" in s. 2 is to restrict the allowances to spouses in a heterosexual union, i.e. those who are legally married or who live in a common law relationship. 2 My colleague, Mr. Justice Cory, has set forth the facts and judicial history as well as the applicable constitutional and statutory provisions. It will be sufficient for me, then, to refer only to such of this material as may be essential to an understanding of what follows. 3 The provision providing for the payment of the allowance is s. 19(1), which reads as follows: 19. (1) Subject to this Act and the regulations, for each month in any fiscal year, a spouse's allowance may be paid to the spouse of a pensioner if the spouse (a) is not separated from the pensioner; (b) has attained sixty years of age but has not attained sixty-five years of age; and (c) has resided in Canada after attaining eighteen years of age and prior to the day on which the spouse's application is approved for an aggregate period of at least ten years and, where that aggregate period is less than twenty years, was resident in Canada on the day preceding the day on which the spouse's application is approved. That provision, without more, would be confined -- as the term "spouse" had been interpreted in the Act before the original enactment of s. 19(1) in 1975 --to spouses in a legal marriage. At the time of the enactment of s. 19(1), however, s. 2 of the Act was amended to define the term "spouse" for the purposes of the Act to include common law spouses described in the definition. The definition reads: 2. In this Act, . . . "spouse", in relation to any person, includes a person of the opposite sex who is living with that person, having lived with that person for at least one year, if the two persons have publicly represented themselves as husband and wife. The effect of this new definition was to extend the allowances to spouses living in a common law relationship as well as those in a legal marriage. 4 The appellants, James Egan and John Norris Nesbit, are homosexuals who have since 1948 lived together in a relationship marked by commitment and interdependence similar to that which one expects to find in a marriage. When Egan became 65 in 1986, he began to receive old age security and guaranteed income supplements under the Act. On reaching age 60, Nesbit applied for a spousal allowance under s. 19(1), which as I mentioned is available to spouses as defined in the Act between the ages of 60 and 65 whose combined income falls below a fixed level. His application was rejected because the relationship between Nesbit and Egan did not fall within the Act. 5 The appellants' claim before this Court is that the Act contravenes s. 15 of the Charter in that it discriminates on the basis of sexual orientation. To establish that claim, it must first be determined that s. 15's protection of equality without discrimination extends to sexual orientation as a ground analogous to those specifically mentioned in the section. This poses no great hurdle for the appellants; the respondent Attorney General of Canada conceded this point. While I ordinarily have reservations about concessions of constitutional issues, I have no difficulty accepting the appellants' contention that whether or not sexual orientation is based on biological or physiological factors, which may be a matter of some controversy, it is a deeply personal characteristic that is either unchangeable or changeable only at unacceptable personal costs, and so falls within the ambit of s. 15 protection as being analogous to the enumerated grounds. As the courts below observed, this is entirely consistent with a number of cases on the point. Indeed, there is a measure of support for this position in this Court. In Canada (Attorney General) v. Ward, [1993] 2 S.C.R. 689, at pp. 737-39, speaking for my colleagues as well, I observed that the analogous grounds approach in s. 15 was appropriate to a consideration of the character of "social groups" subject to protection as Convention refugees. These, I continued, encompass groups defined by an innate or unchangeable characteristic which, I added, would include sexual orientation. 6 The only difficulty I have with the Crown's concession is that it would seem difficult in the absence of more precise argument to consider the point mentioned by Gonthier J. in Miron v. Trudel, S.C.C., No. 22744, issued concurrently, regarding the need to consider the nature of a ground, be it enumerated in s. 15 or analogous, as a basis for discrimination and its necessary limitation where the distinction drawn by legislation merely reflects or is the expression of a fundamental reality or value. I need not pursue this, however, since I do not think that, in the circumstances of this case, the appellants have suffered discrimination under s. 15 as that term has been defined by this Court. 7 The nature of discrimination within the meaning of s. 15(1) of the Charter was first discussed by this Court in the seminal case of Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143. In the principal reasons in that case, McIntyre J., at p. 175, underlined the importance in a constitutional document, which is not easily modified, of achieving a workable balance that permits government to perform effectively its function of making ongoing choices in the interests of society and the work of the courts in ensuring protection for the equality rights described in s. 15. As he stated, what we must do is "to provide a continuing framework for the legitimate exercise of governmental power and, at the same time, for the unremitting protection" of equality rights. And he warned (see p. 168), as I did in my separate reasons, that not all distinctions resulting in disadvantage to a particular group will constitute discrimination. It would bring the legitimate work of our legislative bodies to a standstill if the courts were to question every distinction that had a disadvantageous effect on an enumerated or analogous group. This would open up a s. 1 enquiry in every case involving a protected group. As I put it in Andrews, at p. 194, "it was never intended in enacting s. 15 that it become a tool for the wholesale subjection to judicial scrutiny of variegated legislative choices in no way infringing on values fundamental to a free and democratic society". 8 What then is discrimination? There are several comments in the course of McIntyre J.'s remarks in Andrews that go a long way towards clarifying the concept. Thus, at p. 174, he stated: . . . I would say then that discrimination may be described as a distinction, whether intentional or not but based on grounds relating to personal characteristics of the individual or group, which has the effect of imposing burdens, obligations, or disadvantages on such individual or group not imposed upon others, or which withholds or limits access to opportunities, benefits, and advantages available to other members of society. This statement cannot, however, be looked at in isolation. It must be read in conjunction with McIntyre J.'s earlier comment, at p. 165, as follows: In other words, the admittedly unattainable ideal [of equality] should be that a law expressed to bind all should not because of irrelevant personal differences have a more burdensome or less beneficial impact on one than another. Similarly in my separate reasons, at p. 193, I observed that "the relevant question . . . is . . . whether the impugned provision amounts to discrimination in the sense in which my colleague has defined it, i.e. on the basis of `irrelevant personal differences' such as those listed in s. 15 . . .". 9 As Gonthier J. has noted in Miron v. Trudel, this involves a three-step analysis, which he puts this way (at pp. 8-9): The first step looks to whether the law has drawn a distinction between the claimant and others. The second step then questions whether the distinction results in disadvantage, and examines whether the impugned law imposes a burden, obligation or disadvantage on a group of persons to which the claimant belongs which is not imposed on others, or does not provide them with a benefit which it grants other (Andrews, supra). . . . The third step assesses whether the distinction is based on an irrelevant personal characteristic which is either enumerated in s. 15(1) or one analogous thereto. 10 There is no question that the first step is satisfied in this case. Parliament has clearly made a distinction between the claimant and others. This, of course, does not carry one very far. Parliament is in the business of making such distinctions in developing programs and policies which is the task assigned to it in our democratic system. Further ingredients must be added to warrant a distinction being discriminatory. 11 The second step will also, in general at least, not be of great assistance. Ordinarily decisions do result in advantages or disadvantages to individuals and groups, sometimes intentionally, sometimes unintentionally. Parliament, as I mentioned, is in the business of making choices, and this inevitably involves the distribution of benefits and burdens in our society. 12 In this case, however, the respondent contends that the appellants have suffered no prejudice because by being treated as individuals they have received considerably more in combined federal and provincial benefits than they would have received had they been treated as "spouses". I would simply dispose of this argument on the ground that, while this may be true in this specific instance, there is nothing to show that this is generally the case with homosexual couples, which is the point the respondent must establish. My colleague Cory J. also makes this point in his reasons. 13 I turn then to the third step of the analysis described by Gonthier J. Since it has already been accepted that "sexual orientation" is an analogous ground under s. 15(1), all that remains to be considered under this step is whether the distinction made by Parliament is relevant, what Gonthier J. describes in Miron v. Trudel, at p. 10, as the second aspect of this third step. He there notes that in assessing relevancy for this purpose one must look at "the nature of the personal characteristic and its relevancy to the functional values underlying the law". At this stage, he adds, one must necessarily undertake a form of comparative analysis to determine whether particular facts give rise to inequality; see also Symes v. Canada, [1993] 4 S.C.R. 695, at p. 754. This proposition, too, derives from McIntyre J.'s reasons in Andrews, supra, who, at p. 164, states: It [equality] is a comparative concept, the condition of which may only be attained or discerned by comparison with the condition of others in the social and political setting in which the question arises. It must be recognized at once, however, that every difference in treatment between individuals under the law will not necessarily result in inequality and, as well, that identical treatment may frequently produce serious inequality. Gonthier J. adds, citing Wilson J. in R. v. Turpin, [1989] 1 S.C.R. 1296, at pp. 1331-32, that this comparative analysis must be linked to an examination of the larger context, and in particular with an understanding that the Charter was, in Dickson J.'s words, "not enacted in a vacuum", but must "be placed in its proper linguistic, philosophic and historical contexts" if we are to avoid mechanical and sterile categorization; see R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295, at p. 344; see also United States of America v. Cotroni, [1989] 1 S.C.R. 1469, at pp. 1490-91. 14 In embarking upon this comparative analysis, I shall begin with an examination of the statute with a view to determining "the functional values underlying the law". I shall then examine the personal characteristic here in issue to determine its relevancy to these functional values. 15 In undertaking an examination of the statute, it is now settled that one must not focus narrowly on a provision that has the effect of depriving a group of a benefit that others who initially appear to be in a similar position are accorded. This is the "similarly situated test" which has been categorically rejected by this Court; see Andrews; see also McKinney v. University of Guelph, [1990] 3 S.C.R. 229, at p. 279. Among the reasons for its rejection is that put forth by Kerans J.A. in Mahe v. Alta. (Gov't) (1987), 54 Alta. L.R. (2d) 212, at p. 244 (in a passage approved by McIntyre J. in Andrews, at p. 168), namely: . . . the test accepts an idea of equality which is almost mechanical, with no scope for considering the reason for the distinction. In consequence, subtleties are found to justify a finding of dissimilarity which reduces the test to a categorization game. Moreover, the test is not helpful. After all, most laws are enacted for the specific purpose of offering a benefit or imposing a burden on some persons and not on others. The test catches every conceivable difference in legal treatment. [Emphasis added.] 16 Having quoted this passage, McIntyre J. in Andrews immediately stated that such a fixed rule or formula cannot be accepted, and then concluded by setting forth how the relevant statute must be approached in the following remarks, at p. 168: Consideration must be given to the content of the law, to its purpose, and its impact upon those to whom it applies, and also upon those whom it excludes from its application. The issues which will arise from case to case are such that it would be wrong to attempt to confine these considerations within such a fixed and limited formula. It is in that spirit that I propose to examine the Old Age Security Act. 17 Mahoney J.A. in the Court of Appeal has usefully described the broad structure of the Act in the following passage (at p. 411): Part I of the Act provides for the payment of a monthly pension at age 65 to Canadian citizens and permanent residents. That pension is payable on application regardless of need. Part II provides for an additional payment to a pensioner, called a guaranteed income supplement, on the basis of need. The income of the pensioner's spouse, if any, is taken into account in determining need. Part III provides for the payment of a monthly spouse's allowance, at age 60, to the spouse of a pensioner who is in receipt of a guaranteed income supplement. As is evident from this description, Parliament, in addition to providing greater benefits to the elderly in need, long ago took special account of married couples in need; as I mentioned earlier, before 1975 the term "spouse" only applied to persons who were legally married. This special interest is clearly expressed by the Minister of Health and Welfare, the Honourable Marc Lalonde when testifying before the Standing Committee on Health, Welfare and Social Affairs in relation to the amendment adding the spousal allowance in 1975. He stated: Its objective is clear and singular in purpose. It is to ensure that when a couple is in a situation where one of the spouses has been forced to retire, and that couple has to live on the pension of a single person, that there should be a special provision, when the breadwinner has been forced to retire at or after 65, to make sure that particular couple will be able to rely up on an income which would be equivalent to both members of the couple being retired or 60 [sic] years of age and over. That is the purpose of the Bill, no more than that, no less than that. See Minutes of Proceedings and Evidence, June 12, 1975, at p. 25:7. 18 I add that other evidence of Parliament's continuing concern for the needs of married couples is the benefits for widows. To complete the picture, I repeat that in 1975 Parliament, by defining spouse as above described, extended the benefits under the Act beyond those who were legally married to common law couples, and it is that definition that has formed the principal focus of the appellants' attack. As I earlier noted, however, it is dangerous to focus narrowly on a particular provision. A more comprehensive contextual approach must be taken to determine the relevancy of the personal characteristic in question to the functional values underlying the law. That is how I propose to proceed. 19 The singling out of legally married and common law couples as the recipients of benefits necessarily excludes all sorts of other couples living together such as brothers and sisters or other relatives, regardless of sex, and others who are not related, whatever reasons these other couples may have for doing so and whatever their sexual orientation. Mahoney J.A., at p. 412 of his reasons in the Court of Appeal, lucidly describes these various couples in the following passage: Many couples live together in relationships excluded from the definition. Cohabitation by siblings is a commonplace example; persons otherwise related by blood or marriage do so as well and so do persons not related. They do so for countless personal reasons and combinations thereof, for example: mere convenience, the advantage of pooling resources, shared interests, congeniality, friendship and affection not involving sexual attraction, to have someone with them in disability, failing health or in fear of it, or simply to avoid loneliness and seclusion. Unless subjective pressures are in play, sex, whether same or opposite, need not be a consideration in the choice of a live-in companion. There are those, like the appellants, whose sexual orientation is a determining factor in their choice of partner. Many, possibly most, of those couples do not publicly represent themselves as spouses so that they would benefit from the remedy the appellants seek. 20 What reason or purpose, then, can be assigned to the distinction made by Parliament? It seems to me that it is both obvious and deeply rooted in our fundamental values and traditions, values and traditions that could not have been lost on the framers of the Charter. Simply stated, what Parliament clearly had in mind was to accord support to married couples who were aged and elderly, and this for the advancement of public policy central to society. Moreover, in recognition of changing social realities, s. 2 was amended so that whenever the term "spouse" was used in the Act it was to be construed to extend beyond legal married couples to couples in a common law marriage. 21 My colleague Gonthier J. in Miron v. Trudel has been at pains to discuss the fundamental importance of marriage as a social institution, and I need not repeat his analysis at length or refer to the authorities he cites. Suffice it to say that marriage has from time immemorial been firmly grounded in our legal tradition, one that is itself a reflection of long standing philosophical and religious traditions. But its ultimate raison d'etre transcends all of these and is firmly anchored in the biological and social realities that heterosexual couples have the unique ability to procreate, that most children are the product of these relationships, and that they are generally cared for and nurtured by those who live in that relationship. In this sense, marriage is by nature heterosexual. It would be possible to legally define marriage to include homosexual couples, but this would not change the biological and social realities that underlie the traditional marriage. 22 The marital relationship has special needs with which Parliament and the legislatures and indeed custom and judge-made law have long been concerned. The legal institution of marriage exists both for the protection of the relationship and for defining the obligations that flow from entering into a legal marriage. Because of its importance, legal marriage may properly be viewed as fundamental to the stability and wellbeing of the family and, as such, as Gonthier J. argued in Miron v. Trudel, Parliament may quite properly give special support to the institution of marriage. It is spouses in legal marriage who constitute the bulk of the beneficiaries of spousal allowances. 23 But many of the underlying concerns that justify Parliament's support and protection of legal marriage extend to heterosexual couples who are not legally married. Many of these couples live together indefinitely, bring forth children and care for them in response to familial instincts rooted in the human psyche. These couples have need for support just as legally married couples do in performing this critical task, which is of benefit to all society. Language has long captured the essence of this relationship by the expression "common law marriage". 24 Faced with the social reality that increasing numbers choose not to enter a legal marriage but live together in a common law relationship, Parliament has elected to support these relationships. The legal institution of marriage has long been viewed as the fundamental instrument to promote the underlying values I have referred to. But Parliament cannot force people to get married, and I see no reason why it should not take the necessary means to promote the basic social interests and policies that inform the institution of legal marriage through other instrumentalities. Support of common law relationships with a view to promoting their stability seems well devised to advance many of the underlying values for which the institution of marriage exists. For example, children brought up by single parents more often end up in poverty and impose greater burdens on society. Parliament, it seems to me, is wholly justified in extending support to heterosexual couples like this, which is not to say, however, that it is obligated to do so and may not treat married and unmarried couples differently (see the reasons of Gonthier J. in Miron). 25 Viewed in the larger context, then, there is nothing arbitrary about the distinction supportive of heterosexual family units. And for the reasons set forth by Gonthier J. in Miron, I am not troubled by the fact that not all these heterosexual couples in fact have children. It is the social unit that uniquely has the capacity to procreate children and generally cares for their upbringing, and as such warrants support by Parliament to meet its needs. This is the only unit in society that expends resources to care for children on a routine and sustained basis. As counsel for the intervener the Interfaith Coalition on Marriage and the Family put it, whether the mother or the father leaves the paid workforce or whether both parents are paying after-tax dollars for daycare, this is the unit in society that fundamentally anchors other social relationships and other aspects of society. I add that I do not think the courts should attempt to require meticulous line drawing that would ensure that only couples that had children were included. This could impose on Parliament the burden of devising administrative procedures to ensure conformity that could be both unnecessarily intrusive and difficult to administer, thereby depriving Parliament of that "reasonable room to manoeuvre" which this Court has frequently recognized as necessary; see United States of America v. Cotroni, supra, at p. 1495, and the cases there cited. This I think is wholly consistent with the workable balance between Parliament and the courts sought to be achieved in Andrews, supra, to which I have earlier referred. 26 Neither in its purpose or effect does the legislation constitute an infringement of the fundamental values sought to be protected by the Charter. None of the couples excluded from benefits under the Act are capable of meeting the fundamental social objectives thereby sought to be promoted by Parliament. These couples undoubtedly provide mutual support for one another, and that, no doubt, is of some benefit to society. They may, it is true, occasionally adopt or bring up children, but this is exceptional and in no way affects the general picture. I fail to see how homosexuals differ from other excluded couples in terms of the fundamental social reasons for which Parliament has sought to favour heterosexuals who live as married couples. Homosexual couples, it is true, differ from other excluded couples in that their relationships include a sexual aspect. But this sexual aspect has nothing to do with the social objectives for which Parliament affords a measure of support to married couples and those who live in a common law relationship. 27 In a word, the distinction made by Parliament is grounded in a social relationship, a social unit that is fundamental to society. That unit, as I have attempted to explain, is unique. It differs from all other couples, including homosexual couples. Other excluded couples, it is true, do not have to be described by reference to sex or sexual preferences, but this is of no moment. The distinction adopted by Parliament is relevant, indeed essential, to describe the relationship in the way the statute does so as to differentiate the couples described in the statute from all couples that do not serve the social purposes for which the legislature has made the distinction. Homosexual couples are not, therefore, discriminated against; they are simply included with these other couples. 28 I add that this distinction exists in a plethora of statutes, both federal and provincial, and, indeed, it directly or indirectly forms the substratum of an abundance of legal principles and rules at common law and under the civil law system. I realize, of course, that the distinction could in certain circumstances be used in a discriminatory manner, but that is not this case. It is relevant here to describe a fundamental social unit, indeed the fundamental unit in society, to which some measure of support is given. I add, interstitially, that this support does not exacerbate an historic disadvantage; rather it ameliorates an historic economic disadvantage, both for couples who are legally married and those who live in a common law relationship. If the distinction is thought to be irrelevant here, this would, in my view, mean that the courts would have to embark upon a s. 1 justification every time a distinction was made on the basis of marriage, legal or common law. Moreover, it would interfere with the appropriate balance between legislative and judicial power described in Andrews, which I have discussed earlier in these reasons. 29 Had I concluded that the impugned legislation infringed s. 15 of the Charter, I would still uphold it under s. 1 of the Charter for the considerations set forth in my reasons in McKinney v. University of Guelph, [1990] 3 S.C.R. 229, especially at pp. 316-18, some of which are referred to in the reasons of my colleague Justice Sopinka, as well as for those mentioned in my discussion of discrimination in the present case. 30 I would dismiss the appeal and would answer the constitutional questions as follows: 1.Does the definition of "spouse" in s. 2 of the Old Age Security Act, R.S.C., 1985, c. O-9, infringe or deny s. 15(1) of the Canadian Charter of Rights and Freedoms? No. 2.If the answer to question 1 is yes, is the infringement or denial demonstrably justified in a free and democratic society pursuant to s. 1 of the Canadian Charter of Rights and Freedoms? Assuming it is an infringement, it is justifiable under s. 1 of the Charter. L'HEUREUX-DUBE J.: 31 This appeal raises the question of whether a legislative distinction that limits eligibility for a spousal supplement under the Old Age Security Act, R.S.C., 1985, c. 0-9, to "opposite sex" spouses is discriminatory within the meaning of s. 15 of the Canadian Charter of Rights and Freedoms and, if so, whether it is saved by s. 1 of the Charter. Although I agree with much of what is said by my colleagues Justices Cory and Iacobucci, as well as with the result they reach, I have some concerns as to the proper approach to be taken to s. 15 of the Charter, which I shall outline below. I. Analysis 32 There is no more important task in approaching any Charter right than that of characterizing properly its purpose: R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295. Defining with accuracy and sensitivity the purpose of a particular right is, in short, the starting point for rights analysis. By implication, a right is said to be violated when the purpose of that right is denied, undermined, or frustrated by legislative action. Disagreement, no matter how small, at the foundational level of establishing the right's purpose will only magnify over time in terms of how that right is applied. More difficult cases, quite naturally, will only make these differences more apparent. I believe that this phenomenon is beginning to manifest itself in the divergent approaches to s. 15 taken in recent cases before this Court, of which this case, Miron v. Trudel, No. 22794, and Thibaudeau v. Canada, No. 24154, released concurrently, are no exception. The emergence of these differences suggests to me that we may not necessarily be operating with the same underlying purpose in mind. For s. 15 jurisprudence to continue to develop along principled lines, I believe that two things are necessary: (1) we must revisit the fundamental purpose of s. 15; and (2) we must seek out a means by which to give full effect to this fundamental purpose. I set out below my efforts to begin this dialogue. A. The Purpose of s. 15 33 Section 15 does not guarantee the complete absence of distinctions. Nor, for that matter, does it guarantee equality in the abstract. Indeed, as this Court recognized in Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143, at p. 169, "for the accommodation of differences, which is the essence of true equality, it will frequently be necessary to make distinctions". The key, then, is to define when those distinctions are constitutionally permissible, and when they are not. 34 As an important starting point to evaluating the purpose of s. 15(1), we need look no further than its text. It reads: 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. [Emphasis added.] It is plain from the language of s. 15 that its fundamental purpose is to guarantee to all individuals a certain kind of equality: equality without discrimination. By implication, where "discrimination" is not present, then the Charter guarantee of equality is satisfied. The nine "grounds" enumerated after this basic guarantee to freedom from discrimination are particular applications and illustrations of the ambit of s. 15. They are not the guarantee itself. 35 If the fundamental purpose of s. 15 is to guarantee equality "without discrimination", then it follows that the pivotal question is, "How do we define 'discrimination'"? Under the approach set out in Andrews, this Court has sought to define "discrimination" by reference to the nine grounds enumerated in s. 15(1) as well as by reference to "analogous grounds", which embody characteristics seen to be held in common by the enumerated grounds. As several commentators have suggested, and as I shall argue shortly, this is an indirect means by which to define discrimination. A preferable approach would be to give independent content to the term "discrimination", and to develop s. 15 along the lines of that definition. In so doing, however, we must heed the warning sounded by Professor Lynn Smith in "Judicial Interpretation of Equality Rights under the Canadian Charter of Rights and Freedoms: Some Clear and Present Dangers" (1988), 23 U.B.C. L. Rev. 65, at p. 86: ... I think that there is a real risk that if the courts continue to review all claims which the ingenious legal mind can create, and if the courts decide that the standard of review will be the same no matter what the claim, then that standard will tend toward the lowest common denominator, or at least toward the standard appropriate to the bulk of the cases being brought. Such a standard will not be a high one- -that is, it will be an easy one for government to pass. This tendency, if it occurs, will then result in a watered-down level of protection for all claims. [Emphasis added.] In short, the standard must not be so broad or vague as to risk undermining in practice the very purposes that s. 15 is intended to further. 36 This Court has recognized that inherent human dignity is at the heart of individual rights in a free and democratic society: Big M Drug Mart Ltd., supra, at p. 336 (per Dickson J. (as he then was)). More than any other right in the Charter, s. 15 gives effect to this notion. Building upon this foundation, I believe that the essence of "discrimination" was largely captured by McIntyre J., speaking for the majority of the Court on this point, in Andrews, supra, at p. 171: It is clear that the purpose of s. 15 is to ensure equality in the formulation and application of the law. The promotion of equality entails the promotion of a society in which all are secure in the knowledge that they are recognized as human beings equally deserving of concern, respect and consideration. [Emphasis added.] Equality, as that concept is enshrined as a fundamental human right within s. 15 of the Charter, means nothing if it does not represent a commitment to recognizing each person's equal worth as a human being, regardless of individual differences. Equality means that our society cannot tolerate legislative distinctions that treat certain people as second-class citizens, that demean them, that treat them as less capable for no good reason, or that otherwise offend fundamental human dignity. In a similar vein, I refer to the words of Wilson J. in McKinney v. University of Guelph, [1990] 3 S.C.R. 229, at p. 387 (dissenting, but not on this point): It is, I think, now clearly established that what lies at the heart of s. 15(1) is the promise of equality in the sense of freedom from the burdens of stereotype and prejudice in all their subtle and ugly manifestations. However, the nature of discrimination is such that attitudes rather than laws or rules may be the source of the discrimination. [Emphasis added.] 37 We can further inform our understanding of the purpose of s. 15 by recognizing what it is not. The Charter is a document of civil, political and legal rights. It is not a charter of economic rights. This is not to say, however, that economic prejudices or benefits are irrelevant to determinations under s. 15 of the Charter. Quite the contrary. Economic benefits or prejudices are relevant to s. 15, but are more accurately regarded as symptomatic of the types of distinctions that are at the heart of s. 15: those that offend inherent human dignity. 38 Finally, we must bear in mind that it has been recognized by this Court that an important, though not necessarily exclusive, purpose of s. 15 is the prevention or reduction of distinctions that may worsen the circumstances of those who have already suffered marginalization or historical disadvantage in our society: Andrews, supra; R. v. Turpin, [1989] 1 S.C.R. 1296. See also Lepofsky, "The Canadian Judicial Approach to Equality Rights: Freedom Ride or Rollercoaster?" (1992), 1 N.J.C.L. 315. In many ways, this recognition flows from the fact that discrimination is as much an effect as a purpose, and that those individuals and groups that are more vulnerable in society are also more likely to experience the effects of a distinction more severely. This important relationship was recognized in Judge Abella's groundbreaking Report of the Commission on Equality in Employment (1984), at p. 9: The impact of behaviour is the essence of "systemic discrimination". It suggests that the inexorable, cumulative effect on individuals or groups of behaviour that has an arbitrarily negative impact on them is more significant than whether the behaviour flows from insensitivity or intentional discrimination.... The effect of the system on the individual or group, rather than its attitudinal sources, governs whether or not a remedy is justified. [Emphasis added.] As such, s. 15 requires that we place both the group and the particular distinction adversely affecting that group within the "social, political and legal fabric of our society": Andrews, supra, at p. 152. 39 To summarize, at the heart of s. 15 is the promotion of a society in which all are secure in the knowledge that they are recognized at law as equal human beings, equally capable, and equally deserving. A person or group of persons has been discriminated against within the meaning of s. 15 of the Charter when members of that group have been made to feel, by virtue of the impugned legislative distinction, that they are less capable, or less worthy of recognition or value as human beings or as members of Canadian society, equally deserving of concern, respect, and consideration. These are the core elements of a definition of "discrimination"-- a definition that focuses on impact (i.e. discriminatory effect) rather than on constituent elements (i.e. the grounds of the distinction). 40 Three clarifications should be made at this juncture. First, I acknowledge that the above definition essentially tries to put into words the notion of fundamental human dignity. Dignity being a notoriously elusive concept, however, it is clear that this definition cannot, by itself, bear the weight of s. 15's task on its shoulders. It needs precision and elaboration. I shall attempt to demonstrate shortly how this approach to discrimination can find more concrete and principled expression using many of the criteria that have in the past proven themselves to be highly apposite under the approach taken by this Court in Andrews. As such, it will become evident that the approach I suggest is far less a departure from that developed in Andrews than may appear at first blush. I believe many of those analytical tools to be valid. The problem, in my mind, lies not with the tools but with the framework within which they have in the past been employed. In short, if the framework is not perfectly suited for the tools, then we do not use the tools to their full potential. 41 Second, I note that although the utopian ideal would be a society in which nobody is made to feel debased, devalued or denigrated as a result of legislative distinctions, such an ideal is clearly unrealistic. The guarantee against discrimination cannot possibly hold the state to a standard of conduct consistent with its most sensitive citizens. Clearly, a measure of objectivity must be incorporated into this determination. This being said, however, it would be ironic and, in large measure, self- defeating to the purposes of s. 15 to assess the absence or presence of discriminatory impact according to the standard of the "reasonable, secular, able-bodied, white male". A more appropriate standard is subjective- objective -- the reasonably held view of one who is possessed of similar characteristics, under similar circumstances, and who is dispassionate and fully apprised of the circumstances. The important principle, however, which this Court has accepted, is that discriminatory effects must be evaluated from the point of view of the victim, rather than from the point of view of the state. 42 Third, I feel compelled to comment on the role of "relevance" in s. 15. Relevance has been advocated by commentators such as Professor Gibson as a means of rendering s. 15 more open-ended, so as to extend its ambit beyond that endorsed by this Court in Andrews. If I understand him correctly, he would treat "irrelevance" as an integral aspect of a finding of discrimination. See Gibson, "Equality for Some" (1991), 40 U.N.B.L.J. 2; "Analogous Grounds of Discrimination Under the Canadian Charter: Too Much Ado About Next to Nothing" (1991), 29 Alta. L. Rev. 772; The Law of the Charter: Equality Rights (1990). Although I commend his efforts to ensure that s. 15 not be overly limited in its scope, I cannot agree with the means by which he seeks to achieve such open-endedness. 43 Briefly put, my concern is that "relevance" is a double- edged sword. Although there will certainly be instances in which irrelevant distinctions will trigger s. 15 scrutiny where s. 15 might not otherwise have been engaged, I find it equally easy to envision situations where relevance will be used to deny protections under s. 15 to groups that are otherwise deserving of it. In particular, where a distinction is relevant to the purpose of the legislation, then it is not discriminatory for the purposes of s. 15. In my view, such an approach takes far too narrow a view of discrimination. Relevance can, by definition, only be evaluated as against the purpose of the impugned legislation. Consequently, it fails to take into account the possibility that a distinction that is relevant to the purpose of the legislation may nonetheless still have a discriminatory effect. If s. 15 is about recognizing the equal worth and dignity of each human being, it seems counter- productive to say that this sense of equal worth has not been impugned merely because the legislative distinction is relevant to some legitimate legislative purpose. Who are we, for instance, to say to persons over the age of 65 that, because mandatory retirement is relevant to several important social goals, those persons cannot reasonably feel demeaned as human beings by mandatory retirement legislation? 44 Using relevance to define the absence or presence of discrimination raises other difficulties. It is no good, for instance, for a distinction to be relevant to a legislative purpose if that purpose is, itself, discriminatory. Moreover, including a "relevance" test within a s. 15 determination would place the onus on the rights claimant to characterize properly the purpose of the legislation which it claims is irrelevant to the impugned distinction. This strikes me as undesirable, since between the rights claimant and the government, the government is clearly in the superior position to characterize properly the purpose of its own legislation. Moreover, it is unclear to me what standard of relevance would be appropriate for such a determination. To find a distinction to be discriminatory, it seems quite reasonable, for instance, to require a lesser degree of irrelevance where the consequences of the distinction on the affected individuals are severe than where they are minimal. Finally, a "relevancy" standard would appear to impose an internal limitation on s. 15 that does not arise naturally from its plain language. Given that the decision to impose a limitation at the level of the right itself has consequences in terms of who must bear the onus of overcoming the limitation, this Court has consistently eschewed imposing internal limitations on rights wherever possible: Andrews, supra, at p. 178. 45 In sum, I believe that it is more accurate and more desirable to treat relevance as, in fact, a justification for distinctions that have a discriminatory impact on persons or groups, to be considered under s. 1 of the Charter. I shall elaborate upon this matter below. B. Giving Effect to the Purpose of Section 15 46 I will first discuss why I believe that the current vehicle of choice for fulfilling the purposes of s. 15, the "grounds" approach, is incapable of giving full effect to this purpose. I will then elaborate upon an approach that I believe to be more capable of enabling s. 15 to realize its full potential. (i) The Imperfect Vehicle of "Grounds" 47 In order to realize fully the purpose of Charter rights, it is necessary to look to the effects of impugned legislative actions. In the context of s. 15, no intention to discriminate need be demonstrated in order to render a particular distinction discriminatory. In Andrews, supra, in the course of his discussion on the nature of discrimination, McIntyre J. referred to the conclusions of this Court in the human rights case of Ontario Human Rights Commission v. Simpsons-Sears Ltd., [1985] 2 S.C.R. 536. I believe the following observation, at p. 173 of Andrews, to be at the core of this Court's philosophy regarding the notion of discrimination: ... [n]o intent was required as an element of discrimination, for it is in essence the impact of the discriminatory act or provision upon the person affected which is decisive in considering any complaint. [Emphasis added.] It is here, however, that the emphasis on grounds and, correspondingly, on "analogous grounds", developed in Andrews runs into the greatest difficulty. Rather than focus on "discrimination" directly, the "grounds" approach focuses courts' attentions on the types of grounds which may be a basis for a finding of discrimination. Because a finding that a ground is either enumerated or analogous is a necessary pre- condition for a finding of discrimination, most analysis is devoted toward characterizing the basis for the distinction and, if the basis is not an enumerated ground, deciding whether the ground is "analogous". This approach inquires into whether the characteristics of the ground are sufficient to constitute a basis for discrimination, rather than into the absence or presence of discriminatory effects themselves. 48 We must remember that the grounds in s. 15, enumerated and analogous, are instruments for finding discrimination. They are a means to an end. By focusing almost entirely on the nature, content and context of the disputed ground, however, we have begun to approach it as an end, in and of itself. Such an approach, in effect, approaches s. 15 not by giving primacy to the word "discrimination", but rather by giving primacy to the nine enumerated grounds. In essence, it defines the preconditions to when discrimination will be present exclusively by reference to qualities seen generally to reside in those grounds. 49 It is obvious that this Court could not have adopted an enumerated and analogous grounds approach if, instead of there being nine enumerated grounds in s. 15(1), there had been none. Would the absence of "particularities" in s. 15(1) have changed the basic guarantee to equality without discrimination? In the alternative, what would have happened under the "analogous grounds" approach if, instead of setting out nine enumerated grounds, s. 15(1) had set out only three or four? What if, furthermore, religion was not one of them? Most would agree that the common characteristics of all of the enumerated grounds other than religion is that they involve so-called "immutable" characteristics. Religion, on the other hand, has been described as being premised on a "fundamental choice". Does this mean that s. 15, despite being consciously left open-ended by the drafters, could never have encompassed discrimination on the basis of religion, or any other characteristic which involves a "fundamental choice"? This result seems absurd, yet it seems to flow inevitably from an approach to "discrimination" that relies exclusively on drawing analogies from the essential characteristics of the enumerated grounds. It also demonstrates, in my mind, why reliance on characteristics "analogous" to those in the enumerated grounds is a potentially unsatisfactory means of giving effect to s. 15's open-ended character. 50 Additional problems arise when certain grounds, particularly grounds based upon legal status (marital status, family status, citizenship, province of residence, etc.) may be said to give rise to discriminatory concerns in certain contexts but not in others. Are these grounds therefore sometimes analogous and sometimes not analogous? In these types of circumstances, the finding of "analogousness" will be driven by the result we want to reach. If we want to conclude that the impugned distinction is discriminatory, then we find the grounds to be analogous. If we want to conclude that a distinction is non-discriminatory, then we simply say that although the ground "may be analogous in some contexts", it is not in this case: see, e.g., Turpin, supra, per Wilson J.; R. v. Genereux, [1992] 1 S.C.R. 259, per Lamer C.J.. 51 In addition to defining discrimination indirectly, a "grounds"-based approach suffers an additional shortcoming. Briefly put, "grounds" are, themselves, an imperfect means for discerning discriminatory conduct. This problem was underlined by Wilson J. in McKinney, supra, at pp. 392-93: The grounds enumerated in s. 15 represent some blatant examples of discrimination which society has at last come to recognize as such .... The listing of sex, age and race, for example, is not meant to suggest that any distinction drawn on these grounds is per se discriminatory. Their enumeration is intended rather to assist in the recognition of prejudice when it exists .... It follows, in my opinion, that the mere fact that the distinction drawn in this case has been drawn on the basis of age does not automatically lead to some kind of irrebuttable presumption of prejudice. Rather it compels one to ask the question: is there prejudice? Is the mandatory retirement policy a reflection of the stereotype of old age? Is there an element of human dignity at issue? [Emphasis added.] If a finding of discrimination does not flow automatically from a finding that a distinction has been drawn on the basis of an enumerated or analogous ground (see Weatherall v. Canada (Attorney General), [1993] 2 S.C.R. 872; R. v. Hess, [1990] 2 S.C.R. 906), then it follows that reliance on "grounds" may not contemplate the entire picture. An additional dimension of analysis is needed. 52 At this juncture, an important question must be asked. If the purpose of s. 15 is really to provide a broad guarantee of protection against discrimination in all of its forms, then why does it matter if the basis for distinction is abstractly "analogous" to the enumerated categories? The answer, I think, is that it does not matter. As this Court has frequently acknowledged, the essence of discrimination is its impact, not its intention. The enumerated or analogous nature of a given ground should not be a necessary precondition to a finding of discrimination. If anything, a finding of discrimination is a precondition to the recognition of an analogous ground. The effect of the "enumerated or analogous grounds" approach may be to narrow the ambit of s. 15, and to encourage too much analysis at the wrong level. 53 We will never address the problem of discrimination completely, or ferret it out in all its forms, if we continue to focus on abstract categories and generalizations rather than on specific effects. By looking at the grounds for the distinction instead of at the impact of the distinction on particular groups, we risk undertaking an analysis that is distanced and desensitized from real people's real experiences. To make matters worse, in defining the appropriate categories upon which findings of discrimination may be based, we risk relying on conventions and stereotypes about individuals within these categories that, themselves, further entrench a discriminatory status quo. More often than not, disadvantage arises from the way in which society treats particular individuals, rather than from any characteristic inherent in those individuals. 54 For all of these reasons, I am led inevitably to the conclusion that a truly purposive approach to s. 15 must place "discrimination" first and foremost in the Court's analysis. This is not to say that the essential characteristics of the nine enumerated grounds are irrelevant to our inquiry. They are, in fact, highly relevant. I turn now to a discussion of their important role in an approach that looks to groups rather than grounds, and discriminatory impact rather than discriminatory potential. (ii) Putting "Discrimination" First 55 In my view, for an individual to make out a violation of their rights under s. 15(1) of the Charter, he or she must demonstrate the following three things: (1) that there is a legislative distinction; (2) that this distinction results in a denial of one of the four equality rights on the basis of the rights claimant's membership in an identifiable group; (3) that this distinction is "discriminatory" within the meaning of s. 15. The following remarks are devoted to elaborating upon the last criterion. 56 A distinction is discriminatory within the meaning of s. 15 where it is capable of either promoting or perpetuating the view that the individual adversely affected by this distinction is less capable, or less worthy of recognition or value as a human being or as a member of Canadian society, equally deserving of concern, respect, and consideration. This examination should be undertaken from a subjective-objective perspective: i.e. from the point of view of the reasonable person, dispassionate and fully apprised of the circumstances, possessed of similar attributes to, and under similar circumstances as, the group of which the rights claimant is a member. 57 The means by which courts may give principled expression to this notion is perhaps best illustrated by a simple analogy. If a projectile were thrown against a soft surface, then it would leave a larger scar than if it were thrown against a resilient surface. In fact, the depth of the scar inflicted will generally be a function of both the nature of the affected surface and the nature of the projectile used. In my view, assessing discriminatory impact is, in principle, no different. In order for a court to determine from a subjective- objective perspective whether the impugned distinction will leave a non-trivial discriminatory "scar" on the group affected, it is instructive to consider two categories of factors: (1) the nature of the group adversely affected by the distinction and (2) the nature of the interest adversely affected by the distinction. In my view, neither is completely meaningful without the other. The nature of the group affected 58 No one would dispute that two identical projectiles, thrown at the same speed, may nonetheless leave a different scar on two different types of surfaces. Similarly, groups that are more socially vulnerable will experience the adverse effects of a legislative distinction more vividly than if the same distinction were directed at a group which is not similarly socially vulnerable. As such, a distinction may be discriminatory in its impact upon one group yet not discriminatory in its impact upon another group. While it may be discriminatory against women to prohibit female guards from searching male prisoners, it may not be discriminatory against men to prohibit male guards from searching female prisoners: Weatherall v. Canada (Attorney General), supra. While it may be discriminatory to define a particular criminal offence as only applying to women, it may not be discriminatory to restrict the applicability of the offence of sexual assault of a minor to men: R. v. Hess, supra. In the same way that it does not really matter why the affected surface is soft, it is not necessary that there be a formal nexus between the social vulnerability of the affected group and the prejudice flowing from the impugned distinction in order for that vulnerability to be relevant to determining whether the distinction is discriminatory. Put another way, it is merely admitting reality to acknowledge that members of advantaged groups are generally less sensitive to, and less likely to experience, discrimination than members of disadvantaged, socially vulnerable or marginalized groups. See, by analogy, Schachtschneider v. Canada, [1994] 1 F.C. 40 (C.A.), per Linden J.A.. 59 Most of the factors identified in Andrews under the "analogous grounds" approach as characteristic of the enumerated grounds in s. 15 are, not surprisingly, integral to evaluating the nature of the group affected by the impugned distinction. It is highly relevant, for instance, to inquire into whether the impugned distinction is based upon fundamental attributes, such as those enumerated in s. 15, that are generally considered to be essential to our popular conception of 'personhood' or 'humanness'. Furthermore, it is important to ask ourselves questions such as "Is the adversely affected group already a victim of historical disadvantage?"; "Is this distinction reasonably capable of aggravating or perpetuating that disadvantage?"; "Are group members currently socially vulnerable to stereotyping, social prejudice and/or marginalization?"; and "Does this distinction expose them to the reasonable possibility of future social vulnerability to stereotyping, social prejudice and/or marginalization?". Membership in a "discrete and insular minority", lacking in political power and thus susceptible to having its interests overlooked, is yet another consideration that may be taken into account. 60 Consideration of these factors involves the recognition that differently situated groups are starting on different levels of the s. 15 playing field. In my view, our approach to s. 15 must reflect that reality. Indeed, I reiterate McIntyre J.'s words in Andrews, supra, at p. 169 that "for the accommodation of differences, which is the essence of true equality, it will frequently be necessary to make distinctions". Treating historically vulnerable, disadvantaged or marginalized groups in the same manner as groups which do not generally suffer from such vulnerability may not accommodate, or even contemplate, those differences. In fact, ignoring such differences may compound them, by making access to s. 15 relief most difficult for those groups that are the most disempowered of all in Canadian society. 61 To summarize, the more socially vulnerable the affected group and the more fundamental to our popular conception of "personhood" the characteristic which forms the basis for the distinction, the more likely that this distinction will be discriminatory. Of course, these factors only provide half of the picture. We do not enjoy the full spectrum of colours and contexts until we have also turned our attention to a second set of considerations. The nature of the affected interest 62 In the same way that a very dense projectile will impact upon a surface more sharply than a less dense projectile, an examination of the nature of the interest affected by the impugned distinction is helpful in determining whether that distinction is discriminatory. This examination requires an evaluation of both economic and non-economic elements. 63 As I noted earlier, the Charter is not a document of economic rights and freedoms. Rather, it only protects "economic rights" when such protection is necessarily incidental to protection of the worth and dignity of the human person (i.e. necessary to the protection of a "human right"). Nonetheless, the nature, quantum and context of an economic prejudice or denial of such a benefit are important factors in determining whether the distinction from which the differing economic consequences flow is one which is discriminatory. If all other things are equal, the more severe and localized the economic consequences on the affected group, the more likely that the distinction responsible for these consequences is discriminatory within the meaning of s. 15 of the Charter. 64 Although a search for economic prejudice may be a convenient means to begin a s. 15 inquiry, a conscientious inquiry must not stop here. The discriminatory calibre of a particular distinction cannot be fully appreciated without also evaluating the constitutional and societal significance of the interest(s) adversely affected. Other important considerations involve determining whether the distinction somehow restricts access to a fundamental social institution, or affects a basic aspect of full membership in Canadian society (e.g.. voting, mobility). Finally, does the distinction constitute a complete non- recognition of a particular group? It stands to reason that a group's interests will be more adversely affected in cases involving complete exclusion or non-recognition than in cases where the legislative distinction does recognize or accommodate the group, but does so in a manner that is simply more restrictive than some would like. 65 Referring back to our analogy once again, if the projectile is dense enough and thrown hard enough, then it will leave a mark on even the most resilient of surfaces. Similarly, the more fundamental the interest affected or the more serious the consequences of the distinction, the more likely that the impugned distinction will have a discriminatory impact even with respect to groups that occupy a position of advantage in society. 66 To summarize, tangible economic consequences are but one manifestation of the more intangible and invidious harms flowing from discrimination, which the Charter seeks to root out. In other cases, the prejudice will be to an important individual interest rather than to one that is economic in nature. The nature of the interest affected is therefore highly relevant to whether the distinction that adversely affects that interest is discriminatory in nature. In all but the most extreme cases, this factor cannot be considered in isolation. It only assumes meaningful proportions when assessed in light of the nature of the group affected. Frameworks vs. rigid legal tests 67 It must be emphasized that there are no absolute preconditions to, or preclusions from, a finding of discrimination. Although the presence of one or more of the aforementioned factors in either of these two categories may tend toward the conclusion that the impugned distinction is discriminatory, it does not inevitably lead that way. Conversely, the absence of one or more of these factors does not necessarily preclude there still being a finding of discrimination. Courts must treat these considerations as a matrix rather than as a single equation, and as the microscope rather than as the object being studied. 68 Equality and discrimination are notions that are as varied in form as they are complex in substance. Attempts to evaluate them according to legal formulas which incorporate rigid inclusionary and exclusionary criteria are doomed to become increasingly complex and convoluted over time as "hard" cases become the rule rather than the exception. I prefer to steer clear of those rocky shoals, if at all possible, and to adopt a pragmatic and functional approach to s. 15. I believe that an analysis that examines both sets of factors in the basic framework set out above will enable courts to arrive in a principled manner at an answer that reflects as closely as possible the experience of those in the affected group. If, after examining the nature of both the group and the interest affected, a court concludes that the impact of the impugned distinction is capable of inflicting a non-trivial discriminatory "scar" on the affected group, then it must conclude that this distinction is discriminatory. 69 Once it is found that s. 15(1) of the Charter has been violated by a distinction which is discriminatory, examination must pass on to s. 1. C. Section 1 of the Charter 70 Section 15 of the Charter only guarantees freedom from discrimination subject to reasonable limits, demonstrably justifiable in a free and democratic society. In my view, there is no possible justification for a discriminatory distinction other than that it is relevant to an important objective. As such, a distinction found to violate s. 15(1) may only be saved under s. 1 if it is found to be relevant to a proportionate extent to a pressing and substantial objective. This is accomplished by reference to the framework to s. 1 analysis set down in R. v. Oakes, [1986] 1 S.C.R. 103 and modified by the majority of this Court in Dagenais v. Canadian Broadcasting Corp.,[1994] 3 S.C.R. 835. (i) Pressing and substantial objective 71 In Oakes, it was held that the objective of the legislation, which the measures responsible for a limit on a Charter right or freedom are designed to serve, must be of sufficient importance to warrant overriding a constitutionally protected right or freedom. It therefore must relate to concerns which are pressing and substantial in a free and democratic society. Moreover, in Big M, supra, this Court held that a purpose may never, itself, be unconstitutional. By that same token, where the purpose of impugned legislation is, itself, discriminatory, it cannot be saved under s. 1. I would add, however, that where the court has available to it several possible, and equally likely, interpretations of the purpose of the legislation, then it should prefer one that is consistent with Charter values over one that is not. (ii) Relevant to a proportionate extent 72 Where the purpose of the legislation is found to be pressing and substantial and non-discriminatory, it then becomes necessary to decide whether there is a sufficient degree of proportionality between the impugned distinction (i.e. the means to achieve the purpose) and the rights violation. 73 As I noted earlier, an important concern with including relevance within the ambit of s. 15 analysis is that a court would have little guidance on how relevant/irrelevant a particular distinction need be to the purpose of the legislation in order for that distinction to be discriminatory. Professor Bayefsky, in "A Case Comment on the First Three Equality Rights Cases Under the Canadian Charter of Rights and Freedoms: Andrews, Workers' Compensation Reference, Turpin" (1990), 1 Supreme Court L.R. (2d) 503, at p. 528, favours a sliding scale of scrutiny which recognizes that all trade- offs between furtherance of a pressing and substantial objective and a rights violation occur along a continuum rather than at a single point. To date, this Court has developed no better approach for evaluating whether this equilibrium has been appropriately respected than through the proportionality test in s. 1. Therefore, for a distinction that is discriminatory to be justified under s. 1, it must be shown to be relevant to a proportionate extent to the purpose of the legislation. This determination flows from the traditional three-part proportionality test set out in Oakes and modified in Dagenais. 74 First, the distinction must be rationally connected to the pressing and substantial objective of the legislation. This standard sets down a basic relevancy requirement. Where a distinction is, essentially, irrelevant to the purpose which the legislation seeks to advance, then the distinction cannot be saved under s. 1. Discrimination on the basis of the legal status of the group affected (e.g. citizenship, province of residence, marital status) may raise special problems in this respect, since legal status always comes attached with specific rights and obligations. Because of the various rights and obligations which differing status-based groups may enjoy, part of the rational connection determination in such instances may require some inquiry into whether the distinction drawn in the impugned legislation is relevant to one or more of those rights and/or obligations. If the distinction does not relate rationally to either a right or an obligation which attaches to the affected status-based group, then I do not see how a distinction drawn on the basis of membership in that status-based group would not be irrelevant. 75 The next stage of the proportionality analysis requires that the government demonstrate that the legislation impairs as little as reasonably possible the claimant's s. 15 rights. Depending on the circumstances, the difficulty or impossibility of finding a workable alternative basis of distinction may be a valid consideration under this branch of the proportionality test. So, too, may be the fact that the government has had to make a reasonable trade-off in a field where conflicting social science views, or conflicting rights and interests between groups, are at stake. The mere fact, however, that legislation is "social" in nature is not, by itself, a reason for increased deference. In fact, to defer to the legislative prerogative in circumstances where social science views do not substantially conflict, and where there is a reasonable, alternative means of fulfilling the legislative objective in a way that would materially lessen the magnitude of the rights violation, would frustrate the purpose of the Charter. 76 Finally, there must be a proportionality between the discriminatory effects of the impugned distinction and the salutary effects of the distinction: Dagenais v. Canadian Broadcasting Corp., supra. Factors such as the importance of the state interest, the extent to which it is furthered by the impugned distinction, the constitutional and societal significance of the interests adversely affected, the severity of the rights deprivation suffered by the individual, and the potential for entrenching marginalization or stigmatization of particular groups will all be relevant considerations to this branch of the s. 1 examination. The government must shoulder a heavier justificatory burden when the Charter infringement is severe: Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), [1990] 1 S.C.R. 1123, at p. 1190. 77 It should be noted, finally, that neither s. 1 nor s. 15 calls for a balance sheet approach to discrimination (i.e. summing up all direct and incidental economic benefits to a particular distinction and comparing them against the sum of the economic prejudices, in order to see if there is a net economic prejudice). Such an approach to discrimination loses the forest for the trees. In the context of s. 15, it must be recalled that actual economic prejudice is only one (albeit sometimes significant) ingredient in the more fundamental determination of whether a particular legislative distinction is discriminatory. Similarly, when conducting a s. 1 analysis, it must be recalled that the rights deprivation that triggered s. 15 is not economic. It is something which is fundamental to the person. As such, the mere fact that other economic benefits may flow from the same distinction does not necessarily render that rights-violation any less severe, any less impairing, or any less worthy of condemnation. D. Implications and "Adverse Effects" 78 Adopting an effects-based approach to s. 15 that looks to groups rather than grounds recognizes the importance of adverse effects discrimination in s. 15 without requiring us to resolve some of the intractable issues that have sprung up around that doctrine. A good example is that raised in Bayefsky, supra, at pp. 518-19: Restricting the possible grounds of discrimination puts litigants in the position of first having to prove that their distinguishing feature is caught. This marks the introduction of the same sort of counter-productive, formalistic and artificial debates that have been conducted under anti-discrimination legislation on such issues as the following: does "sex" include differentiation on the basis of "pregnancy" or does "sex" include distinctions made on the basis of sexual orientation? The Court, for example, will have to ask whether legislation involving differential treatment of domestic workers differentiates on the basis of sex because most domestic workers are women. In order to apply section 15 they will be required to find that a differential impact on women should be classed as sex discrimination, rather than facing squarely the issue of differential treatment of domestic workers. And if legislation discriminating against domestic workers can be caught by proving such differential impact upon an enumerated ground, what about similar legislation drawing distinctions that disadvantage farm workers? If one cannot prove that such legislation has a differential impact upon an enumerated or analogous ground, are farm workers excluded from making a section 15 argument, while domestic workers are covered? What logic can there be to this approach to interpreting section 15? [Emphasis added.] See also N. Iyer, "Categorical Denials: Equality Rights and the Shaping of Social Identity" (1993), 19 Queen's L.J. 179. 79 To expand briefly upon the example of domestic workers, under traditional adverse effects doctrine, what percentage of the group would have to have been women in order to succeed in a sex-based discrimination claim? Fifty percent? Ninety percent? As this Court found in Symes v. Canada, [1993] 4 S.C.R. 695, it is difficult to draw a principled distinction along such lines. I believe that it is both easier and more intellectually honest to examine the effect of the distinction on the group affected. In this case, that group would be domestic workers, and the only decision is: does the distinction discriminate against domestic workers? 80 As I noted in Canada (Attorney General) v. Mossop, [1993] 1 S.C.R. 554 at p. 645, categories of discrimination cannot be reduced to watertight compartments, but rather will often overlap in significant measure. When assessing the social context of the impugned distinction, it is therefore of relevance that a significant majority of domestic workers are immigrant women, a subgroup that has historically been both exploited and marginalized in our society. Awareness of, and sensitivity to, the realities of those experiencing the distinction is an important task that judges must undertake when evaluating the impact of the distinction on members of the affected group. Discrimination cannot be fully appreciated or addressed unless courts' analysis focuses directly on the issue of whether these workers are victims of discrimination, rather than becoming distracted by ancillary issues such as "grounds", be they enumerated or analogous. 81 In conclusion, the approach to s. 15 set out in Andrews was an extremely good start. The ensuing jurisprudence has contributed in diverse and important respects to giving content to s. 15 of the Charter. As cases become more and more difficult, however, I believe that it is becoming increasingly evident that we may have been putting the cart before the horse. Although s. 15 is a general guarantee of "equality without discrimination", we have failed to put "discrimination", itself, at the forefront of our analysis. Instead, we have begun to define ourselves into boxes by making "grounds" a precondition to discrimination. As such, we may be denying s. 15 relief to persons who are victims of legislatively sanctioned discrimination, but who are unable to fit themselves into an established or analogous "ground". 82 In this day and age, discrimination is rarely an explicit purpose, and all too frequently an effect. The shortcomings of relying upon "grounds of distinction" are becoming increasingly evident -- we are coming to realize more and more that some "grounds" may give rise to discrimination in some contexts and not in others. In reality, it is no longer the "grounds" that are dispositive of the question of whether discrimination exists, but the social context of the distinction that matters. Given the growing reliance on "context", and the declining role of the "grounds of distinction", I believe that it is a natural next step in s. 15 jurisprudence to admit that, in fact, context is of primary importance and that abstract "grounds of distinction" are simply an indirect method to achieve a goal which could be achieved more simply and truthfully by asking the direct question: "Does this distinction discriminate against this group of people?" See Wilson J. in McKinney, supra, at pp. 392-93. There is no need to use the vehicle of "analogous grounds" to make s. 15 open- ended. An effects-based approach to the notion of "discrimination" will, itself, be open-ended. At the same time, as long as we view "discrimination" through a principled framework of analysis, the open-endedness will be constrained along reasonable and principled lines. 83 This approach bears far greater similarity than difference to jurisprudence inspired by Andrews. It will not, for instance, apply very differently to the vast number of distinctions drawn according to place of residence, or between professionals, employees, or different types of occupations. It is not enough to demonstrate that one has been treated differently by the legislation, and that this differential treatment has been less advantageous than that accorded to other parties. The distinction must, itself, be shown somehow to impugn a basic right to equal human dignity and worth. This determination, as I discussed above, will largely be a function of the nature of both the group and the interest adversely affected by the impugned distinction. As a practical matter, all other things being equal, the greater the degree of unfettered choice involved in membership in the group that is adversely affected, the more difficult it will be to demonstrate that differential treatment undermines basic human rights. As well, in many such situations it will be fair to say that the nature of the interest being adversely affected by the distinction, as well as the constitutional and societal significance of that interest, will generally be comparatively low. 84 It is hoped that this approach offers to individuals and groups that may otherwise be denied the protections of s. 15 the opportunity to claim the fullest protections of our constitution. In my view, s. 15 will never realize its full potential until we adopt an approach that is truly effects-sensitive. The courts of this land must not rest until all individuals are able to enjoy the fundamental right to "equality without discrimination". II. Application to the Facts 85 Our first task is to determine whether there is a legislative distinction between the rights claimants, and Nesbit, and others, and whether this distinction results in the denial of one of the claimants' four basic equality rights (equality before the law, equality under the law, equal protection of the law, and equal benefit of the law) on the basis of their membership in an identifiable group. 86 For the reasons he sets out, I agree with Cory J. that it is clear that homosexual couples are denied the equal benefit of the law on the basis of the legislative distinction in s. 2 of the Old Age Security Act, which defines couples as relationships of "opposite sex". That Egan and Nesbit are able to claim higher benefits as separate individuals does not alter the fact that they have been denied the benefits, both tangible and intangible, of filing for old age benefits as a couple. It would take too narrow a view of the phrase "benefit of the law" to define it strictly in terms of economic interests. Official state recognition of the legitimacy and acceptance in society of a particular type of status or relationship may be of greater value and importance to those affected than any pecuniary gain flowing from that recognition. 87 Furthermore, I share my colleague's reasoning and analysis to the effect that the impugned distinction in s. 2 of the Old Age Security Act is made on the basis of sexual orientation. It would be inconsistent with an effects-oriented approach to the Charter to require as a precondition for protection on this basis that sexual orientation be directly identified in the impugned legislation. Rather, it is sufficient that the couple's sexual orientation cannot in any meaningful way be separated from the "opposite sex" requirement in the Act. This reasoning translates equally well into the framework that I have developed above. In short, the impugned distinction excludes the rights claimants because they are homosexual. To the extent that I frame the problem somewhat more narrowly than Cory J., it is because I prefer to analyze the problem from the point of view of the group actually affected by the distinction, rather than in light of the somewhat illusory neutrality afforded by speaking of the "ground" of "sexual orientation". 88 The last step within s. 15 is to ask ourselves if the distinction is one which is capable of either promoting or perpetuating a view that the appellants and Nesbit are, by virtue of their homosexuality, less capable or less worthy of recognition or value as human beings or as members of Canadian society, equally deserving of concern, respect, and consideration. This evaluation is accomplished by reference to both the nature of the group affected and the nature of the interest affected. These two factors constitute, in short, the all-important socio- economic context of the impugned distinction. 89 In my view, consideration of these two factors leads inevitably to the conclusion that the impugned distinction is discriminatory. My colleague Cory J. cites extensive evidence that same-sex couples are a highly socially vulnerable group, in that they have suffered considerable historical disadvantage, stereotyping, marginalization and stigmatization within Canadian society. The distinction, moreover, is on the basis of an aspect of 'personhood' that is quite possibly biologically based and that is at the very least a fundamental choice. Finally, I cannot help but note that, in addition to being homosexuals, those individuals directly affected by the distinction are all, by definition, also elderly and poor. They are therefore at the margins of an already marginalized group within society. 90 I turn next to the nature of the interest adversely affected. The impugned legislation, the Old Age Security Act, is a cornerstone in Canada's social security net, which is, in turn, a cherished and fundamental institution in our society. It involves the guarantee of a minimum level of income for elderly couples in conjugal relationships. The interest affected is therefore quite a fundamental one, from both a societal and, quite possibly, a constitutional perspective. Although the claimants cannot be said to suffer any economic prejudice from the distinction since they are each entitled as individuals to a certain minimum income level, it cannot be overlooked that the rights claimants have been directly and completely excluded, as a couple, from any entitlement to a basic shared standard of living for elderly persons cohabiting in a relationship analogous to marriage. This interest is an important facet of full and equal membership in Canadian society. Given the marginalized position of homosexuals in society, the metamessage that flows almost inevitably from excluding same-sex couples from such an important social institution is essentially that society considers such relationships to be less worthy of respect, concern and consideration than relationships involving members of the opposite sex. This fundamental interest is therefore severely and palpably affected by the impugned distinction. 91 Blending my analysis of the nature of the interest together with my conclusions regarding the nature of the group affected by the impugned distinction, I am convinced that this distinction is reasonably capable of exacting a discriminatory toll upon the group affected. I would therefore find the impugned distinction to be discriminatory within the meaning of s. 15(1) of the Charter. 92 A violation of s. 15(1) of the Charter may be saved under s. 1 if the impugned distinction is demonstrably justifiable in our free and democratic society. I am of the opinion, however, that the instant violation cannot be salvaged by s. 1, as it is not relevant to a proportionate extent to a pressing and substantial objective. 93 Like Iacobucci J., I would characterize the purpose of the legislation, which s. 2 of the Act is intended to further, in the way it was framed by Linden J.A. (at p. 446): In general terms ... to ensure that when one partner in a couple retires, that couple will continue to receive income equivalent to the amount that would be earned if both members of the couple were retired. I believe that this objective is pressing and substantial and that it is not, itself, animated by a discriminatory purpose. I find, however, that the impugned distinction (i.e. the impugned means to achieve this objective) fails all three branches of the proportionality test. 94 With respect to the first branch of the proportionality test set out in Oakes, I have a great deal of difficulty even concluding that the exclusion of same-sex couples is rationally connected to the objective cited above. The legislation excludes couples who would fill all of the other criteria in the Act except the requirement that they are of the opposite sex. To find that this distinction is rationally connected to the objective of the legislation requires us to conclude that same-sex couples are so different from married couples that it would be unreasonable to make the same benefits available to both. At best, the government has only demonstrated that this is its assumption. I am in entire agreement with Iacobucci J.'s observation that the presumption that same-sex relationships are somehow less interdependent then opposite-sex relationships is, itself, a fruit of stereotype rather than one of demonstrable, empirical reality. See Mossop, supra, at pp. 630-31, per L'Heureux- Dube J. It would be strange, indeed, to permit the government to justify a discriminatory distinction on the basis of presumptions which are, themselves, discriminatory. This would defeat the very purpose which s. 15(1) is intended to further. 95 In this context, I would also reject any argument that homosexual relationships have a distinct biological reality -- namely that homosexuality is non-procreative - - as dangerously reminiscent of the type of biologically based arguments that this Court has now firmly rejected. Compare Bliss v. Attorney General of Canada, [1979] 1 S.C.R. 183 at p. 190, with Brooks v. Canada Safeway Ltd., [1989] 1 S.C.R. 1219, at pp. 1243-44. Moreover, as Iacobucci J. points out, it is evident that the absence or presence of children has nothing whatsoever to do with eligibility for the old age spousal supplement under the Act. I therefore find that the impugned distinction fails the rational connection branch of the proportionality test. Though it is not necessary to address the remaining branches of the Oakes test, I shall do so briefly to demonstrate that I would reach the same result in any event. 96 I agree with Iacobucci J.'s application of the minimal impairment branch of the test. To this, I would add that it cannot be argued that the fact that greater economic benefits may be available to Nesbit, as a single person, under the provincial Guaranteed Available Income for Need Act, R.S.B.C. 1979, c. 158, is evidence that the infringement of s. 15 is minimally impairing. As I have already noted elsewhere, a balance sheet approach to economic benefits and prejudices trivializes and, in many ways, misconstrues the nature of the human rights violation at issue. 97 The Attorney General of Canada argues that the impugned legislation is social, not punitive, and should thereby enjoy a somewhat more deferential standard of review by this Court. As I have already noted, I do not believe that such deference is appropriate when there is a reasonable alternative, readily available, that is not the subject of conflicting social science views, that would materially lessen the effect of the rights violation to the affected group, and that would not result in a concomitant prejudice to another group. To accord deference merely because the issue is a "social" one would be to issue a licence to discriminate in favour of the status quo. In the present case, I believe that a reasonable alternative remedy is, indeed, available. The discriminatory effect would, in fact, be eliminated without prejudice to the rights or interests of any other group, by extending coverage to same-sex couples who otherwise fulfil all of the other non-discriminatory criteria required in the Act. 98 Finally, the third branch of the test requires that there be a proportionality between the discriminatory and salutary effects of the distinction. At this stage, it is appropriate to recall that the Charter breach (i.e. the discriminatory effect) is quite severe. The discrimination against homosexuals arises on the face of the legislation and flows inevitably from the "opposite sex" requirement in s. 2 of the Act. In addition, the impugned distinction is in an Act that plays an important role in a very important Canadian social institution. The interest at issue is a fundamental one -- the right to a basic level of income for the elderly -- and the non- recognition is complete, rather than partial. Although the claimants are not necessarily economically worse off as a result of their exclusion from the Old Age Security Act, the complete exclusion from the program of same-sex couples has a significant discriminatory impact in terms of perpetuating prejudice, stereotyping, and marginalization of same-sex couples, and homosexuals and lesbians individually. 99 It can be argued that the primary salutary effect of the distinction, on the other hand, is the savings it ostensibly entails to the public purse. The government's expert estimates this saving as ranging between $12 million and $37 million. The appellants' cross- examination at trial of that expert suggests that this figure may be considerably less. I would nonetheless make three observations in relation to this argument. First, by the government's own account, these sums account for only between two and four percent of the total cost of the old age supplement program. Second, I have referred to these savings as "ostensible" because if the affected persons had been in heterosexual relationships instead of homosexual relationships, the government would have to have paid out this money anyway. Finally, I note that the majority of this Court recognized in Schachter v. Canada, [1992] 2 S.C.R. 679, at p. 709, that budgetary considerations should not be determinative of a s. 1 analysis, and should more properly be considered when attempting to formulate an appropriate remedy. On this basis, I conclude that the deleterious effects of the impugned distinction outweigh its salutary effects. 100 It goes without saying that I cannot agree with the novel approach to s. 1 taken by Sopinka J. in this case, particularly in light of the following remarks by Dickson C.J. in R. v. Oakes, [1986] 1 S.C.R. 103 at p. 136: A second contextual element of interpretation of s. 1 is provided by the words "free and democratic society". Inclusion of these words as the final standard of justification for limits on rights and freedoms refers the Court to the very purpose for which the Charter was originally entrenched in the Constitution: Canadian society is to be free and democratic. The Court must be guided by the values and principles essential to a free and democratic society which I believe embody, to name but a few,