Supreme Court of Canada

Quebec (Attorney General) v. A, 2013 SCC 5, [2013] 1 S.C.R. 61

Date: 2013-01-25 Docket: 33990

Between:

Attorney General of Quebec — Appellant

and

A — Respondent

And Between:

B — Appellant

and

A — Respondent

And Between:

A — Appellant

and

B and Attorney General of Quebec — Respondents

- and -

Attorney General of New Brunswick, Attorney General of Alberta, Fédération des associations de familles monoparentales et recomposées du Québec and Women's Legal Education and Action Fund — Interveners

Official English Translation: Reasons of LeBel J. and Deschamps J.

Coram: McLachlin C.J. and LeBel, Deschamps, Fish, Abella, Rothstein, Cromwell, Moldaver and Karakatsanis JJ.

Reasons for Judgment: (paras. 1 to 282) LeBel J. (Fish, Rothstein and Moldaver JJ. concurring) Reasons Dissenting in Result: (paras. 283 to 381) Abella J. Reasons Dissenting in Part in Result: (paras. 382 to 409) Deschamps J. (Cromwell and Karakatsanis JJ. concurring) Reasons Concurring in Result: (paras. 410 to 450) McLachlin C.J.

Indexed as: Quebec (Attorney General) v. A

2013 SCC 5

File No.: 33990.

2012: January 18; 2013: January 25.

Present: McLachlin C.J. and LeBel, Deschamps, Fish, Abella, Rothstein, Cromwell, Moldaver and Karakatsanis JJ.

on appeal from the court of appeal for quebec

Constitutional law — Charter of rights — Right to equality — Discrimination based on marital status — De facto spouses — Whether provisions of Civil Code of Québec dealing with family residence, family patrimony, compensatory allowance, partnership of acquests and obligation of spousal support infringe guaranteed right to equality because their application is limited to private legal relationships between married spouses and civil union spouses — If so, whether infringement justified — Civil Code of Québec, S.Q. 1991, c. 64, arts. 401 to 430, 432, 433, 448 to 484, 585 — Canadian Charter of Rights and Freedoms, ss. 1 , 15(1) .

Constitutional law — Charter of rights — Right to equality — Analytical framework applicable to claim under s. 15(1) of Canadian Charter of Rights and Freedoms — Whether prejudice and stereotyping are separate elements into which claim of discrimination must fit — Distinction between two stages of analysis on right to equality, namely stage of review under s. 15 and that of justification under s. 1 — Stage of analysis at which freedom of choice and autonomy of spouses should be considered in relation to partition of property and support.

Family law — De facto spouses — Separation — Support — Spousal support — Family assets — De facto spouses not being covered by protections granted in Civil Code of Québec to married and civil union spouses in relation to support and partition of property — Whether failure to grant same rights to de facto spouses infringes right to equality guaranteed by s. 15(1) of Canadian Charter of Rights and Freedoms — Civil Code of Québec, S.Q. 1991, c. 64, arts. 401 to 430, 432, 433, 448 to 484, 585.

A and B met in A’s home country in 1992. A, who was 17 years old at the time, was living with her parents and attending school. B, who was 32, was the owner of a lucrative business. From 1992 to 1994, they travelled the world together several times a year. B provided A with financial support so that she could continue her schooling. In early 1995, the couple agreed that A would come to live in Quebec, where B lived. They broke up soon after, but saw each other during the holiday season and in early 1996. A then became pregnant with their first child. She gave birth to two other children with B, in 1999 and 2001. During the time they lived together, A attempted to start a career as a model, but she largely did not work outside of the home and often accompanied B on his travels. B provided for all of A’s needs and for those of the children. A wanted to get married, but B told her that he did not believe in the institution of marriage. He said that he could possibly envision getting married someday, but only to make a long‑standing relationship official. The parties separated in 2002 after living together for seven years.

In February 2002, A filed a motion in court seeking custody of the children. The motion was accompanied by a notice to the Attorney General of Quebec stating that A intended to challenge the constitutionality of several provisions of the Civil Code of Québec (“ C.C.Q. ”) in order to obtain the same legal regime for de facto spouses that existed for married spouses. A thus claimed support for herself, a lump sum, partition of the family patrimony and the legal matrimonial regime of partnership of acquests. She also sought to reserve her right to claim a compensatory allowance. A’s claim concerning the use of the family residence was settled in an agreement between A and B. These appeals relate solely to the constitutional aspect of the case. The Quebec Superior Court rejected A’s constitutional arguments and found that the impugned provisions did not violate the right to equality guaranteed by s. 15 of the Charter . A appealed to the Quebec Court of Appeal, which allowed A’s appeal in part and declared the provision that provides for the obligation of spousal support to be of no force or effect. However, the Court of Appeal upheld the Superior Court’s decision as regards the constitutionality of the provisions concerning the family residence, the family patrimony, the compensatory allowance and the partnership of acquests. The majority of the court suspended the declaration of constitutional invalidity of art. 585 C.C.Q. for 12 months. B and the Attorney General of Quebec are appealing the Court of Appeal’s decision to strike down art. 585. A appeals the conclusion that the provisions concerning the partition of property are constitutionally valid.

Held (Deschamps, Cromwell and Karakatsanis JJ. dissenting in part in the result and Abella J. dissenting in the result) : The appeals of the Attorney General of Quebec and B should be allowed, and the appeal of A should be dismissed. Articles 401 to 430 , 432 , 433 , 448 to 484 and 585 of the Civil Code of Québec are constitutional.

The constitutional questions should be answered as follows:

1. Do arts. 401 to 430, 432, 433, 448 to 484 and 585 of the Civil Code of Québec , S.Q. 1991, c. 64, infringe s. 15(1) of the Canadian Charter of Rights and Freedoms ?

Answers: McLachlin C.J. and Deschamps, Abella, Cromwell and Karakatsanis JJ. would answer yes. LeBel, Fish, Rothstein and Moldaver JJ. would answer no.

2. If so, is the infringement a reasonable limit prescribed by law that can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

Answers: LeBel, Fish, Rothstein and Moldaver JJ. would answer that it is not necessary to answer this question. McLachlin C.J. would answer yes. Deschamps, Cromwell and Karakatsanis JJ. would answer that only art. 585 is not justified under s. 1 . Abella J. would answer no.

(1) Section 15(1) of the Charter

Per LeBel, Fish, Rothstein and Moldaver JJ. (minority on s. 15(1) ): The Civil Code of Québec establishes a mandatory primary regime in a chapter that defines the fundamental effects of marriage. This regime creates mutual rights, duties and obligations and radically alters each spouse’s patrimonial rights. More specifically, the primary regime results in the formation of a partial economic union between the spouses. Aside from the primary regime, where there is no marriage contract providing for separation as to property or for changes to the legal regime, the legal matrimonial regime of partnership of acquests applies to the spouses as a result of their marriage. Like the primary regime, the regime of partnership of acquests significantly changes the rights of both spouses in relation to their patrimony. The Quebec legislature has imposed these regimes only on those who, by agreement with another person, have demonstrated that they wish to adhere to them. Their consent must be explicit, and must take the form of marriage or a civil union. The Civil Code of Québec does not lay down the terms of the union of de facto spouses. Since the de facto union is not subject to the mandatory legislative framework that applies to marriage and the civil union, de facto spouses are free to shape their relationships as they wish, having proper regard for public order. They can enter into agreements to organize their patrimonial relationships while they live together and to provide for the consequences of a possible breakdown.

By arguing that arts. 401 to 430, 432, 433, 448 to 484 and 585 of the Civil Code of Québec are contrary to s. 15(1) of the Charter and not justified under s. 1 , A is claiming the benefit of certain aspects of the primary regime that applies in cases of separation from bed and board, divorce, or dissolution of a civil union. She is also seeking the automatic and mandatory application of the legal matrimonial regime of partnership of acquests. In R. v. Kapp , 2008 SCC 41 , [2008] 2 S.C.R. 483 , and Withler v. Canada (Attorney General) , 2011 SCC 12 , [2011] 1 S.C.R. 396 , the Court reworked and provided important clarifications to the analytical framework for applying the equality guarantee provided for in s. 15(1) of the Charter . As can be seen from this framework, a discriminatory distinction is as a general rule an adverse distinction that perpetuates prejudice or that stereotypes. The existence of a pre‑existing or historical disadvantage will make it easier to prove prejudice or a stereotype. However, the existence or perpetuation of a disadvantage cannot in itself make a distinction discriminatory. Substantive equality is not denied solely because a disadvantage is imposed. Rather, it is denied by the imposition of a disadvantage that is unfair or objectionable, which is most often the case if the disadvantage perpetuates prejudice or stereotypes. Thus, according to the established analytical framework, a court analyzing the validity of an allegation that s. 15(1) has been infringed must address the following questions: (1) Does the law create a distinction based on an enumerated or analogous ground? (2) Does the distinction create a disadvantage by perpetuating prejudice or stereotyping? The claimant can show that the impugned law creates a distinction expressly or that it creates one indirectly.

The majority of the Court would have reached the same conclusion in Nova Scotia (Attorney General) v. Walsh , 2002 SCC 83 , [2002] 4 S.C.R. 325 , if its analysis had been based on the reworked analytical framework from Kapp and Withler . Although the statute at issue in Walsh imposed differential treatment based on marital status by limiting the presumption of equal division of matrimonial property to married couples and excluding persons in common law relationships, that distinction did not create a disadvantage by perpetuating prejudice or stereotyping. The majority’s analysis was thus based on the wish to promote substantive equality. Walsh was based on a principle of freedom to choose between different marital statuses that had different consequences for spouses, and that principle did not in that context infringe the constitutional equality guarantee. The principle in question continues to be valid in the circumstances of the case at bar despite the subsequent developments in the case law. Although Walsh concerned not the obligation of support, but the equal division of family assets, the majority’s comments on the sources of the distinctions between the various forms of relationships and the consequences of those distinctions remain relevant.

To dispose of these appeals, it would be inappropriate to distinguish the partition of property from the obligation of support. Such a distinction disregards the character of an “economic partnership” that the Quebec legislature has established for marriage and the civil union. It also disregards the fact that this partnership is structured around a mandatory primary regime that has both patrimonial and extrapatrimonial aspects and that the primary regime establishes the obligation of support as an effect of marriage and of the civil union. In this sense, the obligation of support is tied to the other effects of marriage and of the civil union, such as the obligation to contribute to household expenses, rights and obligations with respect to the family residence, and the creation of a family patrimony. It forms an integral and indissociable part of the set of measures that constitute Quebec’s primary regime. What must therefore be determined in these appeals is not whether the exclusion of de facto spouses from the obligation of support is discriminatory, but whether their exclusion from the entire statutory framework imposed on married and civil union spouses is discriminatory under s. 15(1) of the Charter .

To prove that she has been discriminated against, A must show on a balance of probabilities that the provisions of the Civil Code of Québec at issue create an adverse distinction based on an enumerated or analogous ground and that the disadvantage is discriminatory because it perpetuates prejudice or stereotypes. The provisions relating to the family patrimony, the family residence, the compensatory allowance, the partnership of acquests and the obligation of support apply only to persons who are married or in a civil union, and do not apply to de facto spouses. These provisions therefore have the effect of creating a distinction based on the analogous ground of marital status. That distinction may result in disadvantages for those who are excluded from the statutory framework applicable to a marriage or a civil union. Generally speaking, when de facto spouses separate, one of them will likely end up in a more precarious patrimonial situation than if the couple had been married or in a civil union. As a result, unless these de facto spouses have exactly the same earning capacity and exactly the same patrimony, one of them will be in a worse position after the relationship ends than would a married or civil union spouse in a similar patrimonial situation.

However, the distinction is not discriminatory, because it does not create a disadvantage by expressing or perpetuating prejudice or by stereotyping. Although there was a period of Quebec history during which de facto spouses were subjected to both legislative hostility and social ostracism, nothing in the evidence suggests that de facto spouses are now subject to public opprobrium. The expert reports filed by the parties tend to show the contrary. According to them, the de facto union has become a respected type of conjugality and is not judged unfavourably by Quebec society as a whole. Likewise, the legislature’s traditional hostility generally seems to have changed into acceptance of the de facto union. In this regard, Quebec social legislation no longer draws distinctions between the various types of conjugality either in granting benefits to or imposing obligations on spouses where their relations with government institutions are concerned. The distinction continues to exist in the context of relations between the spouses themselves, within their conjugal relationship, where there is still a will to preserve the possibility of choosing between various types of conjugality.

Nor is the exclusion of de facto spouses from the application of the impugned provisions discriminatory on the basis of an expression of prejudice. The legislature has not established a hierarchy between the various forms of conjugality, nor has it expressed a preference for marriage and the civil union at the expense of the de facto union. It has merely defined the legal content of the different forms of conjugal relationships. It has made consent the key to changing the spouses’ mutual patrimonial relationship. In this way, it has preserved the freedom of those who wish to organize their patrimonial relationships outside the mandatory statutory framework. Express, and not deemed, consent is the source of the obligation of support and of that of partition of spouses’ patrimonial interests. This consent is given in Quebec law by contracting marriage or a civil union, or entering into a cohabitation agreement. Participation in the protective regimes provided for by law depends necessarily on mutual consent. In this regard, the conclusion of a cohabitation agreement enables de facto spouses to create for themselves the legal relationship they consider necessary without having to modify the form of conjugality they have chosen for their life together. In this context in which the existence of a set of rights and obligations depends on mutual consent in one of a variety of forms, it is hard to speak of discrimination against de facto spouses. The resulting choice has become a key factor in the determination of the scope of the right at issue, and not only in the justification of a limit on that right. It is not imperative that there be an identical framework for each form of union in order to remain true to the purpose of s. 15(1). In the instant case, the fact that there are different frameworks for private relationships between spouses does not indicate that prejudice is being expressed or perpetuated, but, rather, connotes respect for the various conceptions of conjugality. Thus, no hierarchy of worth is established between the different types of couples.

The articles of the Civil Code of Québec whose constitutional validity is being challenged by A therefore do not express or perpetuate prejudice against de facto spouses. On the contrary, it appears that, by respecting personal autonomy and the freedom of de facto spouses to organize their relationships on the basis of their needs, those provisions are consistent with two of the values underlying s. 15(1) of the Charter . They were enacted as part of a long and complex legislative process during which the Quebec National Assembly was concerned about keeping step with changes in society and about adapting family law to new types of conjugal relationships in a manner compatible with the freedom of spouses.

Furthermore, there is no evidence in the Court’s record that would justify finding that the exclusion of de facto spouses from the primary regime and the regime of partnership of acquests is based on a stereotypical characterization of the actual circumstances of such spouses. More specifically, none of A’s evidence tends to show that the policy of freedom of choice, consensualism and autonomy of the will does not correspond to the reality of the persons in question. Nor can judicial notice be taken of the fact that the choice of type of conjugality is not a deliberate and genuine choice that should have patrimonial consequences but necessarily results from the spouses’ ignorance of the consequences of their status. Such a fact is clearly controversial and not beyond reasonable dispute. It is not unreasonable to believe that, in theory, individuals sometimes make uninformed choices and that some individuals may be unaware of the consequences of their choice of conjugal lifestyle. Nevertheless, to take judicial notice of the fact that the voluntary choice not to marry does not reflect an autonomous decision to avoid the legal regimes would be to exceed the limits of legitimate judicial notice, especially in relation to an issue at the centre of the controversy. In this case, A has not established that it is stereotypical to believe that couples in a de facto union have chosen not to be bound by the regimes applicable to marriage and civil unions. The Quebec scheme, the effect of which is to respect each person’s freedom of choice to establish his or her own form of conjugality, and thus to participate or not to participate in the legislative regime of marriage or civil union with its distinct legal consequences, is not based on a stereotype. In this sense, recognition of the principle of autonomy of the will, which is one of the values underlying the equality guarantee in s. 15 of the Charter , means that the courts must respect choices made by individuals in the exercise of that autonomy. In this context, it will be up to the legislature to intervene if it believes that the consequences of such autonomous choices give rise to social problems that need to be remedied.

In conclusion, although arts. 401 to 430, 432, 433, 448 to 484 and 585 of the Civil Code of Québec draw a distinction based on marital status between de facto spouses and married or civil union spouses, they do not create a disadvantage by expressing or perpetuating prejudice or by stereotyping. These provisions accordingly do not violate the right to equality guaranteed by s. 15 of the Charter .

Per Abella J. (majority on s. 15(1) ): The total exclusion of de facto spouses — the term used in Quebec for those who are neither married nor in a civil union — from the legal protections for both support and property given to spouses in formal unions is a violation of s. 15(1) of the Canadian Charter of Rights and Freedoms . When spouses who are married or in civil unions separate or divorce in Quebec, they are guaranteed certain legal protections. They have the right to claim support from each other and an equal division of the family property. The spousal support and family property provisions in Quebec are aimed at recognizing and compensating spouses for the roles assumed within the relationship and any resulting interdependence and vulnerability on its dissolution. Many de facto spouses share the characteristics that led to the protections for spouses in formal relationships. They form long‑standing unions; they divide household responsibilities and develop a high degree of interdependence; and, critically, the economically dependent, and therefore vulnerable, spouse is faced with the same disadvantages when the relationship is dissolved. Yet de facto dependent spouses in Quebec have no right to claim support, no right to divide the family patrimony, and are not governed by any matrimonial regime.

As the history of modern family law demonstrates, fairness requires that we look at the content of the relationship’s social package, not at how it is wrapped. In Quebec and throughout the rest of Canada, the right to support does not rest on the legal status of either husband or wife, but on the reality of the dependence or vulnerability that the spousal relationship creates. The law dealing with division of family property also rests on a protective basis rather than a contractual one. The provisions in Quebec on compensatory allowance and the family patrimony regime are part of public order, applying mandatorily to all married spouses and those in civil unions. The mandatory nature of both the compensatory allowance and family patrimony regimes highlights the preeminent significance Quebec has given to concerns for the protection of vulnerable spouses over other values such as contractual freedom or choice.

Historically, unmarried spouses in Canada were stigmatized; but as social attitudes changed, so did the approaches of legislatures and courts, which came to accept conjugal relationships outside a formal marital framework. This change reflected an enhanced understanding of what constitutes a “family”. As attitudes shifted and the functional similarity between many unmarried relationships and marriages was accepted, this Court expanded protection for unmarried spouses. In Miron v. Trudel , [1995] 2 S.C.R. 418 , for example, the Court found that “marital status” was an analogous ground under s. 15(1) of the Charter because of the historic disadvantage of unmarried spouses. Notably too, the Court observed that while in theory an individual is free to choose whether to marry, there are, in reality, a number of factors that may place the decision beyond his or her effective control. This was a recognition of the complex and mutual nature of the decision to marry and the myriad factors at play in that decision. It was also an acknowledgment that the decision to live together as unmarried spouses may, for some, not in fact be a choice at all.

The purpose of the s. 15 equality provision is to eliminate the exclusionary barriers faced by individuals in the enumerated or analogous groups in gaining meaningful access to what is generally available. In Kapp , this Court reaffirmed its commitment to the test that was set out in Andrews v. Law Society of British Columbia , [1989] 1 S.C.R. 143 , whereby s. 15 was seen as an anti‑discrimination provision. The claimant’s burden under the Andrews test is to show that the government has made a distinction based on an enumerated or analogous ground and that the distinction’s impact on the individual or group perpetuates disadvantage. If this has been demonstrated, the burden shifts to the government to justify the reasonableness of the distinction under s. 1 . Kapp , and later Withler restated these principles as follows: (1) Does the law create a distinction based on an enumerated or analogous ground? (2) Does the distinction create a disadvantage by perpetuating prejudice or stereotyping?

In referring to prejudice and stereotyping in the second step of the Kapp reformulation of the Andrews test, the Court was not purporting to create a new s. 15 test. Prejudice and stereotyping are not discrete elements of the test which a claimant is obliged to demonstrate. Prejudice is the holding of pejorative attitudes based on strongly held views about the appropriate capacities or limits of individuals or the groups of which they are a member. Stereotyping, like prejudice, is a disadvantaging attitude, but one that attributes characteristics to members of a group regardless of their actual capacities. Attitudes of prejudice and stereotyping can undoubtedly lead to discriminatory conduct, and discriminatory conduct in turn can reinforce these negative attitudes. But Kapp and Withler should not be seen as establishing an additional requirement on s. 15 claimants to prove that a distinction will perpetuate prejudicial or stereotypical attitudes towards them. Such an approach improperly focuses attention on whether a discriminatory attitude exists, not a discriminatory impact, contrary to Andrews , Kapp and Withler . I t is the discriminatory conduct that s. 15 seeks to prevent, not the underlying attitude or motive. Requiring claimants, therefore, to prove that a distinction perpetuates negative attitudes about them imposes a largely irrelevant, not to mention ineffable burden.

The root of s. 15 is our awareness that certain groups have been historically discriminated against, and that the perpetuation of such discrimination should be curtailed. The key is whether a distinction has the effect of perpetuating arbitrary disadvantage on the claimant because of his or her membership in an enumerated or analogous group. If the state conduct widens the gap between the historically disadvantaged group and the rest of society rather than narrowing it, then it is discriminatory.

Assessment of legislative purpose is an important part of a Charter analysis, but it is conducted under s. 1 once the burden has shifted to the state to justify the reasonableness of the infringement. To focus on the legislative purpose — freedom of choice — at the s. 15(1) stage is not only contrary to the approach in Andrews , it is also completely inconsistent with Miron and undermines the recognition of marital status as an analogous ground . Having accepted marital status as an analogous ground, it is contradictory to find not only that de facto spouses have a choice about their marital status, but that it is that very choice that excludes them from the protection of s. 15(1) to which Miron said they were entitled. Moreover, this Court has repeatedly rejected arguments that choice protects a distinction from a finding of discrimination.

Because the equality analysis under s. 15(1) of the Charter has evolved substantially in the decade since Walsh was decided, Walsh need not be followed. In particular, the majority in Walsh relied on the dignity test and on comparator groups, neither of which is any longer required as part of the s. 15(1) analysis.

The exclusion of de facto spouses from the economic protections available to formal spousal relationships is a distinction based on marital status, an analogous ground. That it imposes a disadvantage is clear: the law excludes vulnerable and economically dependent de facto spouses from protections considered so fundamental to the welfare of vulnerable married or civil union spouses that one of those protections is presumptive, and the rest are of public order, explicitly overriding freedom of contract or choice for those spouses. The disadvantage this exclusion perpetuates is an historic one: it continues to deny de facto spouses access to economic remedies of which they have always been deprived, remedies Quebec considered indispensable for the protection of married and civil union spouses. There is little doubt that some de facto couples are in relationships that are functionally similar to formally recognized spousal relationships. Since many spouses in de facto couples exhibit the same functional characteristics as spouses in formal unions, with the same potential for one partner to be left economically vulnerable or disadvantaged when the relationship ends, their exclusion from similar protections perpetuates historic disadvantage against them based on their marital status. There is no need to look for an attitude of prejudice motivating or created by the exclusion of de facto couples from the presumptive statutory protections. There is no doubt that attitudes have changed towards de facto unions in Quebec, but what is relevant is not the attitudinal progress towards them, but the continuation of their discriminatory treatment.

Per Deschamps, Cromwell and Karakatsanis JJ. (concurring with Abella J. on s. 15(1)): There is agreement with Abella J.’s analysis of s. 15 of the Charter and with her conclusion that the right protected by that section has been infringed. The Quebec legislature has infringed the guaranteed right to equality by excluding de facto spouses from all the measures adopted to protect persons who are married or in civil unions should their family relationships break down. The Court has recognized the fact of being unmarried as an analogous ground because, historically, unmarried persons were considered to have adopted a lifestyle less worthy of respect than that of married persons. For this reason, they were excluded from the social protections. Even though society’s perception of de facto spouses has changed in recent decades and there is no indication that the Quebec legislature intended to stigmatize them, the denial of the benefits in question perpetuates the disadvantage such people have historically experienced. The Attorney General of Quebec therefore had to justify this distinction.

Per McLachlin C.J. (concurring with Abella J. on s. 15(1)): The s. 15 analysis set out in Abella J.’s reasons is agreed with, as is her conclusion that there is a breach. While the promotion or the perpetuation of prejudice, on the one hand, and false stereotyping, on the other, are useful guides, what constitutes discrimination requires a contextual analysis, taking into account matters such as pre‑existing disadvantage of the claimant group, the degree of correspondence between the differential treatment and the claimant group’s reality, the ameliorative impact or purpose of the law, and the nature of the interests affected. The issue of whether the law is discriminatory must be considered 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 claimant.

It is important to maintain the analytical distinction between s. 15 and s. 1 . Walsh does not bind the Court in the present case. Public policy considerations such as freedom of choice and individual autonomy, which were held in Walsh to negate a breach of s. 15, are better considered at the s. 1 stage of the analysis.

Here, the Quebec approach of applying mandatory protections only to married and civil union spouses limits the s. 15 equality right of de facto spouses. A reasonable person in A’s position would conclude that the law in fact shows less concern for people in A’s position than for married and civil union spouses on break‑up of a relationship. As it applies to people in A’s situation, the law perpetuates the effects of historical disadvantage rooted in prejudice and rests on a false stereotype of choice rather than on the reality of the claimant’s situation. While the legislative animus against de facto spouses in Quebec has disappeared, the present law continues to exclude de facto spouses from the protective schemes of Quebec family law. Moreover, the law assumes that de facto partners choose to forgo the protections it offers to married and civil union partners. This assumption fails to accord to the reality of the situation of de facto spouses such as A.

(2) Section 1 of the Charter

Per LeBel, Fish, Rothstein and Moldaver JJ.: Since the exclusion of de facto spouses from the scope of the provisions of the Civil Code of Québec at issue is not discriminatory within the meaning of s. 15(1) of the Charter and does not violate the constitutional right to equality, it is not necessary to proceed to the s. 1 stage of the Charter analysis.

Per McLachlin C.J.: The limit on the equality right of de facto spouses is justified under s. 1 of the Charter . The objective of the Quebec legislature, which is to promote choice and autonomy for all Quebec spouses with respect to property division and support, was pursued in response to rapidly changing attitudes in Quebec with respect to marriage and is sufficiently important to justify an infringement to the right to equality. The distinction made by the law is rationally connected to the state objective: the Quebec approach only imposes state‑mandated obligations on spouses who have made a conscious and active choice to accept those obligations. The law falls within a range of reasonable alternatives for maximizing choice and autonomy in the matter of family assets and support. While schemes adopted in other Canadian provinces impair the equality right of de facto spouses to a lesser degree, such approaches would be less effective in promoting Quebec’s goals of maximizing choice and autonomy for couples in Quebec. The question at the minimum impairment stage is whether the legislative goal could be achieved in a way that impacts the right less, not whether the goal should be altered. Finally, the effects of the Quebec scheme on the equality rights of de facto spouses are proportionate to the scheme’s overall benefits for the group. The scheme enhances the freedom of choice and autonomy of many spouses as well as their ability to give personal meaning to their relationship. Having regard to the need to allow legislatures a margin of appreciation on difficult social issues and the need to be sensitive to the constitutional responsibility of each province to legislate for its population, the unfortunate dilemma faced by women such as A is not disproportionate to the benefits of the scheme to an extent that warrants a finding of unconstitutionality.

Per Deschamps, Cromwell and Karakatsanis JJ.: Although support and the measures relating to patrimonial property have some of the same functions and objectives, they cannot and must not be confused with one another. The needs they address and how the legislature has dealt with them in the past warrant their being considered separately. The measures that protect the patrimony of spouses are not, like support, focused on the basic needs of the vulnerable spouse. Their purpose is to ensure autonomy and fairness for couples who have been able to, or wanted to, accumulate property. The process that leads to the acquisition of a right of ownership is different from the one that causes a spouse to become economically dependent. Whereas a plan to live together takes shape gradually and can result in the creation of a relationship of interdependence over which one of the parties has little or no control, property can be acquired only as a result of a conscious act.

This analysis leads to the conclusion that only the exclusion of de facto spouses from support is not justified under s. 1 of the Charter . The objective of promoting the autonomy of the parties is pressing and substantial. There is also a rational connection. However, the minimal impairment test is not met. The affected interest is vital to persons who have been in a relationship of interdependence. The rationale for awarding support on a non‑compensatory basis applies equally to persons who are married or in a civil union and to de facto spouses. If the legal justification for support is based on, among other things, the satisfaction of needs resulting from the breakdown of a relationship of interdependence created while the spouses lived together, it is difficult to see why a de facto spouse who may not have been free to choose to have the relationship with his or her spouse made official through marriage or a civil union, but who otherwise lives with the latter in a “family unit”, would not be entitled to support. For someone in such a position, the possibility the parties have, according to the Attorney General, of choosing to marry or to enter into a civil union does not really exist. The concept of “mutual obligation” as the non‑compensatory basis for the obligation of support must guide legislators in seeking ways to promote the autonomy of the parties while interfering as little as reasonably possible with the right to support itself. A total exclusion from the right to support benefits only de facto spouses who want to avoid the obligation of support, and it impairs the interests of dependent and vulnerable former spouses to a disproportionate extent.

Per Abella J.: The breach of s. 15(1) is not saved under s. 1 , failing the minimal impairment and proportionality steps of the Oakes test. Th e exclusion of de facto spouses from spousal support and property regimes in Quebec was a carefully considered policy choice. It was discussed and reaffirmed during successive family law reforms from 1980 onwards. But the degree of legislative time, consultation and effort cannot act as a justificatory shield to guard against constitutional scrutiny. What is of utmost relevance is the resulting legislative choice. Neither the deliberative policy route nor the popularity of its outcome is a sufficient answer to the requirement of constitutional compliance.

An outright exclusion of de facto spouses cannot be said to be minimally impairing of their equality rights. This Court has generally been reluctant to defer to the legislature in the context of total exclusions from a legislative scheme. The antipathy towards complete exclusions is not surprising, since the government is required under s. 1 to explain why a significantly less intrusive and equally effective measure was not chosen . This will be a difficult burden to meet when, as in this case, a group has been entirely left out of access to a remedial scheme. The current opt- in protections may well be adequate for some de facto spouses who enter their unions with sufficient financial security, legal information, and the intent to avoid the consequences of a more formal union. But their ability to exercise freedom of choice can be equally protected under a regime with an opt- out mechanism. The needs of the economically vulnerable, however, require presumptive protection no less in de facto unions than in more formal ones. The evidence discloses that many de facto spouses simply do not turn their minds to the eventuality of separation. This lack of awareness speaks to the relative merit of a system of presumptive protection, under which they would be protected whether aware of their legal rights or not, while leaving de facto spouses who wish to do so the freedom to choose not to be protected. A further weakness of the current opt‑in system is its failure to recognize that the choice to formally marry is a mutual and complex decision, as Miron pointed out . Where one member of a couple refuses to marry or enter into a civil union, he or she thereby deprives the other of the benefit of needed economic support when the relationship ends.

Every other province has extended spousal support to unmarried spouses. They have set minimum periods of cohabitation before couples are subject to their regimes, and have preserved freedom of choice by allowing couples to opt out. Some have also extended statutory division of property to unmarried spouses. These presumptively protective schemes with a right on the part of de facto spouses to opt out are examples of alternatives that would provide economically vulnerable spouses with the protection they need, without in any way interfering with the legislative objective of giving freedom of choice to those de facto spouses who want to exercise it. At the end of the day, the methodology for remedying the s. 15 breach lies with the Quebec legislature, and Quebec is in no way obliged to mimic any other province’s treatment of de facto spouses. But t he fact of these other regimes can be helpful in determining that there is a less impairing way to fulfill the objective of preserving freedom of choice without infringing the equality rights of de facto spouses.

The choices for de facto spouses in Quebec are to enter into a contract to enshrine certain protections, to marry and receive all the protections provided by law, or to remain unbound by any mutual rights or obligations. It is entirely possible for Quebec to design a regime that retains all of these choices without violating s. 15. Spouses who are aware of their legal rights, and choose not to marry so they can avoid Quebec’s support and property regimes, would be free to choose to remove themselves from a presumptively protective regime. Changing the default situation of the couple, however, so that spousal support and division of property protection of some kind applies to them, would protect those spouses for whom the choices are illusory and who are left economically vulnerable at the end of the relationship.

The deleterious effect of excluding all de facto spouses, who represent over a third of Quebec couples, from the protection of the family support and division of property regimes is profound. Being excluded requires potentially vulnerable de facto spouses, unlike potentially vulnerable spouses in formal unions, to expend time, effort and money to try to obtain some financial assistance. If the vulnerable spouse fails to take these steps, either through a lack of knowledge or resources, or because of the limits on his or her options imposed by an uncooperative partner, he or she will remain unprotected. The outcome for such a spouse in the event of a separation can be, as it is for economically dependent spouses in formal unions, catastrophic. The difference is that economically dependent spouses in formal unions have automatic access to the possibility of financial remedies. De facto spouses have no such access. The salutary impact of the exclusion, on the other hand, is the preservation of de facto spouses’ freedom to choose not to be in a formal union. Those for whom a de facto union is truly a chosen means to preserve economic independence would still be able to achieve this result by opting out. Since the salutary effect can be achieved without in any way compromising a de facto spouse’s freedom of choice, it cannot be said to outweigh the serious harm for economically vulnerable de facto spouses that results from their exclusion from the family support and property regimes.

Cases Cited

By LeBel J.

Applied: R. v. Kapp , 2008 SCC 41 , [2008] 2 S.C.R. 483 ; Withler v. Canada (Attorney General) , 2011 SCC 12 , [2011] 1 S.C.R. 396 ; Nova Scotia (Attorney General) v. Walsh , 2002 SCC 83 , [2002] 4 S.C.R. 325 ; distinguished : M. v. H. , [1999] 2 S.C.R. 3 ; discussed : Law v. Canada (Minister of Employment and Immigration) , [1999] 1 S.C.R. 497 ; Andrews v. Law Society of British Columbia , [1989] 1 S.C.R. 143 ; referred to: Miron v. Trudel , [1995] 2 S.C.R. 418 ; M.T. v. J.‑Y.T. , 2008 SCC 50 , [2008] 2 S.C.R. 781 ; Droit de la famille — 977 , [1991] R.J.Q. 904 ; Bracklow v. Bracklow , [1999] 1 S.C.R. 420 ; G.B. v. C.C. , [2001] R.J.Q. 1435 ; Couture v. Gagnon , [2001] R.J.Q. 2047 , leave to appeal refused, [2002] 3 S.C.R. vii; Ponton v. Dubé , 2005 QCCA 413 ; Bourbonnais v. Pratt , 2006 QCCS 5611 , [2007] R.D.F. 124 ; M.B. v. L.L. , [2003] R.D.F. 539 ; Peter v. Beblow , [1993] 1 S.C.R. 980 ; Cie Immobilière Viger Ltée v. Lauréat Giguère Inc. , [1977] 2 S.C.R. 67 ; Benzina v. Le , 2008 QCCA 803 ; Barrette v. Falardeau , 2010 QCCA 989 ; C.L. v. J.Le. , 2010 QCCA 2370 ; Droit de la famille — 121120 , 2012 QCCA 909 ; Blencoe v. British Columbia (Human Rights Commission) , 2000 SCC 44 , [2000] 2 S.C.R. 307 ; Gosselin v. Quebec (Attorney General) , 2002 SCC 84 , [2002] 4 S.C.R. 429 ; R. v. Big M Drug Mart Ltd. , [1985] 1 S.C.R. 295 ; R. v. Morgentaler , [1988] 1 S.C.R. 30 ; Rodriguez v. British Columbia (Attorney General) , [1993] 3 S.C.R. 519 ; Eldridge v. British Columbia (Attorney General) , [1997] 3 S.C.R. 624 ; Hodge v. Canada (Minister of Human Resources Development) , 2004 SCC 65 , [2004] 3 S.C.R. 357 ; Egan v. Canada , [1995] 2 S.C.R. 513 ; Corbiere v. Canada (Minister of Indian and Northern Affairs) , [1999] 2 S.C.R. 203 ; Thibaudeau v. Canada , [1995] 2 S.C.R. 627 ; R. v. Oakes , [1986] 1 S.C.R. 103 ; Alberta v. Hutterian Brethren of Wilson Colony , 2009 SCC 37 , [2009] 2 S.C.R. 567 ; Ermineskin Indian Band and Nation v. Canada , 2009 SCC 9 , [2009] 1 S.C.R. 222 ; A.C. v. Manitoba (Director of Child and Family Services) , 2009 SCC 30 , [2009] 2 S.C.R. 181 ; Alberta (Aboriginal Affairs and Northern Development) v. Cunningham , 2011 SCC 37 , [2011] 2 S.C.R. 670 ; R. v. Turpin , [1989] 1 S.C.R. 1296 ; Lavoie v. Canada , 2002 SCC 23 , [2002] 1 S.C.R. 769 ; Trociuk v. British Columbia (Attorney General) , 2003 SCC 34 , [2003] 1 S.C.R. 835 ; Eaton v. Brant County Board of Education , [1997] 1 S.C.R. 241 ; R. v. Williams , [1998] 1 S.C.R. 1128 ; R. v. Find , 2001 SCC 32 , [2001] 1 S.C.R. 863 ; R. v. Spence , 2005 SCC 71 , [2005] 3 S.C.R. 458 .

By Abella J.

Applied: R. v. Kapp , 2008 SCC 41 , [2008] 2 S.C.R. 483 ; Andrews v. Law Society of British Columbia , [1989] 1 S.C.R. 143 ; Withler v. Canada (Attorney General) , 2011 SCC 12 , [2011] 1 S.C.R. 396 ; not followed: Nova Scotia (Attorney General) v. Walsh , 2002 SCC 83 , [2002] 4 S.C.R. 325 ; discussed : Miron v. Trudel , [1995] 2 S.C.R. 418 ; referred to: Québec (Procureure générale) v. B.T. , 2005 QCCA 748 , [2005] R.D.F. 709 ; Moge v. Moge , [1992] 3 S.C.R. 813 ; Bracklow v. Bracklow , [1999] 1 S.C.R. 420 ; M. (M.E.) v. L. (P.) , [1992] 1 S.C.R. 183 ; Droit de la famille — 977 , [1991] R.J.Q. 904 ; Rathwell v. Rathwell , [1978] 2 S.C.R. 436 ; Murdoch v. Murdoch , [1975] 1 S.C.R. 423 ; Pettkus v. Becker , [1980] 2 S.C.R. 834 ; M. v. H. , [1999] 2 S.C.R. 3 ; Ontario Human Rights Commission v. Simpsons‑Sears Ltd. , [1985] 2 S.C.R. 536 ; Canadian National Railway Co. v. Canada (Canadian Human Rights Commission) , [1987] 1 S.C.R. 1114 ; Law v. Canada (Minister of Employment and Immigration) , [1999] 1 S.C.R. 497 ; Griggs v. Duke Power Co. , 401 U.S. 424 (1971); Corbiere v. Canada (Minister of Indian and Northern Affairs) , [1999] 2 S.C.R. 203 ; Brooks v. Canada Safeway Ltd. , [1989] 1 S.C.R. 1219 ; Lavoie v. Canada , 2002 SCC 23 , [2002] 1 S.C.R. 769 ; Janzen v. Platy Enterprises Ltd. , [1989] 1 S.C.R. 1252 ; Nova Scotia (Workers’ Compensation Board) v. Martin , 2003 SCC 54 , [2003] 2 S.C.R. 504 ; Alberta v. Hutterian Brethren of Wilson Colony , 2009 SCC 37 , [2009] 2 S.C.R. 567 ; RJR‑MacDonald Inc. v. Canada (Attorney General) , [1995] 3 S.C.R. 199 ; Eldridge v. British Columbia (Attorney General) , [1997] 3 S.C.R. 624 ; McKinney v. University of Guelph , [1990] 3 S.C.R. 229 ; Tétreault‑Gadoury v. Canada (Employment and Immigration Commission) , [1991] 2 S.C.R. 22 ; Vriend v. Alberta , [1998] 1 S.C.R. 493 .

By Deschamps J.

Not followed: Nova Scotia (Attorney General) v. Walsh , 2002 SCC 83 , [2002] 4 S.C.R. 325 ; referred to: R. v. Kapp , 2008 SCC 41 , [2008] 2 S.C.R. 483 ; Withler v. Canada (Attorney General) , 2011 SCC 12 , [2011] 1 S.C.R. 396 ; Law v. Canada (Minister of Employment and Immigration) , [1999] 1 S.C.R. 497 ; Miron v. Trudel , [1995] 2 S.C.R. 418 ; Bracklow v. Bracklow , [1999] 1 S.C. R. 420 ; Moge v. Moge , [1992] 3 S.C.R. 813 ; M. v. H. , [1999] 2 S.C. R. 3 .

By McLachlin C.J.

Not followed: Nova Scotia (Attorney General) v. Walsh , 2002 SCC 83 , [2002] 4 S.C.R. 325 ; referred to: Miron v. Trudel , [1995] 2 S.C.R. 418 ; R. v. Kapp , 2008 SCC 41 , [2008] 2 S.C.R. 483 ; Withler v. Canada (Attorney General) , 2011 SCC 12 , [2011] 1 S.C.R. 396 ; Law v. Canada (Minister of Employment and Immigration) , [1999] 1 S.C.R. 497 ; Andrews v. Law Society of British Columbia , [1989] 1 S.C.R. 143 ; R. v. Oakes , [1986] 1 S.C.R. 103 ; R. v. Edwards Books and Art Ltd. , [1986] 2 S.C.R. 713 ; Irwin Toy Ltd. v. Quebec (Attorney General) , [1989] 1 S.C.R. 927 ; Alberta v. Hutterian Brethren of Wilson Colony , 2009 SCC 37 , [2009] 2 S.C.R. 567 ; Lavoie v. Canada , 2002 SCC 23 , [2002] 1 S.C.R. 769 ; RJR‑MacDonald Inc. v. Canada (Attorney General) , [1995] 3 S.C.R. 199 ; McKinney v. University of Guelph , [1990] 3 S.C.R. 229 ; R. v. Advance Cutting & Coring Ltd. , 2001 SCC 70 , [2001] 3 S.C.R. 209 .

Statutes and Regulations Cited

Act instituting civil unions and establishing new rules of filiation , S.Q. 2002, c. 6, s. 143.

Act respecting duties on transfers of immovables , R.S.Q., c. D‑15.1.

Act respecting financial assistance for education expenses , R.S.Q., c. A‑13.3.

Act respecting financial services cooperatives , R.S.Q., c. C‑67.3.

Act respecting industrial accidents and occupational diseases , R.S.Q., c. A‑3.001.

Act respecting insurance , R.S.Q., c. A‑32.

Act respecting labour standards , R.S.Q., c. N‑1.1.

Act respecting legal aid and the provision of certain other legal services , R.S.Q., c. A‑14.

Act respecting matrimonial regimes , S.Q. 1969, c. 77.

Act respecting school elections , R.S.Q., c. E‑2.3.

Act respecting the Civil Service Superannuation Plan , R.S.Q., c. R‑12.

Act respecting the conditions of employment and the pension plan of the Members of the National Assembly , R.S.Q., c. C‑52.1.

Act respecting the Government and Public Employees Retirement Plan , R.S.Q., c. R‑10.

Act respecting the legal capacity of married women , S.Q. 1964, c. 66.

Act respecting the Pension Plan of Certain Teachers , R.S.Q., c. R‑9.1.

Act respecting the Pension Plan of Elected Municipal Officers , R.S.Q., c. R‑9.3.

Act respecting the Pension Plan of Peace Officers in Correctional Services , R.S.Q., c. R‑9.2.

Act respecting the Québec Pension Plan , R.S.Q., c. R‑9.

Act respecting the Québec sales tax , R.S.Q., c. T‑0.1.

Act respecting the Teachers Pension Plan , R.S.Q., c. R‑11.

Act respecting trust companies and savings companies , R.S.Q., c. S‑29.01.

Act to amend the Civil Code and the Code of Civil Procedure respecting the civil rights of women , S.Q. 1931, c. 101, s. 27.

Act to amend the Civil Code as regards marriage , S.Q. 2004, c. 23, s. 7.

Act to amend the Civil Code of Québec and other legislation in order to favour economic equality between spouses , S.Q. 1989, c. 55.

Act to amend various legislative provisions concerning de facto spouses , S.Q. 1999, c. 14.

Act to establish a new Civil Code and to reform family law , S.Q. 1980, c. 39 .

Automobile Insurance Act , R.S.Q., c. A‑25.

Canadian Charter of Rights and Freedoms , ss. 1, 15 .

Civil Code of Lower Canada , arts. 173, 768, 1056.

Civil Code of Québec [en. S.Q. 1980, c. 39, s. 1 ], art. 633 .

Civil Code of Québec , S.Q. 1991, c. 64, arts. 15, 264, 266, 269, 391, 392, 394, 395, 396, 401 to 430, 404, 409, 410, 414 et seq. , 416, 419, 423, 427 et seq. , 431, 432 et seq. , 433, 438, 448 to 484, 461, 467, 485, 486, 487, 492, 507, 511, 512, 516, 521.1, 521.6, 521.8, 521.12, 521.13, 521.17, 522, 585 to 596.1, 587, 597 et seq. , 604, 653 et seq. , 655, 1030, 1493 to 1496, 1818, 1819, 1938, 1958, 2380, 2419.

Civil Marriage Act , S.C. 2005, c. 33.

Code of Civil Procedure , R.S.Q., c. C‑25, art. 553.

Common‑Law Partners’ Property and Related Amendments Act , S.M. 2002, c. 48.

Cooperatives Act , R.S.Q., c. C‑67.2.

Courts of Justice Act , R.S.Q., c. T‑16.

Divorce Act , R.S.C. 1985, c. 3 (2nd Supp.), s. 15.2 .

Divorce Act , S.C. 1967‑68, c. 24.

Domestic Relations Act , R.S.A. 2000, c. D‑14.

Family Law Act , R.S.N.L. 1990, c. F‑2.

Family Law Act , R.S.O. 1990, c. F.3.

Family Law Act , R.S.P.E.I. 1988, c. F‑2.1.

Family Law Act , S.N.W.T. 1997, c. 18.

Family Maintenance Act , R.S.M. 1987, c. F20.

Family Maintenance Act, 1997 , S.S. 1997, c. F‑6.2.

Family Relations Act , R.S.B.C. 1996, c. 128.

Family Services Act , S.N.B. 1980, c. F‑2.2.

Individual and Family Assistance Act , R.S.Q., c. A‑13.1.1.

Insurance Act , R.S.O. 1980, c. 218.

Interpretation Act , R.S.Q., c. I‑16, s. 61.1.

Law Reform (2000) Act , S.N.S. 2000, c. 29.

Maintenance and Custody Act , R.S.N.S. 1989, c. 160, s. 52(1).

Maintenance Enforcement Act , S.N.S. 1994‑95, c. 6, s. 2(e).

Matrimonial Property Act , R.S.N.S. 1989, c. 275, s. 2(g), “spouse”.

Miscellaneous Statutes (Domestic Relations) Amendment Act, 2001 (No. 2) , S.S. 2001, c. 51.

Supplemental Pension Plans Act , R.S.Q., c. R‑15.1.

Taxation Act , R.S.Q., c. I‑3.

Vital Statistics Act , R.S.N.S. 1989, c. 494.

Workers’ Compensation Act , R.S.Q., c. A‑3.

Authors Cited

Bala, Nicholas. “Controversy Over Couples in Canada: The Evolution of Marriage and Other Adult Interdependent Relationships” (2003), 29 Queen’s L.J. 41.

Barak, Aharon. Proportionality: Constitutional Rights and their Limitations , trans. by Doron Kalir. Cambridge: Cambridge University Press, 2012.

Baudouin, Jean‑Louis. “Examen critique de la situation juridique de l’enfant naturel” (1966), 12 McGill L.J. 157.

Baudouin, Jean‑Louis, et Pierre‑Gabriel Jobin. Les obligations , 6 e éd. par Pierre‑Gabriel Jobin avec la collaboration de Nathalie Vézina. Cowansville, Qué.: Yvon Blais, 2005.

Bredt, Christopher D., and Adam M. Dodek. “Breaking the Law ’s Grip on Equality: A New Paradigm for Section 15 ” (2003), 20 S.C.L.R. (2d) 33 .

British Columbia. Royal Commission on Family and Children’s Law. Family Maintenance . Vancouver: The Commission, 1975.

British Columbia Law Institute. Report on Recognition of Spousal and Family Status . Vancouver: The Institute, 1998.

Brodsky, Gwen. “Case Comment: Gosselin v. Quebec (Attorney General) : Autonomy with a Vengeance” (2003), 15 C.J.W.L. 194.

Burman, Danielle. “Politiques législatives québécoises dans l’aménagement des rapports pécuniaires entre époux: d’une justice bien pensée à un semblant de justice — un juste sujet de s’alarmer” (1988), 22 R.J.T. 149.

Cameron, B. Jamie. “A Work in Progress: The Supreme Court and the Charter ’s Equation of Rights and Limits”, in Debra M. McAllister and Adam M. Dodek, eds., The Charter at Twenty: Law and Practice 2002 . Toronto: Ontario Bar Association, 2002, 31.

Canada. Law Reform Commission. Studies on Family Property Law . Ottawa: Information Canada, 1975.

Canada. Law Reform Commission. Working Paper 12. Maintenance on Divorce . Ottawa: Information Canada, 1975.

Caparros, Ernest. “Le patrimoine familial: une qualification difficile” (1994), 25 R.G.D. 251 .

Caparros, Ernest. Les régimes matrimoniaux au Québec , 3 e éd. Montréal: Wilson & Lafleur, 1988.

Comtois, Roger. Traité théorique et pratique de la communauté de biens . Montréal: Recueil de droit et de jurisprudence, 1964.

Cossette, André. “Le concubinage au Québec” (1985), 88 R. du N. 42.

D.‑Castelli, Mireille, et Dominique Goubau. Le droit de la famille au Québec , 5 e éd. Saint‑Nicholas, Qué.: Presses de l’Université Laval, 2005.

Deleury, Edith, et Marlène Cano. “Le concubinage au Québec et dans l’ensemble du Canada: Deux systèmes juridiques, deux approches”, dans Jacqueline Rubellin‑Devichi, dir., Des concubinages dans le monde . Paris: Centre national de la recherche scientifique, 1990, 85.

Dworkin, Ronald. Taking Rights Seriously . Cambridge, Mass.: Harvard University Press, 1977.

Goubau, Dominique. “La conjugalité en droit privé: comment concilier ‘autonomie’ et ‘protection’?”, dans Pierre‑Claude Lafond et Brigitte Lefebvre, dir., L’union civile: nouveaux modèles de conjugalité et de parentalité au 21 e siècle . Cowansville, Qué.: Yvon Blais, 2003, 153.

Goubau, Dominique, Ghislain Otis et David Robitaille. “La spécificité patrimoniale de l’union de fait: le libre choix et ses ‘dommages collatéraux’” (2003), 44 C. de D. 3.

Greschner, Donna. “The Purpose of Canadian Equality Rights” (2002), 6 Rev. Const. Stud. 291.

Hogg, Peter W. Constitutional Law of Canada , 5th ed. Supp., vol. 2. Toronto: Thomson/Carswell, 2007 (loose-leaf updated 2012, release 1).

Holland, Winifred. “Intimate Relationships in the New Millennium: The Assimilation of Marriage and Cohabitation?” (2000), 17 Can. J. Fam. L. 114.

Jarry, Jocelyne. Les conjoints de fait au Québec: vers un encadrement légal. Cowansville, Qué.: Yvon Blais, 2008.

Koshan, Jennifer, and Jonnette Watson Hamilton. “Meaningless Mantra: Substantive Equality after Withler ” (2011), 16 Rev. Const. Stud. 31.

Langevin, Louise. “Liberté de choix et protection juridique des conjoints de fait en cas de rupture: difficile exercice de jonglerie” (2009), 54 McGill L.J. 697.

Leckey, Robert. “Chosen Discrimination” (2002), 18 S.C.L.R. (2d) 445 .

Lefebvre, Brigitte. “L’évolution de la notion de conjoint en droit québécois”, dans Pierre‑Claude Lafond et Brigitte Lefebvre, dir., L’union civile: nouveaux modèles de conjugalité et de parentalité au 21 e siècle . Cowansville, Qué.: Yvon Blais, 2003, 3.

Manitoba. Law Reform Commission. Reports on Family Law , Part I — The Support Obligation , Report #23. Winnipeg: The Commission, 1976.

Massé, Sylvie. “Les régimes matrimoniaux au Canada — Analyse comparative des législations provinciales” (1985), 88 R. du N. 103.

McIntyre, Sheila. “Deference and Dominance: Equality Without Substance”, in Sheila McIntyre and Sanda Rodgers, eds., Diminishing Returns: Inequality and the Canadian Charter of Rights and Freedoms . Markham, Ont.: LexisNexis Canada, 2006, 95.

Mignault, Pierre‑Basile. Le droit civil canadien , t. 6. Montréal: C. Théoret, 1902.

Moore, Benoît. “Culture et droit de la famille: de l’institution à l’autonomie individuelle” (2009), 54 McGill L.J. 257.

Moreau, Sophia. “ R. v. Kapp : New Directions for Section 15 ” (2008‑2009), 40 Ottawa L. Rev. 283.

Moreau, Sophia. “The Promise of Law v. Canada ” (2007), 57 U.T.L.J. 415.

Moreau, Sophia R. “The Wrongs of Unequal Treatment” (2004), 54 U.T.L.J. 291.

Nova Scotia. Law Reform Commission. Final Report: Reform of the Law Dealing with Matrimonial Property in Nova Scotia . Halifax: The Commission, 1997.

Ontario. Law Reform Commission. Report on the Rights and Responsibilities of Cohabitants under the Family Law Act . Toronto: The Commission, 1993.

Ontario. Ministry of the Attorney General. Family Law Reform . Toronto: The Ministry, 1976.

Pineau, Jean, et Danielle Burman. Effets du mariage et régimes matrimoniaux . Montréal: Thémis, 1984.

Pineau, Jean, et Marie Pratte. La famille . Montréal: Thémis, 2006.

Proulx, Daniel. “Le concept de dignité et son usage en contexte de discrimination: deux Chartes, deux modèles”, [2003] R. du B. (numéro spécial) 485.

Quebec. Assemblée nationale. Commission permanente des institutions. Journal des débats , vol. 37, n o 46, 2 e sess., 36 e lég., 12 février 2002, p. 4‑5.

Quebec. Assemblée nationale. Journal des débats , vol. 23, n o 15, 6 e sess., 31 e lég., 4 décembre 1980, p. 608 et 663.

Quebec. Assemblée nationale. Journal des débats , vol. 30, n o 125, 2 e sess., 33 e lég., 8 juin 1989, p. 6487, 6489 et 6497.

Quebec. Assemblée nationale. Journal des débats , n o 197, 2 e sess., 35 e lég., 18 juin 1998, p. 12069‑12070.

Quebec. Assemblée nationale. Journal des débats , vol. 37, n o 96, 2 e sess., 36 e lég., 7 mai 2002, p. 5816.

Quebec. Assemblée nationale. Sous‑commission des institutions. Journal des débats , n o 22, 1 re sess., 34 e lég., 19 novembre 1991, p. 859.

Quebec. Civil Code Revision Office. Report on Matrimonial Regimes . Montréal: Official Publisher, 1968.

Quebec. Civil Code Revision Office. Report on the Québec Civil Code , vol. I — Draft Civil Code . Québec: Éditeur officiel, 1978.

Quebec. Civil Code Revision Office. Report on the Québec Civil Code , vol. II — Commentaries , t. 1. Québec: Éditeur officiel, 1978.

Quebec. Conseil du statut de la femme. Mémoire du Conseil du statut de la femme présenté lors de la consultation générale sur les droits économiques des conjoints . Québec: Le Conseil, 1988.

Quebec. Conseil du statut de la femme. Mémoire présenté à la Commission parlementaire sur la réforme du droit de la famille . Québec: Le Conseil, 1979.

Quebec. Les droits économiques des conjoints , par Herbert Marx et Monique Gagnon Tremblay. Sainte‑Foy, Qué.: Ministère de la Justice, 1988.

Réaume, Denise G. “Discrimination and Dignity” (2003), 63 La. L. Rev. 645.

Roy, Alain. “La charte de vie commune ou l’émergence d’une pratique réflexive du contrat conjugal” (2007), 41 R.J.T. 399.

Roy, Alain. “Le contrat de mariage en droit québécois: un destin marqué du sceau du paradoxe” (2006), 51 McGill L.J. 665.

Roy, Alain. Le contrat de mariage réinventé: Perspectives socio‑juridiques pour une réforme . Montréal: Thémis, 2002.

Roy, Alain. “Le régime juridique de l’union civile: entre symbolisme et anachronisme”, dans Pierre‑Claude Lafond et Brigitte Lefebvre, dir., L’union civile: nouveaux modèles de conjugalité et de parentalité au 21 e siècle . Cowansville, Qué.: Yvon Blais, 2003, 165.

Ryder, Bruce, Cidalia C. Faria and Emily Lawrence. “What’s Law Good For? An Empirical Overview of Charter Equality Rights Decisions” (2004), 24 S.C.L.R. (2d) 103 .

Saskatchewan. Law Reform Commission. “Common Law Relationships Under the Matrimonial Property Act ”, discussion paper, July 1997 (online: http://www.lawreformcommission.sk.ca/Commonlaw.pdf).

Senécal, Jean‑Pierre. Droit de la famille québécois , vol. 1 et 2. Brossard, Qué.: Publications CCH, 1985 (feuilles mobiles mises à jour septembre 2012, envoi n o 374).

Sylvestre, Jean. “Les accords entre concubins”, [1981] 1 C.P. du N. 195.

Tarnopolsky, Walter Surma. Discrimination and The Law in Canada . Toronto: Richard De Boo, 1982.

Tétrault, Michel. Droit de la famille , vol. 1, 4 e éd. Cowansville, Qué.: Yvon Blais, 2010.

Tétrault, Michel. “L’union civile: j’me marie, j’me marie pas”, dans Pierre‑Claude Lafond et Brigitte Lefebvre, dir., L’union civile: nouveaux modèles de conjugalité et de parentalité au 21 e siècle . Cowansville, Qué.: Yvon Blais, 2003, 101.

Traité de droit civil du Québec , t. 10, par Léon Faribault. Montréal: Wilson et Lafleur, 1952.

Tremblay, Luc B. “Promoting Equality and Combating Discrimination Through Affirmative Action: The Same Challenge? Questioning the Canadian Substantive Equality Paradigm” (2012), 60 Am. J. Comp. L. 181.

Young, Margot. “Blissed Out: Section 15 at Twenty”, in Sheila McIntyre and Sanda Rodgers, eds., Diminishing Returns: Inequality and the Canadian Charter of Rights and Freedoms . Markham, Ont.: LexisNexis Canada, 2006, 45.

Young, Margot. “Unequal to the Task: ‘Kapp’ing the Substantive Potential of Section 15”, in Sanda Rodgers and Sheila McIntyre, eds., The Supreme Court of Canada and Social Justice: Commitment, Retrenchment or Retreat . Markham, Ont.: LexisNexis, 2010, 183.

APPEALS from a judgment of the Quebec Court of Appeal (Beauregard, Dutil and Giroux JJ.A.), 2010 QCCA 1978 , [2010] R.J.Q. 2259 , [2010] R.D.F. 659 , 89 R.F.L. (6th) 1 , [2010] Q.J. No. 11091 (QL) , 2010 CarswellQue 15654 , SOQUIJ AZ‑50685017 , affirming in part a decision of Hallée J., 2009 QCCS 3210 , [2009] R.J.Q. 2070 , [2009] R.D.F. 545 , 67 R.F.L. (6th) 315 , [2009] Q.J. No. 7153 (QL) , 2009 CarswellQue 14051 , SOQUIJ AZ‑50566038 . Appeals of the Attorney General of Quebec and B allowed, appeal of A dismissed, Deschamps, Cromwell and Karakatsanis JJ. dissenting in part in the result and Abella J. dissenting in the result.

Benoît Belleau and Hugo Jean , for the appellant/respondent the Attorney General of Quebec.

Guy J. Pratte and Mark Phillips , for the appellant/respondent A.

Pierre Bienvenu , Suzanne H. Pringle , Catherine Martel and Azim Hussain , for the appellant/respondent B.

Gaétan Migneault , for the intervener the Attorney General of New Brunswick.

Robert J. Normey , for the intervener the Attorney General of Alberta.

Jocelyn Verdon , Dominique Goubau and Mireille Pélissier‑Simard , for the intervener Fédération des associations de familles monoparentales et recomposées du Québec.

Martha McCarthy and Johanne Elizabeth O’Hanlon , for the intervener the Women’s Legal Education and Action Fund.

English version of the judgment of LeBel, Fish, Rothstein and Moldaver JJ. delivered by

LeBel J. —

I. Introduction

[ 1 ] The issue raised by the parties in these appeals is whether it is valid to exclude de facto spouses from the patrimonial and support rights granted to married and civil union spouses. Does this exclusion violate the right to equality guaranteed by s. 15 of the Canadian Charter of Rights and Freedoms (“ Charter ”)?

[ 2 ] The Court must determine whether the provisions of the Civil Code of Québec , S.Q. 1991, c. 64 (“ C.C.Q. ”), dealing with the family residence (arts. 401 et seq. ), the family patrimony (arts. 414 et seq. ), the compensatory allowance (arts. 427 et seq. ), the partnership of acquests (arts. 432 et seq. ) and the obligation of spousal support ( art. 585 ) infringe s. 15(1) of the Charter because their application is limited to private legal relationships between married spouses and civil union spouses (see arts. 521.6 and 521.8 C.C.Q. ).

[ 3 ] The Court must therefore decide whether the exclusion of de facto spouses from the scope of these provisions is discriminatory within the meaning of s. 15(1). For the reasons that follow, I am of the opinion that the exclusion is not discriminatory within the meaning of s. 15(1) and accordingly does not violate the right to equality guaranteed by s. 15 of the Charter .

II. The Dispute

[ 4 ] Parties Ms. A and Mr. B met in 1992 in A’s native country. A, who was 17 years old at the time, was living with her parents and attending secondary school. B, who was 32 years old, was running a large international business. From 1992 to 1994, the parties travelled the world together several times a year. B provided A with financial support so that she could continue her schooling. In early 1995, the parties agreed that A would come to live in Quebec, where B lived.

[ 5 ] The parties broke up for the first time in late July 1995. They saw each other again during the holiday season, and then again in February 1996. A then became pregnant with her first child. The couple had three children together, born in 1996, 1999 and 2001. During the time they lived together, A did not hold employment. She regularly accompanied B on his trips, and he provided for all her needs and for the children’s needs. A wanted to get married, but B told her that he did not believe in the institution of marriage and that he might consider getting married after living with her for 25 years. The parties separated in 2002. They had lived together for a total of seven years.

[ 6 ] In February 2002, A filed a motion in the Quebec Superior Court seeking custody of the children, support, a lump sum, use of the family residence, a provision for costs and an interim order. The motion was accompanied by a notice to the Attorney General of Quebec stating that A intended to challenge the constitutionality of several provisions of the Civil Code of Québec in order to obtain the same legal regime for de facto spouses that existed for married spouses. More specifically, A claimed support for herself, a lump sum, partition of the family patrimony and the legal matrimonial regime of partnership of acquests. She also sought to reserve her right to claim a compensatory allowance. A claim concerning the use of the family residence was settled in an agreement between A and B.

[ 7 ] Since the constitutionality of the provisions relating to child custody and the child support obligation had not been challenged, the Superior Court awarded the parties joint custody of the children and awarded A $34,260.24 a month in child support and a provision for costs on May 16, 2006. The court also made a series of orders requiring B to pay certain specific expenses, including the children’s tuition fees, expenses related to their extracurricular activities, the salaries of two nannies and the salary of a cook working for A. As well, the court ordered B to continue paying all costs, school and municipal taxes, home insurance premiums and general maintenance and renovation costs required for the residence where the parties had agreed that A and the children would live. B remained the owner of that residence.

[ 8 ] The appeals relate solely to the constitutional aspect of the case and concern only the provisions of the Civil Code of Québec alleged by A to be invalid under s. 15 of the Charter . Hallée J. of the Superior Court ruled on the constitutional issues on July 16, 2009. She found that the impugned provisions did not violate the right to equality guaranteed by s. 15(1), and she denied A’s requests for a declaration of constitutional invalidity, which had been opposed by B and the Attorney General of Quebec. A then appealed to the Quebec Court of Appeal.

[ 9 ] On November 3, 2010, the Quebec Court of Appeal allowed A’s appeal in part. Dutil J.A., with whom Giroux J.A. concurred, declared art. 585 C.C.Q. , which provides for the obligation of spousal support, to be of no force or effect on the basis of an unjustified infringement of the right to equality set out in s. 15(1) of the Charter . However, Dutil J.A. upheld the Superior Court’s decision as regards the constitutionality of the provisions concerning the family residence, the family patrimony, the compensatory allowance and the partnership of acquests. In her opinion, those provisions are not discriminatory and therefore do not infringe s. 15(1). Dutil J.A. also suspended the declaration of constitutional invalidity of art. 585 for 12 months to give the Quebec legislature time to amend the provision in order to make it consistent with the Charter . Beauregard J.A. dissented on the issue of the appropriate remedy. He concluded that the declaration of constitutional invalidity of arts. 511 and 585 C.C.Q. should apply immediately so that A could benefit from the obligation of spousal support without delay.

[ 10 ] In this Court, B and the Attorney General of Quebec are appealing the Court of Appeal’s decision to strike down art. 585 on the obligation of spousal support. A is also appealing that decision. She takes issue with the conclusion that the Civil Code ’s provisions concerning the family residence, the family patrimony, the compensatory allowance and the partnership of acquests are constitutionally valid. To ensure that the issues in these appeals are fully understood, I will begin by reviewing the proceedings in the Superior Court and the Court of Appeal in greater detail.

III. Judicial History

A. Quebec Superior Court, 2009 QCCS 3210 , [2009] R.J.Q. 2070

[ 11 ] This case came before Hallée J. by way of a Charter motion. In addition to her claims based on s. 15 of the Charter , A originally made certain arguments concerning the division of constitutional powers between Parliament and the provincial legislatures as regards the definition of marriage. Hallée J. rejected all those arguments, and A abandoned them on appeal. In this Court, the only remaining issues have to do with the equality guarantee set out in the Charter .

[ 12 ] Hallée J. began with an overview of the legal situation of de facto spouses in Quebec. She noted that they cannot bring support proceedings against one another or partition the family patrimony, and that they are not governed by a legal matrimonial regime. However, they are treated in the same way as married spouses for the purposes of life insurance (art. 2419 C.C.Q. ), annuities (art. 2380 C.C.Q. ) and the protective supervision of incapable or vulnerable persons (arts. 264, 266 and 269 C.C.Q. ). Moreover, they are authorized by art. 15 C.C.Q. to consent to care for a person of full age who is incapable of giving consent. Hallée J. added that the Quebec legislature has enacted a number of social or tax laws (including the Act respecting the Québec Pension Plan , R.S.Q., c. R‑9, and the Taxation Act , R.S.Q., c. I‑3) that grant de facto spouses benefits similar to the ones already available to married spouses. With these exceptions, only a cohabitation agreement can govern the rights of de facto spouses. In such an agreement, de facto spouses can provide for, among other things, an obligation of support in the event of a breakdown.

[ 13 ] Hallée J. then summarized the most relevant points from the expert reports filed by the parties. Although none of the experts had been heard during the trial, some of them had been examined out of court. Hallée J. found on the basis of these reports that de facto unions were a growing phenomenon in contemporary Quebec society:

[ translation ] After having read the expert reports attentively, the Court finds that the phenomenon of de facto unions is growing in Quebec. From 1981 to 2006, the proportion of couples living in a de facto union grew from 7.9% to 34.6%.

The 2006 Statistics Canada census indicates that 34.6% of Quebeckers live in a de facto union, while an average of 18.4% of couples throughout Canada choose to live in this type of relationship. Thus, Quebec is far in the lead in terms of the number of couples living in a de facto union. Moreover, according to the Institut de la statistique du Québec, 60% of children in Quebec are born out of wedlock.

Some experts would therefore like to see the legislature intervene to regulate these unions, while others believe that further study is required before drawing [actual] conclusions about this phenomenon. [paras. 59‑61]

[ 14 ] Hallée J. then noted that, in R. v. Kapp , 2008 SCC 41 , [2008] 2 S.C.R. 483 , this Court had adopted a two‑part test for finding that a distinction is discriminatory in the constitutional sense: (1) Does the law create a distinction based on an enumerated or analogous ground? (2) Does the distinction create a disadvantage by perpetuating prejudice or stereotyping?

[ 15 ] Hallée J. added that marital status has been recognized as a ground analogous to the grounds of discrimination enumerated in s. 15(1). She also noted that, to find that s. 15(1) has been infringed, it is not enough to establish the existence of a legislative distinction based on an analogous ground. According to the case law, the claimant must prove that the differential treatment has a purpose or effect that discriminates in a substantive sense.

[ 16 ] In this case, Hallée J. concluded that A had not shown that the distinction between de facto spouses and married spouses resulting from the impugned provisions had substantively discriminatory effects and added that the lack of evidence in this regard was fatal to A’s action. Hallée J. stressed the limitations of the expert reports filed by the parties. Those reports indicated that living conditions were by and large better for intact married families than for the new forms of family, but they did not assess the impact of the impugned provisions, particularly in the event of a breakdown. Hallée J. found on the basis of this evidence that de facto spouses in Quebec are not subject to [ translation ] “any stereotypical disadvantages or prejudice”. In her opinion, “the legislature’s purpose in preserving a distinction between marriage and de facto union is to safeguard freedom of choice and to respect the dignity and autonomy of de facto spouses” (para. 222). Finally, she concluded that A had not established concrete effects of the distinctions between de facto and married spouses either during the conjugal relationship or upon its breakdown.

[ 17 ] However, Hallée J. did not stop there. She also took the precedential value of Nova Scotia (Attorney General) v. Walsh , 2002 SCC 83 , [2002] 4 S.C.R. 325 , into account. In that case, this Court had held that, when the rights and obligations of common law spouses vis‑à‑vis each other are in issue, the spouses’ choice to marry or not to marry becomes the most important factor for the purposes of s. 15(1) of the Charter . Since the distinctions between the rights and obligations of married spouses and those of common law spouses reflect choices made by those individuals, the consequences of the choices do not infringe s. 15 . In short, Hallée J. found that Walsh fully disposed of A’s Charter ‑based constitutional arguments. In her view, Walsh showed that the possibility for de facto spouses to make such choices is consistent with the fundamental purpose of s. 15 of the Charter :

[ translation ] In the view of this Court, this is the fundamental point in Walsh . Whether to establish an identical protective regime, regardless of the choice made regarding marital status, is not a decision that falls within the purview of the courts, provided the choices made by the legislature are not discriminatory. It is the legislature’s task to determine whether it is necessary to impose, in whole or in part, a universal and standardized protective regime that does not take into account the matrimonial status of the de facto spouses. [para. 249]

[ 18 ] Hallée J. then rejected A’s argument that Walsh could be distinguished from the case at bar because there was an obligation of support between common law spouses in Nova Scotia that does not exist in Quebec law. In her opinion, the majority’s reasons in Walsh were not based on the existence of such an obligation in Nova Scotia. Hallée J. found that Walsh instead reflected the fundamental importance of freedom of choice and that this factor is just as applicable in Quebec as in Nova Scotia. The rights of married couples are not denied to de facto spouses, who can choose to benefit from them in different ways, by, for example, contracting a civil union or entering into an agreement. In addition to freedom of choice, Hallée J. stated, other contextual factors supported the finding that there was no discrimination in the instant case. As in Walsh , the evidence showed that distinctions exist between married and de facto couples, and that there is significant heterogeneity within the group consisting of de facto spouses. Moreover, the legislative history of the Civil Code , like that of the legislation at issue in Walsh , shows that the legislature’s consistent and considered purpose for the past 30 years has been to respect the freedom of every individual to choose whether or not to marry. As a result, Hallée J. found that the differential treatment of de facto spouses and married persons with respect to the obligation of support and the partition of property does not perpetuate prejudice or result from stereotyping.

[ 19 ] For all these reasons, Hallée J. denied the constitutional conclusions sought by A and found that the provisions of the Civil Code of Québec A challenges are constitutional.

B. Quebec Court of Appeal, 2010 QCCA 1978 , [2010] R.J.Q. 2259

(1) Reasons of Dutil J.A.

[ 20 ] Dutil J.A. considered A’s appeal in reasons concurred in by Giroux J.A. She found first that the main issues raised by the appeal concerned, on the one hand, the obligation of support of de facto spouses and, on the other hand, the partition of property upon separation, that is, the right to partition of the family patrimony, protection of the family residence, the partnership of acquests and the compensatory allowance.

[ 21 ] According to Dutil J.A., the trial judge had been correct in concluding that this Court’s decision in Walsh is binding on the Quebec courts with respect to the partition of property between de facto spouses upon separation. She explained this as follows:

[ translation] In the case before us, the impugned C.C.Q. provisions pertaining to the division of property govern patrimonial relations between married spouses. On this issue, the Supreme Court has spoken clearly, stating that the freedom to choose whether to marry or not is paramount. Although in Quebec the legislature has stipulated that the C.C.Q. provisions governing the effects of marriage are of public order (article 391 C.C.Q.), while in Nova Scotia married spouses can choose not to be subject to the MPA [ Matrimonial Property Act ], this does not in my opinion permit Walsh to be distinguished from the present case on this point.

The Quebec legislature has addressed the issue of conjugal status and de facto unions on a number of occasions (1980, 1989, 1991, 1999, 2002) and has deliberately decided to allow spouses the freedom to choose the type of relationship they wish. If the issue is to be revisited from the perspective of the division of assets, this should be done by the legislature in light of the changes that have taken place in society, since the Supreme Court has determined that the legislative choice already made on this issue does not contravene section 15 of the Charter . [paras. 59‑60]

[ 22 ] Dutil J.A. then considered the obligation of support provided for in art. 585 C.C.Q. She disagreed with the trial judge on this point, finding that Walsh does not have precedential value with respect to the obligation of spousal support. On the one hand, she noted that common law spouses already had an obligation of support under the Nova Scotia legislation considered by this Court in Walsh . On the other hand, she found that there is an important distinction between the obligation of support and the provisions on partition of property. In her view, [ translation ] “support payments exist to meet basic needs and represent an aspect of social solidarity, whereas the division of property is contractual in origin” (para. 68).

[ 23 ] Dutil J.A. therefore considered whether art. 585 C.C.Q. infringes s. 15(1) of the Charter . Applying the test established by this Court in Kapp , she concluded that the first stage of the analysis was not problematic, since marital status had already been held to be a ground of discrimination analogous to the ones enumerated in s. 15(1). The case therefore turned on the second stage of the analysis, namely whether the distinction created a disadvantage by perpetuating prejudice or stereotyping. She concluded that by failing to mention de facto spouses in art. 585 C.C.Q. , the legislature had created a disadvantage by stereotyping and by expressing a prejudice.

[ 24 ] Although she acknowledged that the legislative disadvantages that once existed for de facto spouses have become less significant and that the de facto union is now socially acceptable, Dutil J.A. concluded that [ translation ] “the fact remains that the legislature’s failure to include them within the protection afforded by article 585 C.C.Q. perpetuates the stereotype that these types of unions are less durable and serious than marriage and civil unions, which are recognized by means of a formal act” (para. 98). She stated that her conclusion concerning the existence of such a stereotype was based on the opinion expressed by McLachlin J. in Miron v. Trudel , [1995] 2 S.C.R. 418 , which Bastarache J. had reproduced in Walsh , about the historical disadvantage suffered by de facto spouses. She also relied on the fact that “concubinage” had been considered a reprehensible lifestyle choice before the 1980 family law reform. She added that, in enacting the new Civil Code , the legislature had deliberately refrained from including the de facto union in the provisions on the family because it considered such unions to be less stable than marriage. She relied in this regard on certain remarks made by the Quebec Minister of Justice in September and November 1991 in the course of the study of Bill 125 on the C.C.Q. by a parliamentary committee. Finally, although many recommendations had been made by an interdepartmental committee that considered the situation of de facto spouses in 1996, the Quebec legislature had not passed a bill to structure their mutual relationships.

[ 25 ] According to Dutil J.A., other signs of the disadvantages suffered by de facto spouses are still present in the Civil Code . For example, a de facto spouse can inherit from his or her spouse only by will (arts. 653 C.C.Q. et seq. ) and cannot make gifts of future property (arts. 1818 and 1819 C.C.Q. ). Finally, referring to M. v. H. , [1999] 2 S.C.R. 3 , in which this Court explained that one factor which may demonstrate that a distinction violates a person’s dignity is the vulnerability of the group in question, Dutil J.A. stated that de facto spouses are vulnerable just as same‑sex spouses were. In her opinion, since the legislature had not made the obligation of support provided for in art. 585 applicable to de facto spouses, it deemed them less worthy of the protection afforded to married spouses even though de facto unions may be similar in several respects to the other types of conjugal relationships.

[ 26 ] Dutil J.A. referred to Gonthier J.’s comment in Walsh that the obligation of support has an important social objective that differs from the objective of the division of property. By providing in art. 585 C.C.Q. that spouses and relatives in the direct line in the first degree owe each other support, the Quebec legislature also recognized that the obligation of support is different. According to Dutil J.A., it [ translation ] “is not solely the consequence of a contractual agreement; rather, it is a social obligation toward members of the immediate family unit” (para. 101). In Quebec, the family unit concept now also includes families formed by de facto spouses. In ignoring such families, the Quebec legislature excluded more than a third of Quebec couples from the application of a measure that exists precisely to protect the family unit. Dutil J.A. also found that, by requiring marriage or civil union as a precondition for the right to support, the legislature had failed to consider social realities. The purpose of an obligation of support of former spouses is to enable an economically dependent person to obtain support, following the breakdown of a conjugal relationship, from a former spouse who is capable of paying it. The nature of the couple’s relationship, be it a de facto union, civil union or marriage, does not alter the extent to which one of the former spouses needs support after they separate.

[ 27 ] Rather, Dutil J.A. noted that a de facto union that lasts a certain length of time and produces children is very similar to marriage. Finally, she disagreed with the trial judge that the lack of evidence about the concrete effects of the distinction between de facto spouses and married spouses was fatal to A’s action. She referred in this regard to Law v. Canada (Minister of Employment and Immigration) , [1999] 1 S.C.R. 497 , in which this Court had stated that it will often be evident on the basis of facts of which judicial notice is taken and logical reasoning that a distinction is discriminatory within the meaning of s. 15(1) of the Charter . In her opinion, Hallée J. should therefore have taken judicial notice of the concrete effects of the distinction made by the legislature between de facto spouses and married spouses. She concluded that the Quebec legislature’s differential treatment of de facto spouses and married or civil union spouses with respect to the obligation of support has a substantive impact on de facto spouses. More specifically, in her view, art. 585 C.C.Q. deprives certain individuals of a right — the ability to meet basic financial needs following the breakdown of a relationship — however fundamental it may be, since they cannot claim support from a former spouse following separation. This exclusion exists regardless of the length of the union, the birth of children or the creation of a situation of economic dependence.

[ 28 ] Having concluded, in light of certain contextual factors, that the exclusion of de facto spouses from art. 585 C.C.Q. is discriminatory, Dutil J.A. held that this legislative provision is not justified under s. 1 of the Charter . As a result, she decided that the appropriate constitutional remedy in this case would be a declaration that art. 585 C.C.Q. is invalid; this declaration was to be suspended for 12 months without any exemption for A.

(2) Reasons of Beauregard J.A.

[ 29 ] Beauregard J.A. agreed that, because de facto spouses in Quebec are excluded from the right to support after separation, they are discriminated against in violation of s. 15(1) of the Charter and that this discrimination cannot be justified under s. 1 .

[ 30 ] However, Beauregard J.A. disagreed with Dutil J.A. about the appropriate remedy, since many de facto spouses would be deprived of support during the period in which the declaration of invalidity of art. 585 C.C.Q. was to be suspended. To avoid this result, he would have ordered that art. 585 C.C.Q. , which provides for the obligation of support, and art. 511 C.C.Q. , which allows a court to order the payment of support at the time of separation from bed and board, be immediately interpreted as applying to de facto spouses.

IV. Analysis

A. Issues

[ 31 ] After the parties appealed to this Court, the Chief Justice stated the following constitutional questions:

1. Do arts. 401 to 430, 432, 433, 448 to 484 and 585 of the Civil Code of Québec , S.Q. 1991, c. 64, infringe s. 15(1) of the Canadian Charter of Rights and Freedoms ?

2. If so, is the infringement a reasonable limit prescribed by law that can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

[ 32 ] A added the following question concerning the immediate application of a declaration of constitutional invalidity:

3. [ translation ] Did the majority of the Court of Appeal err in the choice of remedy, in particular by failing to immediately enable all Quebeckers to benefit from a remedy rectifying the constitutional invalidity in issue, and also with respect to the question of an individual remedy for the claimant?

[ 33 ] In determining whether the Civil Code ’s provisions on the relationship between spouses are consistent with the equality guarantee of s. 15(1) of the Charter , it will be necessary to consider certain questions related to this main issue. First, I will review the development of the framework for the legal relationship between spouses in Quebec, Quebec’s legislative policy as regards the distinction between married or civil union spouses and de facto spouses, and the development of the de facto union in Quebec society. Second, I will discuss the nature and application of the right to equality. Third, I will consider the nature and precedential value of Walsh . Lastly, I will go on to discuss the constitutional issues raised by the parties.

B. Positions of the Parties

(1) A

[ 34 ] A submits that the Court of Appeal properly struck down art. 585 C.C.Q. , which provides for the obligation of spouses to support one another, on the ground that it unjustifiably infringes s. 15(1) of the Charter . She also argues that the Court of Appeal erred in finding that the provisions on partition of property are not inconsistent with s. 15(1) . In her opinion, both the provisions on partition of property and the provision concerning the obligation of spousal support unjustifiably infringe the right to equality guaranteed by s. 15(1) .

[ 35 ] A adds that it was an error to apply this Court’s reasoning in Walsh to the provisions on partition of property. She submits that the Quebec scheme differs from the Nova Scotia scheme considered by this Court in Walsh and that Walsh did not resolve the issues in the case at bar. She also challenges its precedential value.

[ 36 ] A further argues that the theory of freedom of choice, consensualism or autonomy of the will relied on by the Attorney General and B to support the constitutional validity of the Civil Code provisions she is challenging does not correspond to reality. This Court cannot permit a legislature to justify discriminatory treatment by relying on a theory that has no connection with the reality of the persons being discriminated against. Such an approach would favour the recognition of distinctions based on prejudice and exempt the legislature from the obligation to ensure the substantive equality promised by s. 15 of the Charter . A argues that the exclusion of de facto spouses from the protection the Civil Code affords the family in relation to both the partition of property and the obligation of support disregards their actual situation and their needs.

[ 37 ] In short, A argues that it was up to the Attorney General to defend the legislation by showing that it is not based on prejudice or stereotypes. In her opinion, far from combatting stereotypes, the Attorney General and B are relying on them and basing their arguments on them.

[ 38 ] According to A, the appropriate remedy in this case would be for this Court to read de facto spouses into the impugned provisions. Her case would then have to be remitted to the Superior Court to determine the amount and duration of the support owed to her and partition the parties’ property.

(2) Attorney General of Quebec

[ 39 ] The Attorney General of Quebec submits that the Court of Appeal did not err in finding that the provisions on partition of property are not inconsistent with s. 15(1) of the Charter . However, he argues that the Court of Appeal erred in striking down art. 585 C.C.Q. , which provides for the obligation of spousal support, on the ground that it unjustifiably infringes s. 15(1) . In his opinion, the right to equality guaranteed by s. 15(1) is violated by neither the provisions on partition of property nor the provision concerning the obligation of spousal support.

[ 40 ] The Attorney General submits that, in light of the evidence and in accordance with this Court’s reasons in Walsh , the exclusion of de facto spouses from the impugned provisions of the Civil Code is not discriminatory. In his view, it does not create a disadvantage by perpetuating prejudice or stereotyping. According to the Attorney General, the reasons and principles stated by the majority in Walsh apply to all the impugned statutory provisions, including the one on the obligation of spousal support. He argues that any other conclusion would be illogical and contradictory. Otherwise, the respect shown by the legislature in the Civil Code for the freedom of choice of partners to marry or not to marry would on the one hand be interpreted as a legislative choice that is not based on stereotyping of or prejudice against the de facto union in the case of the matrimonial regime and the partition of property, whereas on the other hand, the same legislative choice would be considered to be tainted by stereotyping of or prejudice against the de facto union in the case of the obligation of support.

[ 41 ] The Attorney General submits that, with regard to both the obligation of support and the patrimony of the spouses, the Quebec legislature’s objective is the same one the Nova Scotia legislature was pursuing in Walsh : to respect individual autonomy. In other words, both for the obligation of spousal support and for patrimonial property, the exclusion of de facto spouses is intended to respect the freedom of every individual to choose whether to marry and thus whether to participate in the statutory scheme applicable to marriage and accept the specific legal consequences flowing from that scheme. According to the Attorney General, the distinction between married or civil union spouses and de facto spouses can be explained by the fact that the former have chosen to make a commitment the nature of which legally entails a certain number of rights and obligations, while the latter have not done so. Under the Quebec scheme, a decision to live together is not enough to show an intention to make such a commitment.

[ 42 ] In short, the Attorney General argues that to extend the Civil Code ’s provisions on the rights and obligations of marriage, including the obligation of support, to persons in de facto unions is a societal choice that falls within the political and not the judicial realm. However, if the Court were to conclude that some of the impugned provisions unjustifiably infringe s. 15, the Attorney General is opposed to reading in as a solution and prefers to have the declaration of invalidity suspended to enable the legislature to correct the constitutional defect. The Attorney General is also opposed to granting A an individual remedy during any period in which the declaration of invalidity is suspended.

(3) B

[ 43 ] B adopts the Attorney General’s positions on the validity of the Quebec legislative scheme and on the remedy the Court should grant if it reaches the conclusion proposed by A on the constitutional questions.

[ 44 ] B also argues that the evidence adduced in the Superior Court, this Court’s determinative case law on discrimination based on marital status, and an analysis of the relevant contextual factors lead necessarily to the conclusion that the impugned distinctions under Quebec law between de facto unions and marriage or civil unions are not discriminatory, as they do not perpetuate prejudice and do not result from stereotyping. Moreover, according to B, the principles from Walsh are directly applicable in the instant case, and the Court of Appeal erred in holding that they do not apply to the obligation of support.

[ 45 ] B points out that he and A, in their own way and for their own reasons, chose to live together outside the institution of marriage. For deeply personal reasons, he never wanted to participate in that institution. As for A, she chose to live with B in a conjugal relationship without making marriage a precondition to their cohabitation.

(4) Interveners

[ 46 ] In this Court, four interveners stated their positions on the constitutional challenge.

[ 47 ] Although New Brunswick has established a support remedy for common law spouses, the Attorney General of that province submits that the respect for freedom of choice shown by the Court in Walsh must apply in relation to both the provisions on partition of property and those on spousal support. In her view, freedom of choice characterizes the type of union contracted by the spouses, so there is no reason to distinguish the obligation of support from obligations with respect to the partition of property in the case of de facto unions.

[ 48 ] The Attorney General of Alberta, whose province also imposes an obligation of support on common law spouses, intervened only with respect to the provisions on partition of property. In his view, the Court disposed of the issue adequately in Walsh by stressing the importance of the factor of freedom of choice in the analysis under s. 15 of the Charter .

[ 49 ] The Women’s Legal Education and Action Fund, adopting some of A’s arguments, contends that spousal relationships are marked by gender inequality and that the Court should take that inequality into account in determining whether the impugned statutory provisions are discriminatory. Finally, this intervener asks that the Civil Code provisions relating to the obligation of support and the partition of property be applied to de facto spouses.

[ 50 ] Similarly, the Fédération des associations de familles monoparentales et recomposées du Québec supports A’s arguments concerning the infringement of the right to equality that results from the exclusion of de facto spouses from art. 585 on the obligation of support. Disagreeing with the weight attached to freedom of choice by the Attorney General and B, it disputes the validity of this factor. Moreover, as with the obligation of support, the Fédération submits that the exclusion of de facto spouses from the provisions on protection of the family residence discriminates against them and cannot be justified under s. 1 of the Charter . The Fédération also argues that this Court should consider the impact on children of the discrimination against their parents.

C. Changes in the Framework for Legal Relationships Between Spouses in Quebec Since 1980

[ 51 ] This case centres on the exclusion of de facto spouses from the scope of certain provisions of the Civil Code of Québec that shape legal relationships between spouses. To determine whether the provisions in question are discriminatory, it will be necessary to consider them in the relevant historical and legislative context. For this purpose, I will discuss the framework applicable to legal relationships between spouses in Quebec, Quebec legislative policy with respect to the distinction between de facto spouses and married or civil union spouses, and the development of the de facto union in Quebec society.

(1) Married Spouses

(a) Historical Review of the Situation of Married Spouses and Development of the Legal Framework for Their Relationships

[ 52 ] Before discussing the situation of de facto spouses, I must review the civil law’s matrimonial regimes for legally married spouses. On enacting the Civil Code of Lower Canada (“ C.C.L.C. ”) in 1866, the legislature gave spouses a choice, at the time of their marriage, between two principal matrimonial regimes: community of property and separation of property. Originally, under the 1866 Code , the choice made at the time of marriage was irrevocable for the duration of the marriage.

[ 53 ] The legal matrimonial regime, that is, the regime that applied where the spouses did not enter into a marriage contract choosing another regime, was that of “community of moveables and acquests” (“community of property”). The community of property was administered by the husband. [ translation ] “[T]he regime was organized around the supremacy of the husband . . ., the head of the community”: B. Lefebvre, “L’évolution de la notion de conjoint en droit québécois”, in P.‑C. Lafond and B. Lefebvre, eds., L’union civile: nouveaux modèles de conjugalité et de parentalité au 21 e siècle (2003), 3, at p. 11. In 1931, to give the wife some autonomy in relation to her husband, the legislature created the category of “reserved property” of the wife. Reserved property was property the wife acquired by working outside the household and over which she had certain powers of administration: An Act to amend the Civil Code and the Code of Civil Procedure respecting the civil rights of women , S.Q. 1931, c. 101, s. 27. After that, the Civil Code of Lower Canada provided that upon being dissolved, the community, including the wife’s reserved property, was partitioned equally between the spouses. If the wife renounced the partition, she kept only her reserved property.

[ 54 ] Alternatively, the spouses could enter into a marriage contract before a notary to establish a regime of separation of property. Under this regime, the spouses’ assets did not constitute a mass of community created during the marriage. Thus, for spouses who opted out of the regime of community of property, there was no partition upon dissolution; they therefore kept their respective patrimonies. In addition, the Civil Code of Lower Canada authorized a variety of stipulations that might change the scope of the regimes. The basic choice was nevertheless between separation of property and community of property.

[ 55 ] New trends in the implementation of matrimonial regimes began emerging in the 1930s. Wives increasingly adopted the conventional regime of separation of property even though the vast majority of them did not have paid employment outside the home through which they could accumulate property. In 1932, 43% of couples chose separation of property; this rose to 70% by 1970: A. Roy, Le contrat de mariage réinventé: Perspectives socio‑juridiques pour une réforme (2002), at pp. 58‑62; J. Pineau and D. Burman, Effets du mariage et régimes matrimoniaux (1984), at p. 123. As a result, in the event of separation from bed and board or, more rarely, of divorce, wives who had chosen the regime of separation of property, and who had probably not accumulated patrimonies because they did not work outside the home, were not entitled to partition of the property owned by their husbands.

[ 56 ] Several explanations have been advanced for this trend, at first glance surprising, in favour of the regime of separation of property during that period. They include rejection of the patriarchal nature of the regime of community of property, the incompatibility of this regime with women’s newly acquired legal capacity, the risk represented by the community of property in the event of bankruptcy and, more generally, the fact that an effective divorce procedure was not accessible.

[ 57 ] According to many commentators, this change could be explained as a reaction to the fact that wives had no power under the regime of community of property. Since the regime of separation of property gave wives some powers of administration over their patrimonies, it seemed to guarantee them more autonomy than that of community of property: see, inter alia , M. Tétrault, Droit de la famille (4th ed. 2010), vol. 1, at p. 562; D. Burman, “Politiques législatives québécoises dans l’aménagement des rapports pécuniaires entre époux: d’une justice bien pensée à un semblant de justice — un juste sujet de s’alarmer” (1988), 22 R.J.T. 149, at pp. 151‑52 and 155.

[ 58 ] Moreover, when the Act respecting the legal capacity of married women , S.Q. 1964, c. 66, came into force, wives obtained the legal capacity to freely dispose of their property and to perform the same acts in relation to it as a person of full age. The concept of “authority of the husband”, which was based on the obedience owed by wives to their husbands, also disappeared from the Civil Code at that time. [ translation ] “As a result, there was a clear dichotomy between the legal regime — community of property, which was administered by the husband — and this new legal capacity”: Lefebvre, at p. 12. The regime of community of property, which failed to reflect this new reality, was therefore abandoned in favour of that of separation of property.

[ 59 ] Another explanation is that, as pointed out by a number of notaries at the time, the regime of separation of property protected a wife if her husband went bankrupt after going into business: see, inter alia , Burman, at p. 151; Tétrault, Droit de la famille , at p. 562. It may therefore have seemed prudent in such situations to opt for the regime of separation rather than that of community of property.

[ 60 ] Finally, the Divorce Act , S.C. 1967‑68, c. 24, did not come into force in Canada until 1968. Before then, divorce was not accessible to the vast majority of the population, since it could be obtained only through a private Act. Some authors assume that many wives viewed separation or the possible dissolution of their union as an unrealistic prospect: see, inter alia , J. Jarry, Les conjoints de fait au Québec: vers un encadrement légal (2008), at p. 87. In such circumstances, the choice of a particular matrimonial regime could have seemed to be of no practical consequence.

[ 61 ] Whatever the reasons for the choice of the regime of separation of property and for its growing popularity, the fact remains that, in a context in which wives were not engaged in remunerative activities outside the matrimonial home, the consequences of this regime could be devastating in the event of separation or divorce. The effects of this choice of regime became clear after the Divorce Act came into force.

[ 62 ] To try to reverse this “separatist” trend that often caused serious problems for married women when their unions were dissolved, the Quebec legislature “modernized” the legal regime of community of property [ translation ] “to make it more attractive”: A. Roy, “Le régime juridique de l’union civile: entre symbolisme et anachronisme”, in Lafond and Lefebvre, 165, at p. 184; see also S. Massé, “Les régimes matrimoniaux au Canada — Analyse comparative des législations provinciales” (1985), 88 R. du N. 103, at p. 148. Thus, the legal matrimonial regime of partnership of acquests came into existence on July 1, 1970 with the coming into force of the Act respecting matrimonial regimes , S.Q. 1969, c. 77. From then on, the community consisted only of property acquired by the spouses during the marriage. Professor Tétrault explains this change as follows:

[ translation ] The change of legal regime from community of property to partnership of acquests is easily explained by the main characteristic of the regime of community of property, which concentrated the administration of the regime in the husband’s hands. This approach was difficult to reconcile with the goals of feminist movements and the granting of full legal capacity to women.

( Droit de la famille , at p. 511)

[ 63 ] Under the new legal regime, each spouse had the full administration of his or her property during the union, and most of the property acquired during the marriage could be partitioned upon dissolution. This regime distinguished two categories of property: acquests, which could be partitioned upon dissolution of the regime, and private property, of which the legislature provided an exhaustive list and which could not be partitioned. Acquests were property acquired during the marriage by either spouse, while private property was property owned before the marriage or property acquired after the marriage that was intrinsically personal, such as clothing or work tools.

[ 64 ] The creation of this legal matrimonial regime was accompanied by the repudiation of the principle of immutability of marriage agreements. Until 1970, once a marriage had been solemnized, the matrimonial regime could not be modified, even by mutual agreement of the spouses. Only one type of change was permitted by the legislature at that time: a wife to whom the regime of community of property applied could, subject to certain conditions, ask a court for permission to opt for the regime of separation of property: see E. Caparros, Les régimes matrimoniaux au Québec (3rd ed. 1988), at p. 97.

[ 65 ] The reason why spouses had been prohibited from amending their marriage agreements while married lay, in part, in the idea that only contracts entered into before marriage were entered into by independent persons capable of making the agreements they wished to make. Once the couple were married, since the wife fell legally under her husband’s power, her interests could not be validly defended if she entered into contracts with him. This was why spouses were prohibited from amending agreements entered into before marriage: see Roy, Le contrat de mariage réinventé , at pp. 99‑100; P.‑B. Mignault, Le droit civil canadien , vol. 6 (1902), at pp. 128‑29; L. Faribault, Traité de droit civil du Québec , vol. 10 (1952), at pp. 44‑45; R. Comtois, Traité théorique et pratique de la communauté de biens (1964), at p. 195.

[ 66 ] Like the rules under which the husband was responsible for the administration of the property of the community of property, the principle of immutability was hard to reconcile with the full legal capacity of married women and the end of the “authority of the husband”, according to which wives had been required to obey their husbands: see Roy, Le contrat de mariage réinventé , at p. 125. The creation of the regime of partnership of acquests was therefore accompanied by the introduction of a principle of mutability of matrimonial agreements, which allowed spouses to change their regime in whole or in part during their marriage: see Caparros, Les régimes matrimoniaux au Québec , at p. 97.

[ 67 ] However, as Professor Burman points out, [ translation ] “[a]t the time of the reform of matrimonial regimes, the new legal regime of partnership of acquests was given a very chilly reception, with spouses continuing to prefer that of separation of property” (p. 156). Thus, between 1971 and 1980, 48% of couples chose the legal regime of partnership of acquests, while 52% chose the regime of separation of property: Roy, Le contrat de mariage réinventé , at pp. 63‑64. However, this trend gradually reversed itself during the 1980s, with the result that fewer than 1% of couples chose the regime of separation as to property between 1995 and 2005: A. Roy, “Le contrat de mariage en droit québécois: un destin marqué du sceau du paradoxe” (2006), 51 McGill L.J. 665, at p. 668.

[ 68 ] Since the regime of partnership of acquests was neither retroactive nor mandatory, its introduction did not change the situation of wives who had chosen the regime of separation of property either before or after 1970. It should be borne in mind that, at the time, married women had not yet joined the labour market in large numbers. As the Committee on Matrimonial Regimes Committee of the Civil Code Revision Office pointed out in its report of May 20, 1968, “[i]t is still usual in Quebec households for the wife to devote all her time to the care of the family and for the husband to be the only one able to amass an estate by his work”: Report on Matrimonial Regimes (1968), at p. 9.

[ 69 ] In 1981, the legislature undertook a major new reform of family law. The Act to establish a new Civil Code and to reform family law , S.Q. 1980, c. 39 , established a primary regime of public order and provided for certain effects of marriage for spouses. This regime applied to all future marriages and to marriages already entered into regardless of the matrimonial regime that had previously been chosen. The reform introduced the principle that spouses had equal rights and obligations in marriage. This principle of equality was reflected in, among other things, the spouses’ obligation to take in hand the moral and material direction of the family together and to choose the family residence together. Other new measures were adopted to ensure adherence to the principle of joint direction by requiring the consent of both spouses for certain acts, such as alienation of the family residence by the spouse who owned it. To the obligations arising out of marriage before the reform, namely fidelity, cohabitation, assistance and succour, the legislature added an obligation for each spouse to contribute toward the expenses of the marriage in proportion to his or her means, including through activities within the home.

[ 70 ] The legislature also dealt directly with the situation of women who had married under the regime of separation of property rather than that of community of property or partnership of acquests during the preceding decades. To remedy their vulnerability, it created the compensatory allowance mechanism, which entitled each spouse to claim compensation for his or her contribution, in property or services, to the enrichment of the other spouse’s patrimony. Payment of such compensation could be ordered by a court in the course of proceedings leading to the spouses’ separation. However, this measure proved ineffective after a few years, as the Minister of Justice and the Minister for the Status of Women explained in 1988:

[ translation ] As for the compensatory allowance, it has not proved effective enough to fully remedy the problems experienced by certain married spouses, particularly those who have chosen separation of property as their matrimonial regime. Thus, according to the majority of the jurisprudence, the work performed by a spouse within the home does not entitle that spouse to a compensatory allowance if the spouse was merely fulfilling his or her obligation to contribute towards the expenses of the marriage; in asserting his or her right, the co‑operating spouse encounters major evidentiary difficulties that, in some cases, render the remedy illusory . . . .

(H. Marx and M. Gagnon Tremblay, Les droits économiques des conjoints (1988), document tabled for consultation, at p. 10.)

[ 71 ] In 1989, because the compensatory allowance had not had the intended effect, the Quebec National Assembly passed the Act to amend the Civil Code of Québec and other legislation in order to favour economic equality between spouses , S.Q. 1989, c. 55, which introduced the concept of family patrimony into the Civil Code . As a result, “a marriage has the immediate effect . . . of establishing a family patrimony, and it creates a claim that can be asserted upon separation from bed and board or upon dissolution of the marriage”: M.T. v. J.‑Y.T. , 2008 SCC 50 , [2008] 2 S.C.R. 781 , at para. 14 (emphasis added) . Like all the other effects of marriage, such as the obligation to provide assistance and succour, the family patrimony was of public order and applied regardless of the legal or conventional matrimonial regime chosen by the parties to govern their patrimonial relationship.

[ 72 ] The claim so created provided a basis for equal partition of the net value of certain property, such as the family’s residences, the household furniture used by the family, the vehicles used by the family and rights under retirement plans, regardless of which spouse had a right of ownership in that property. Residences, furniture and vehicles were included in the family patrimony regardless of whether they were acquired before or during the marriage, but in the case of rights under a retirement plan, only those accrued during the marriage were part of that patrimony.

[ 73 ] The transitional provisions enacted by the legislature indicated that the articles concerning the family patrimony applied to spouses who had married before the Act came into force. However, the legislature allowed such spouses to opt out of the family patrimony provisions by mutual agreement. They had 18 months after the Act came into force to make that choice and record it in an agreement.

[ 74 ] There were three objectives underlying this further reform of the rights and obligations of spouses vis‑à‑vis one another. The legislature wanted to remedy the problems encountered by women who had married under the regime of separation of property, make up for the ineffectiveness of the compensatory allowance and redefine marriage.

[ 75 ] First, as Baudouin J.A. of the Quebec Court of Appeal explained in a decision rendered shortly after the creation of the family patrimony,

[ translation ] [b]y introducing into our law the partition of the family patrimony ( An Act to amend the Civil Code of Québec and other legislation in order to favour economic equality between spouses [S.Q. 1989, c. 55]), the legislature intended to remedy injustices that could be suffered by a certain category of married women and to recognize the value of work done in the home. Upon divorce or separation from bed and board, women who were married under the regime of separation of property were sometimes severely disadvantaged from an economic standpoint when the time came to liquidate the patrimony accumulated while the spouses were living together.

Through Bill 146, the legislature wanted to correct the sometimes perverse effects of choosing the regime of separation of property . . . .

Indeed, these injustices had sometimes been pointed out by this Court [ Droit de la famille — 67 , [1985] C.A. 135 ], which had always found that, in the absence of specific provisions on this point, the courts had no power to change the freely chosen matrimonial regime to alleviate such inequities.

( Droit de la famille — 977 , [1991] R.J.Q. 904, at pp. 907‑8; see also Tétrault, Droit de la famille , at pp. 152‑53; Roy, “Le contrat de mariage en droit québécois: un destin marqué du sceau du paradoxe”, at pp. 668‑69; Roy, “Le régime juridique de l’union civile: entre symbolisme et anachronisme”, at p. 180; E. Caparros, “Le patrimoine familial: une qualification difficile” (1994), 25 R.G.D. 251, at p. 253 .)

[ 76 ] As well, spouses who had previously chosen the regime of separation of property now had a claim on the value of the family patrimony. The legislature was, in a sense, giving wives an opportunity to change the impact of the choice of matrimonial regime they had made in the past by making the regime of separation of property inapplicable to a large portion of the family patrimony, which now became subject to a form of partition. However, this change was not irrevocable, since a wife could choose to renounce her rights in the family patrimony upon the dissolution of their union (art. 423 C.C.Q. ).

[ 77 ] Second, “[t]hat Act represented a partial response to the disappointment and difficulties that had resulted from the implementation of the compensatory allowance in the years prior to its enactment”: M.T. v. J.‑Y.T. , at para. 17 ; see also L. Langevin, “Liberté de choix et protection juridique des conjoints de fait en cas de rupture: difficile exercice de jonglerie” (2009), 54 McGill L.J. 697, at p. 714; Lefebvre, at p. 17.

[ 78 ] Third, the legislature redefined marriage by means of the Act. From the time the Act came into force, marriage became not only a union of persons, but also an egalitarian economic union with a number of patrimonial consequences. In M.T. v. J.‑Y.T. , I elaborated on the basis for the legislature’s objective in introducing the family patrimony:

Marriage represents, first and foremost, a union of persons. However, the legislature also wanted it to be a partial economic union or an association of interests (D. Burman and J. Pineau, Le “patrimoine familial” (projet de loi 146) (1991), No. 31). The adoption of the partnership of acquests as the suppletive matrimonial regime that is to apply unless the spouses make another choice shows that this is what the legislature intended. The creation of the family patrimony confirms that intention even more clearly.

Marriage results in the establishment of a form of economic union to which both spouses must contribute as best they can (Kasirer, at p. 572). Article 396 C.C.Q. clearly imposes on the spouses a legal obligation to contribute toward the expenses of the marriage “in proportion to their respective means”. It also provides that “[t]he spouses may make their respective contributions by their activities within the home.” The law is not really concerned with the size or nature of the contributions, and in fact presumes them to be equal ( Droit de la famille — 1893 , [1993] R.J.Q. 2806 (C.A.), at p. 2809). [Emphasis added; paras. 21‑22.]

[ 79 ] As a result of this reform, spouses who decided to marry were required to accept [ translation ] “a partnership model” (B. Moore, “Culture et droit de la famille: de l’institution à l’autonomie individuelle” (2009), 54 McGill L.J. 257, at p. 268) that involved a willingness to partition: Droit de la famille — 977 , at p. 908. It can therefore be concluded that [ translation ] “[i]n 1989, [the legislature] transformed marriage into a primarily economic partnership by creating a family patrimony”: M. Tétrault, “L’union civile: j’me marie, j’me marie pas”, in Lafond and Lefebvre, 101, at p. 111.

[ 80 ] In Quebec, marriage thus became not only a union, but also, “as this Court held in Moge (at p. 870), . . . a ‘joint endeavour’, a socio‑economic partnership”: Bracklow v. Bracklow , [1999] 1 S.C.R. 420 , at para. 49 . From the time of the legislative reform in which the family patrimony was created, any person who chose to marry was deemed to want to create this socio‑economic partnership on the basis of a number of provisions of public order that established the effects of marriage, such as those governing acts involving the family residence and the spouses’ proportional contributions to the expenses of the marriage. In return for this obligation to form an economic partnership, the new definition of marriage provided for mechanisms of public order to apportion the patrimonial consequences of dissolution of the partnership, such as the partition of the family patrimony and the awarding of support following the breakdown of the marriage. Conversely, persons who did not wish to be subject to these effects or to create a joint endeavour or economic union with a partially predetermined content could choose to remain in a de facto union outside marriage.

[ 81 ] In sum, whereas the legislature’s first two objectives in introducing the family patrimony related to the existing economic context, its third objective was decidedly forward‑looking, as it proposed a new definition of marriage, which now included an economic union or partnership.

(b) Legal Framework for Marriage

[ 82 ] As a result of the reforms outlined above, marriage is now subject to a legal framework that governs the mutual relationships of spouses. This framework is made up of a primary regime and a legal or conventional matrimonial regime the effects of which are felt both during the marriage and when it ends. However, before looking at the effects of each of these regimes during and after marriage, I note that, aside from the death of a spouse, marriage can end as a result of separation from bed and board under the Civil Code of Québec or divorce under the Divorce Act , R.S.C. 1985, c. 3 (2nd Supp.).

(i) Primary Regime

[ 83 ] The Civil Code of Québec establishes a primary regime in a chapter that defines the fundamental effects of marriage. The provisions setting out these effects are mandatory, since the spouses may not derogate from or renounce them in a marriage contract:

391. In no case may spouses derogate from the provisions of this chapter, whatever their matrimonial regime.

[ 84 ] The primary regime thus governs certain aspects of the spouses’ relationship with one another and creates mutual rights, duties and obligations for the spouses. During the marriage, the spouses “owe each other respect, fidelity, succour and assistance” and “are bound to live together” (art. 392, paras. 2 and 3 C.C.Q. ). “[They] together take in hand the moral and material direction of the family” (art. 394 C.C.Q. ), they “choose the family residence together” (art. 395, para. 1 C.C.Q. ) and they “contribute towards the expenses of the marriage in proportion to their respective means” (art. 396, para. 1 C.C.Q. ). And arts. 401 to 408 C.C.Q. limit the exercise of each spouse’s right of ownership in the family residence and the movable property serving for the use of the household during the marriage. Certain acts of alienation, hypothec and lease may not be performed by one spouse without the other spouse’s consent. In addition, as we have seen, “[m]arriage entails the establishment of a family patrimony consisting of certain property of the spouses regardless of which of them holds a right of ownership in that property” (art. 414 C.C.Q. ).

[ 85 ] Some of these duties are moral or extrapatrimonial in nature. For example, the duty of assistance concerns a person’s obligation to support his or her spouse through affection, help, care and devotion: M. D.‑Castelli and D. Goubau, Le droit de la famille au Québec (5th ed. 2005), at pp. 98‑99; J. Pineau and M. Pratte, La famille (2006), at p. 132; Tétrault, Droit de la famille , at pp. 134‑35; J.‑P. Senécal, Droit de la famille québécois , vol. 1 (loose‑leaf), ¶ 11‑615.

[ 86 ] The duty of succour is economic in nature and involves providing the other spouse with the resources or support required for subsistence on the basis of his or her needs: Senécal, vol. 1, ¶ 11‑625; Pineau and Pratte, at pp. 132‑33 and 156‑57; D.‑Castelli and Goubau, at p. 99. More specifically, it relates to the spouses’ obligation of support to one another: arts. 392 and 585 to 596.1 C.C.Q.

[ 87 ] The duty of succour lasts until the marriage is dissolved. Since separation from bed and board — although it loosens the marital bond by releasing the spouses from the obligation to live together — does not terminate the marriage, it does not terminate the other effects of marriage, including the duty of succour: art. 507 C.C.Q. This explains why a court granting a separation from bed and board may order either spouse to pay support to the other: art. 511 C.C.Q. Since the duty of succour is based solely on the respective needs of the spouses, an award of support is never automatic but depends on each spouse’s needs and ability to pay: arts. 512 and 587 C.C.Q.

[ 88 ] Unlike separation from bed and board, divorce dissolves marriage: art. 516 C.C.Q. This is why [ translation ] “in the context of a divorce . . . the right to support is no longer based on the duty of succour, which is an effect of marriage, but derives from ss. 15.2 and 17 [of the Divorce Act ]”: Senécal, vol. 2, ¶ 65‑770; see also Pineau and Pratte, at pp. 132‑33; Bracklow , at paras. 20‑21 . At this stage, a support award will depend on the condition, means, needs and other circumstances of each spouse, including the length of time the spouses cohabited and the functions performed by each spouse during cohabitation.

[ 89 ] In the event of separation from bed and board or of divorce, the primary regime has consequences other than a possible support award. First, the value of the family patrimony is divided equally between the spouses: art. 416 C.C.Q. Second, a compensatory allowance may be awarded by a court on the application of either spouse: art. 427 C.C.Q. Third, where applicable, a court may award to either spouse the lease of the family residence (art. 409 C.C.Q. ) and the ownership or use of certain movable property (art. 410 para. 1 C.C.Q. ), or, to a spouse who has custody of the children, the use of the family residence (art. 410, para. 2 C.C.Q. ) .

(ii) Matrimonial Regime

[ 90 ] After deciding to marry and thus to be subject to the primary regime, spouses must choose a matrimonial regime to govern the rest of their financial relationship. They are free to select the regime they consider the most appropriate. As the Quebec Court of Appeal has pointed out, [ translation ] “the matrimonial regime concept [is] based on the spouses’ freedom of choice of regime”: G.B. v. C.C. , [2001] R.J.Q. 1435 , at para. 22 (emphasis in original) . Spouses are also free to change their matrimonial regime by mutual agreement during their marriage: art. 438 C.C.Q.

[ 91 ] The Civil Code provides that spouses who, before the solemnization of their marriage, have not fixed their matrimonial regime in a marriage contract are subject by default to the legal regime of partnership of acquests, under which [ translation ] “each spouse has the administration and free disposal of all his or her property, both private property and acquests, while subject to the regime but is required to partition the value of his or her acquests equally with the other spouse on the date of separation or dissolution”: Roy, “Le régime juridique de l’union civile: entre symbolisme et anachronisme”, at p. 183; arts. 432, 461 and 467 C.C.Q.

[ 92 ] Alternatively, spouses who decide to enter into a marriage contract are free to organize their patrimonial relationship through “[a]ny kind of stipulation . . . subject to the imperative provisions of law and public order”: art. 431 C.C.Q.

[ 93 ] Spouses may also prefer the regime of conventional separation as to property to that of partnership of acquests, and can establish it by including a declaration to this effect in their marriage contract: art. 485 C.C.Q. Under this regime, the spouses, individually, have the administration, enjoyment and free disposal of all their property both during marriage and upon its breakdown: art. 486 C.C.Q. No distinction is drawn between private property and acquests, and no partition occurs at the time of separation or divorce, except in the case of property held in co‑ownership and property over which neither spouse is able to establish an exclusive right of ownership. The latter type of property is presumed to be held by both spouses in undivided co‑ownership: art. 487 C.C.Q.

(2) Civil Union Spouses

[ 94 ] On June 7, 2002, the Quebec legislature passed the Act instituting civil unions and establishing new rules of filiation , S.Q. 2002, c. 6. This Act provided same‑sex couples, in particular, with a first mechanism for making their unions official, namely the civil union, which would also be available to opposite‑sex couples. As Professor Tétrault notes, [ translation ] “[a]t the time of its enactment, the Quebec legislature was responding to the fact that Parliament did not recognize the right of persons of the same sex to marry”: Droit de la famille , at p. 571. That recognition would come three years later when Parliament enacted the Civil Marriage Act , S.C. 2005, c. 33.

[ 95 ] A civil union is defined as “a commitment by two persons 18 years of age or over who express their free and enlightened consent to live together and to uphold the rights and obligations that derive from that status”: art. 521.1, para. 1 C.C.Q. Unlike marriage, which must be dissolved judicially, a civil union may be dissolved by way of a notarized joint declaration, provided that the spouses consent to the dissolution and settle all the consequences of the dissolution in an agreement: art. 521.13 C.C.Q. However, “[i]n the absence of a joint declaration dissolving the civil union executed before a notary or where the interests of the common children of the spouses are at stake, the dissolution of the union must be pronounced by the court”: art. 521.17, para. 1 C.C.Q . It has also been possible, since the coming into force of the Act to amend the Civil Code as regards marriage , S.Q. 2004, c. 23, s. 7, for spouses to dissolve their civil union by getting married. In such a case, the effects of the civil union are maintained and are considered to be effects of the marriage from the date of the civil union, and the spouses’ civil union regime becomes their marriage regime: art. 521.12, para. 2 C.C.Q.

[ 96 ] Article 521.6 C.C.Q. provides that civil union has significant mandatory effects and refers, for certain principles, to the Civil Code ’s provisions on the effects of marriage:

521.6. The spouses in a civil union have the same rights and obligations.

They owe each other respect, fidelity, succour and assistance.

They are bound to live together.

The effects of the civil union as regards the direction of the family, the exercise of parental authority, contribution towards expenses, the family residence, the family patrimony and the compensatory allowance are the same as the effects of marriage, with the necessary modifications.

Whatever their civil union regime, the spouses may not derogate from the provisions of this article.

[ 97 ] While in their union, civil union spouses are subject to the duty of succour and therefore owe each other support. The duty of succour ends with the dissolution of the union. However, on application by either spouse, a court may, upon or after pronouncing the dissolution, order one of the spouses to pay support to the other: art. 521.17, para. 3 C.C.Q.

[ 98 ] I note in passing that no factors specific to civil union spouses for determining the amount of support are set out in the Civil Code . The courts must therefore refer to the general principles stated in art. 587 C.C.Q. : see, inter alia , Senécal, vol. 2, ¶ 65‑815; Tétrault, Droit de la famille , at p. 587.

[ 99 ] In addition to being subject on a mandatory basis to this primary regime, civil union spouses have a legal regime that is identical to the matrimonial regime of partnership of acquests that applies to married spouses. Should they wish to do so, persons planning to enter into a civil union can also opt out of this legal regime by way of a contract prepared before the solemnization of their union and adopt a regime of separation as to property:

521.8. A civil union regime may be created by and any kind of stipulation may be made in a civil union contract, subject to the imperative provisions of law and public order.

Spouses who, before the solemnization of their civil union, have not so fixed their civil union regime are subject to the regime of partnership of acquests.

Civil union regimes, whether legal or conventional, and civil union contracts are subject to the same rules as are applicable to matrimonial regimes and marriage contracts, with the necessary modifications.

(3) De Facto Spouses

(a) Historical Review of the Situation of De Facto Spouses Under the Civil Code of Lower Canada Before the 1980 Reforms

[ 100 ] It should be noted at the outset that, until the family law reform and the coming into force of part of the draft Civil Code of Québec in 1981, the legislative treatment of the de facto union, then referred to as “concubinage”, was negative. As several authors have pointed out, the de facto union, as an [ translation ] “obstacle to family stability and peace”, was considered to be suspect and “contrary to public order and good morals”, and had an “unfavourable” if not “immoral” character that the government could not encourage: see, inter alia , E. Deleury and M. Cano, “Le concubinage au Québec et dans l’ensemble du Canada: Deux systèmes juridiques, deux approches”, in J. Rubellin‑Devichi, ed., Des concubinages dans le monde (1990), 85, at p. 88; A. Cossette, “Le concubinage au Québec” (1985), 88 R. du N. 42, at pp. 45 and 53; Tétrault, “L’union civile: j’me marie, j’me marie pas”, at p. 127; Lefebvre, at p. 11; Tétrault, Droit de la famille , at pp. 839‑49.

[ 101 ] This disapproval of the de facto union was expressed in legislation in two ways: one concerned the relationships of de facto spouses, the other the treatment of the children born of such unions. With regard to the former, art. 768 C.C.L.C. limited “[g]ifts inter vivos made in favor of the person with whom the donor has lived in concubinage . . . to maintenance”. In this way, the legislature prohibited de facto spouses from organizing their patrimonial relationships. The law denied them the possibility of establishing a legal framework for their cohabitation and limited their freedom of contract. Notary Jean Sylvestre described the impact of this prohibition as follows:

[ translation ] Agreements between concubinaries were, for all practical purposes, unthinkable until [1981].

By prohibiting all gifts inter vivos between concubinaries and persons who had lived in concubinage, art. 768 C.C.L.C. closed the door on all financial arrangements between persons in a de facto union that were gratuitous or in the nature of a liberality or gift.

Since one or more of these characteristics was almost always present in what were known as agreements between spouses or concubinaries, it was impossible to consider making such agreements between concubinaries.

(“Les accords entre concubins”, [1981] C.P. du N. 195, at paras. 1‑3)

[ 102 ] Moreover, the children of de facto spouses, or “natural” children, were denied a number of rights granted to “legitimate” children, those whose parents were married: see J.‑L. Baudouin, “Examen critique de la situation juridique de l’enfant naturel” (1966), 12 McGill L.J. 157, at p. 158. Historically, unless they were legitimated by their parents’ getting married, natural children could not inherit from their parents unless the latter had provided for them in a will. Nor could they claim from their parents performance of the obligations of maintenance and education that married spouses had to their legitimate children. The [ translation ] “ Civil Code of Lower Canada [thus] distinguished legitimate, natural, adulterine and incestuous children”: Moore, at p. 266. Professor Jean‑Louis Baudouin, as he then was, explained the rationale for this distinction:

[ translation ] The legal reasons given to justify ignoring the natural family group can be seen clearly upon reading our Code . . .: a desire to protect the rights of legitimate families, a refusal to condone conduct contrary to good morals, a refusal to encourage the proliferation of de facto unions, etc. [pp. 157‑58]

[ 103 ] When it enacted the Act to establish a new Civil Code and to reform family law , the legislature moved away from this negative, even hostile, view of the de facto union. The prohibition against gifts inter vivos was removed by repealing art. 768 C.C.L.C. De facto spouses became free to organize their relationships with one another through legally valid and binding agreements. From that time on, it was open to them [ translation ] “to enter into agreements freely [in what was] virtually the equivalent of a matrimonial regime for them”: Sylvestre, at para. 8. According to Professor Benoît Moore, this change confirmed the disappearance of the legislature’s hostility toward the de facto union: p. 267.

[ 104 ] The legislature also eliminated the distinctions between legitimate, natural, adulterine and incestuous children. It thus established the principle that children are equal regardless of the circumstances of their birth and the nature of their filiation. Article 522 C.C.Q. now codifies the principle that all children whose filiation is established have the same rights and obligations. The rules on parental authority (arts. 597 C.C.Q. et seq. ), the obligation of support (art. 585 C.C.Q. ) and intestate succession (art. 655 C.C.Q. ) therefore apply to all children. In addition, art. 604 C.C.Q. provides that, “[i]n the case of difficulties relating to the exercise of parental authority, the person having parental authority may refer the matter to the court, which will decide in the interest of the child after fostering the conciliation of the parties”. Thus, when de facto spouses cease living together, a court can rule on child custody and access in the same way as it might in a case involving the separation of married spouses. According to D. Goubau, G. Otis and D. Robitaille, [ translation ] “the Civil Code [thus] reflects the fact that marriage is no longer required as a framework for the family”: “La spécificité patrimoniale de l’union de fait: le libre choix et ses ‘dommages collatéraux’” (2003), 44 C. de D. 3, at p. 13.

(b) Legislative Policy Respecting the De Facto Union Following the 1980 Reform

[ 105 ] In the 1980 reform, the Quebec legislature established the rule of freedom of contract for de facto spouses in organizing their relationships with one another. At the same time, as I mentioned above, it redefined the mandatory content of marriage by introducing a primary regime that no spouse could opt out of. Whether the de facto union should be redefined by imposing a similar mandatory legislative framework on it was also discussed in the Quebec National Assembly at that time.

[ 106 ] One proposal that had already been made by the Civil Code Revision Office was to require de facto spouses to contribute proportionately toward household expenses: Report on the Québec Civil Code (1978), vol. II — Commentaries , t. 1, at pp. 113 and 206; see also Lefebvre, at p. 18. With this in mind, the Office proposed art. 338 in its draft Civil Code :

338. De facto consorts owe each other support as long as they live together.

However, if exceptional circumstances justify it, the court may order a de facto consort to pay support to his spouse once they no longer live together.

( Report on the Québec Civil Code (1978), vol. I — Draft Civil Code , at p. 119)

[ 107 ] However, Quebec’s Conseil du statut de la femme, an independent agency created by the legislature in 1973 to advise the government on all matters relating to the status of women, was against imposing a mandatory legislative framework to govern relationships between de facto spouses. Criticizing the proposals of the Civil Code Revision Office on the basis that they limited the freedom of de facto spouses, the Conseil stated that

[ translation ] . . . this approach violates the animating principle of de facto spouses, namely freedom of choice. To respect the will of the parties involved, the [Conseil] recommends that no obligations result from a de facto union.

Our position on the de facto union is based on true recognition of the equality and autonomy of individuals. This is why we consider it essential to emphasize the non‑institutionalization of this type of union and to respect the will of the parties involved.

( Mémoire présenté à la Commission parlementaire sur la réforme du droit de la famille (1979), at pp. 23‑24; see also Mémoire du Conseil du statut de la femme présenté lors de la consultation générale sur les droits économiques des conjoints (1988), at p. 40.)

[ 108 ] At the end of the parliamentary debate, the then Minister of Justice rejected the recommendation that a legislative framework be imposed on de facto unions. Instead, he decided to preserve the freedom of individuals to choose a form of union whose content was not predetermined. He stated the following:

[ translation ] Another case requiring a concrete application of the principle that individuals should be free to choose how to organize their family unit is that of the de facto union. Most of the briefs submitted to the parliamentary justice committee on family law reform in March 1979 asked the legislature to respect the desire of unmarried couples to distinguish their choice of lifestyle from marriage. We therefore considered it appropriate not to interfere with this freely chosen lifestyle; there is thus no need to institutionalize or regulate it.

Moreover, to fully respect this option, it seemed reasonable to put de facto spouses on the same footing as other individuals by proposing the abolition of the restrictions that continue to apply to them today under article 768 of the Civil Code , which limits their right to make gifts to each other.

(National Assembly, Journal des débats , vol. 23, No. 15, 6th Sess., 31st Leg., December 4, 1980, at p. 608 (second reading of Bill 89). This position was supported by the then leader of the official Opposition, who stated the following at p. 663: [ translation ] “Regarding the de facto union, therefore, I think we should proceed very carefully. If people do not want to give their union a legal or statutory status themselves, such a status cannot be forced on them either.”)

[ 109 ] The legislature would debate but reaffirm this position several times during the successive reforms that have modified Quebec family law since 1980. On the occasion of each reform, the legislature reiterated its choice not to regulate the private relationships of de facto spouses on the basis that their individual autonomy and freedom should be respected: see, for example, the remarks of the Minister for the Status of Women at the time of the creation of the family patrimony (National Assembly, Journal des débats , vol. 30, No. 125, 2nd Sess., 33rd Leg., June 8, 1989, at p. 6487), of the Minister of Justice two years later in the context of the reform of the Civil Code (National Assembly, Subcommittee on Institutions, Journal des débats , No. 22, 1st Sess., 34th Leg., November 19, 1991, at p. 859), of the Minister of Justice at the time of the recognition of same‑sex de facto spouses in social legislation (National Assembly, Journal des débats , No. 197, 2nd Sess., 35th Leg., June 18, 1998, at pp. 12069‑70), and of the Minister of Justice at the time of the establishment of the civil union (National Assembly, Standing Committee on Institutions, Journal des débats , vol. 37, No. 46, 2nd Sess., 36th Leg., February 12, 2002, at pp. 4‑5).

[ 110 ] At the time of the establishment of the civil union in 2002, the Minister of Justice clearly stated that Quebec law now [ translation ] “recognizes three types of conjugality: that of married spouses, that of civil union spouses and that of de facto spouses”. He also confirmed that the bill instituting the civil union [ translation ] “does not propose any amendments that affect the conditions under which de facto spouses live together. They remain free to establish the terms and conditions governing their relationships”: National Assembly, Journal des débats , vol. 37, No. 96, 2nd Sess., 36th Leg., May 7, 2002, at p. 5816.

(c) Relationships Between De Facto Spouses: A Regime of Freedom of Contract

[ 111 ] As we have seen, married and civil union spouses are, both during the marriage or union and upon its breakdown, subject to the mandatory application of the primary regime and the suppletive application of the regime of partnership of acquests. The situation is very different for de facto spouses.

[ 112 ] As I mentioned above, the Civil Code of Québec does not lay down the terms of the union of de facto spouses. The law imposes no duty of assistance and succour on them, and thus no obligation of support. The sharing of household expenses is left to their discretion; they are not required to contribute toward those expenses in proportion to their respective means. Nor are they required to choose the family residence together. No mandatory provisions apply to limit the exercise of their rights of ownership in the family residence. A de facto spouse who is the sole owner of the residence can therefore sell it or lease it without the other spouse’s consent. A de facto union does not create a family patrimony, is not subject to the legal matrimonial regime of partnership of acquests and does not entitle a spouse to a compensatory allowance.

[ 113 ] A de facto spouse continues, both while living with the other spouse and after their relationship breaks down, to own any property he or she acquired before or during their union. Any change in this situation must be consented to by the spouse whose rights are affected. Thus, under the Act respecting the Québec Pension Plan , an application for the partition of pensionable earnings accrued during the period in which de facto spouses lived together requires the consent of both of them. Similarly, the Supplemental Pension Plans Act , R.S.Q., c. R‑15.1, which establishes the legal framework for private pension plans, permits former de facto spouses to partition benefits accumulated under a member’s pension plan only if both of them consent.

[ 114 ] Since the de facto union is not subject to the mandatory legislative framework that applies to marriage and the civil union, de facto spouses are free to shape their relationships as they wish, having proper regard for public order. They can enter into agreements to organize their patrimonial relationships while they live together and to provide for the consequences of a possible breakdown: on the possible content of such agreements, see Tétrault, “L’union civile: j’me marie, j’me marie pas”, at pp. 133‑34; Tétrault, Droit de la famille , at pp. 870‑71; A. Roy, “La charte de vie commune ou l’émergence d’une pratique réflexive du contrat conjugal” (2007), 41 R.J.T. 399. Such agreements are commonly referred to as “cohabitation agreements”.

[ 115 ] As some authors have noted, in light of the Quebec jurisprudence, [ translation ] “there is no longer any doubt that contracts between de facto spouses are valid”: Jarry, at p. 134. The Quebec courts have held that de facto spouses can validly enter into contracts that provide that the rules on partition of the family patrimony will apply should their relationships break down ( Couture v. Gagnon , [2001] R.J.Q. 2047 (C.A.) , leave to appeal refused, [2002] 3 S.C.R. vii), contracts that contemplate an obligation to pay spousal support should cohabitation cease ( Ponton v. Dubé , 2005 QCCA 413 ), and contracts that grant a right to exclusive use of the family residence after separation ( Bourbonnais v. Pratt , 2006 QCCS 5611 , [2007] R.D.F. 124 ). According to the Quebec Court of Appeal, [ translation ] “[s]uch agreements can even provide for the equivalent of a compensatory allowance”: M.B. v. L.L. , [2003] R.D.F. 539 , at para. 30 .

[ 116 ] In the absence of such agreements, the general law applies to any patrimonial dispute that results from the end of the spouses’ cohabitation. Since each of the de facto spouses continues to own any property he or she acquired individually before or while they lived together, a spouse who proves his or her sole ownership of movable property can revendicate it. Where property is owned in undivided co‑ownership, either spouse can force the other to proceed to the partition and licitation of the undivided property, since no one is bound to remain in indivision (art. 1030 C.C.Q. ).

[ 117 ] Finally, de facto spouses who believe that they were wronged when their union broke down can bring an action based on unjust enrichment, the rules for which have been codified in arts. 1493 to 1496 C.C.Q. since 1994. That unjust enrichment is applicable to relationships between de facto spouses was confirmed by this Court in Peter v. Beblow , [1993] 1 S.C.R. 980 . The principle of unjust enrichment must be interpreted cautiously but generously, in a manner consistent with the conditions originally established by the Court in Cie Immobilière Viger Ltée v. Lauréat Giguère Inc. , [1977] 2 S.C.R. 67 , which are now set out in art. 1493 C.C.Q. As Dalphond J.A. of the Quebec Court of Appeal explained, it should be used [ translation ] “solely to compensate one party for a contribution, in property or services, that enabled the other party to be in a better position than he or she would have been in had the parties not lived together, in short, that enriched the other party”: M.B. v. L.L. , at para. 39 .

[ 118 ] To obtain compensation under arts. 1493 C.C.Q. et seq. , a de facto spouse alleging unjust enrichment must therefore prove on a balance of probabilities that the following conditions are met: an enrichment, an impoverishment, a correlation between the enrichment and the impoverishment, and the absence of justification, of evasion of the law and of any other remedy: Cie Immobilière Viger ; Peter v. Beblow ; M.B. v. L.L. , at para. 34 ; J.‑L. Baudouin and P.‑G. Jobin, Les obligations (6th ed. 2005), by P.‑G. Jobin with N. Vézina, at paras. 566 et seq.

[ 119 ] As this Court held in Peter v. Beblow , a de facto spouse can benefit from certain rebuttable presumptions that make it easier to discharge his or her burden of proof. For example, in the case of a long‑term de facto union, a court can presume, on the one hand, that there is a correlation between the enrichment of one spouse and the impoverishment of the other and, on the other hand, that there was no reason for the enrichment: Peter v. Beblow , at pp. 1013 and 1018; M.B. v. L.L. , at para. 37 ; Benzina v. Le , 2008 QCCA 803, at para. 7 ; Barrette v. Falardeau , 2010 QCCA 989, at paras. 26‑27 . Finally, when these conditions are met, the de facto spouse’s action will be allowed in the lesser of the following two amounts: that of the enrichment of his or her spouse and that of his or her own impoverishment ( Cie Immobilière Viger , at p. 77).

[ 120 ] In a recent decision written by Dalphond J.A. with the concurrence of Côté J.A., the Court of Appeal reiterated the principles of the doctrine of unjust enrichment with respect to de facto spouses and the importance of the presumptions in the plaintiff’s favour: C.L. v. J.Le. , 2010 QCCA 2370, at paras. 10‑15 , quoted in full in Droit de la famille — 121120 , 2012 QCCA 909, at para. 65 . The Court of Appeal correctly noted that a court hearing [ translation ] “a claim by a de facto spouse for compensation for unjust enrichment [must] undertake a broad, overall analysis of the parties’ circumstances , taking into account all contributions made by the spouses during the time they lived together”: C.L. v. J.Le. , at para. 12 (emphasis added). As well, rather than a simple accounting exercise, [ translation ] “the analysis of the factual and legal aspects requires a particular flexibility adapted to the nature of relationships between spouses ( Lacroix v. Valois , [1990] 2 S.C.R. 1259 , at p. 1279) ”: C.L. v. J.Le , at para. 13 (emphasis added).

(d) Treatment of De Facto Spouses in the Same Way as Married and Civil Union Spouses in Various Statutes

[ 121 ] For the dealings that couples have with third parties, and more particularly with the government, there are a number of social statutes in which a distinction is no longer drawn between marriage, civil union and de facto union: see, inter alia , Workers’ Compensation Act , R.S.Q., c. A‑3 ; An Act respecting industrial accidents and occupational diseases , R.S.Q., c. A‑3.001; An Act respecting financial assistance for education expenses , R.S.Q., c. A‑13.3; An Act respecting legal aid and the provision of certain other legal services , R.S.Q., c. A‑14; Automobile Insurance Act , R.S.Q., c. A‑25; An Act respecting insurance ; R.S.Q., c. A‑32; An Act respecting financial services cooperatives , R.S.Q., c. C‑67.3; An Act respecting trust companies and savings companies , R.S.Q., c. S‑29.01; An Act respecting school elections , R.S.Q., c. E‑2.3; An Act respecting duties on transfers of immovables , R.S.Q., c. D‑15.1; Cooperatives Act , R.S.Q., c. C‑67.2; Taxation Act , R.S.Q., c. I-3; An Act respecting the Québec sales tax , R.S.Q., c. T‑0.1; An Act respecting labour standards , R.S.Q., c. N‑1.1; Courts of Justice Act , R.S.Q., c. T‑16; An Act respecting the Québec Pension Plan ; An Act respecting the Government and Public Employees Retirement Plan , R.S.Q., c. R‑10; An Act respecting the Civil Service Superannuation Plan , R.S.Q., c. R‑12; Supplemental Pension Plans Act ; An Act respecting the conditions of employment and the pension plan of the Members of the National Assembly , R.S.Q., c. C‑52.1; An Act respecting the Pension Plan of Certain Teachers , R.S.Q., c. R‑9.1; An Act respecting the Pension Plan of Peace Officers in Correctional Services , R.S.Q., c. R‑9.2; An Act respecting the Pension Plan of Elected Municipal Officers , R.S.Q., c. R‑9.3; An Act respecting the Teachers Pension Plan , R.S.Q., c. R‑11; Individual and Family Assistance Act , R.S.Q., c. A‑13.1.1.

[ 122 ] As Professors D.‑Castelli and Goubau explain, [ translation ] “[p]rovided that a union meets the specific conditions of a statute, the situation of de facto spouses is exactly the same as that of married or civil union spouses under Quebec civil law for the purposes of that statute”: p. 173. In this regard, the Quebec Court of Appeal has stated that [ translation ] “a review of these statutes . . . shows that the legislature intended to treat de facto spouses and married spouses on an equal basis”: Couture v. Gagnon , at para. 32 .

[ 123 ] It should be noted that, since the 1999 enactment of the Act to amend various legislative provisions concerning de facto spouses , S.Q. 1999, c. 14, the de facto spouse concept has been extended in these statutes to include same‑sex de facto spouses. Moreover, as a result of the enactment in 2002 of the Act instituting civil unions and establishing new rules of filiation , s. 143, the definition of spouse in the Interpretation Act , R.S.Q., c. I‑16, s. 61.1, includes same‑sex spouses.

[ 124 ] This tendency in social statutes to treat de facto spouses in the same way as married and civil union spouses can also be seen in certain civil law matters involving interaction between de facto spouses and third parties. For example, a de facto spouse, like a married or civil union spouse, can consent to care required by his or her spouse who is incapable of giving consent (art. 15 C.C.Q. ); where his or her spouse alone signed a lease for a dwelling and subsequently leaves the dwelling or dies, maintain occupancy and become the lessee (art. 1938 C.C.Q. ); repossess a dwelling in an immovable owned jointly with his or her spouse (art. 1958 C.C.Q. ); participate in meetings of relatives, persons connected by marriage or a civil union and friends during the process of appointing a curator for his or her spouse (art. 266 C.C.Q. ); avoid the seizure of part of his or her salary if supporting his or her spouse ( art. 553 of the Code of Civil Procedure , R.S.Q., c. C‑25); and bring a direct action in liability where his or her spouse dies (abolition of the rule in art. 1056 C.C.L.C. ).

D. Demographic and Sociological Evolution of the De Facto Union in Quebec Since 1980

[ 125 ] Parallel to the evolution of family law in Quebec, it is interesting to see how the social position of the de facto union in that province has changed. On the basis of data collected by Statistics Canada and the Institut de la statistique du Québec that are studied in some of the expert reports submitted by the parties, some demographic trends with respect to the de facto union can be observed. First, over the last 30 years, the development of the de facto union has taken place largely in Quebec. In Canada, it is in that province that the de facto union has spread most widely. Between 1981 and 2006, the proportion of Quebec couples in de facto unions increased from 7.9% to 34.6%. By comparison, 18.4% of couples in Canada as a whole were in such a relationship in 2006. Also, 60% of Quebec children are now apparently being born out of wedlock. The 2011 census confirms these trends. According to the now available data, 20% of Canadian couples live in de facto unions. In Quebec, the proportion rises to 37.8%.

[ 126 ] The statistics analyzed by the experts suggest that de facto unions are more common among young Quebeckers under the age of 35 than among Canadians in that age group as a whole. Young Quebeckers opt for de facto unions in much higher proportions (51%) than do young Canadians as a whole (29%). In Quebec, and to a lesser extent in Canada as a whole, de facto unions remain proportionately more common among couples in their third unions than among couples in their first.

[ 127 ] Whereas the purely demographic aspect of the de facto union is easily quantified, the same is not true of its sociological or socio‑demographic character. In the instant case, four of the eight expert reports filed by the parties at trial relate specifically to this sociological perspective. Among other things, these experts discuss how conjugality is depicted in Quebec, the differences and similarities between marriage and the de facto union, and how Quebec couples organize themselves. Because of the methodological limitations of the studies, their small sample sizes and the general lack of quantitative data on which to base hypotheses, it seems impossible in the context of this case to draw any definitive conclusions about several aspects of this social phenomenon, including why this type of relationship is chosen and how it functions from an economic standpoint.

[ 128 ] It will be helpful here to quote some remarks made by Professors Le Bourdais and Lapierre‑Adamcyk, at pp. 3 and 4 of their report, on the state of the research on de facto unions in Quebec:

[ translation ] Very little research based on large samples representative of the population as a whole has been conducted into the meaning of the de facto union in relation to marriage and the meaning of the commitment made by the spouses to one another. Moreover, the research that has been done has not gone into detail about the mechanisms underlying the choice of type of union for unions formed recently, that is, since the mid‑1990s. Many unknowns remain, and since the circumstances are changing, relationships that seem to be well established for a given period may no longer be valid for a subsequent period. . . .

[With regard to certain links observed among various facts], the reliable data needed to analyze them correctly are often unavailable. Thus, the link between financial independence and the choice of type of union continues to be the subject of speculation, since prospective longitudinal data following individuals as they advance through their lives, which would make it possible to observe their financial situations at the times they have to make choices, are only rarely or partially available.

Furthermore, in areas like money management by couples or the depictions of conjugality, which have been addressed qualitatively in the studies examined here, quantitative data on large samples are sorely lacking in Quebec and in Canada, making it impossible to compare couples in de facto unions with married couples or to take account of all relevant groups regardless of their stage in the life cycle (at the beginning of the relationship, after several years of living together or the arrival of a child, or immediately after separating). Research on other societies, where it exists, may suggest some elements of an answer, but it does not suffice to shed a satisfactory light on the Canadian reality. [Joint Record, vol. 14, at pp. 7‑8]

[ 129 ] Finally, as the trial judge pointed out, none of the expert reports submitted by the parties discusses the concrete effects of the breakdown of a relationship for de facto spouses or for married or civil union spouses.

E. Scope of A’s Claim

[ 130 ] Before considering the nature and application of the right to equality in the context of this appeal, I must review the scope of A’s claim. All its effects must be understood in order to assess their impact on the application of the right to equality.

[ 131 ] First, by arguing that arts. 401 to 430, 432, 433, 448 to 484 and 585 C.C.Q. are contrary to s. 15(1) of the Charter and not justified under s. 1 , A is claiming the benefit of certain aspects of the primary regime that applies in cases of separation from bed and board, divorce, or dissolution of a civil union. On this basis, she is seeking to partition the family patrimony and reserve her right to claim a compensatory allowance. She also submits that she is entitled to support for herself and to a lump sum.

[ 132 ] Although A is challenging the constitutionality of the provisions protecting the family residence, she makes no specific claim concerning those provisions in this case. The issues relating to the place of residence of A and the parties’ children appear to have been dealt with in an agreement the parties entered into before the trial.

[ 133 ] Second, along with the application of certain parts of the primary regime, A is seeking the automatic and mandatory application of the legal matrimonial regime of partnership of acquests. As I mentioned above, in the case of marriage and civil union, the legal regime applies automatically only if the spouses do not enter into a marriage contract or a civil union contract. Under Quebec civil law, the regime of partnership of acquests is suppletive and applies only if there is no contract designating a regime of separation as to property or containing stipulations tailored to the spouses’ personal circumstances. In essence, as regards matrimonial regimes, A wants her de facto union to be treated not only like a marriage, but like a marriage entered into without a marriage contract. She is therefore asking this Court to recognize the retroactive creation of a partnership of acquests that applies on a mandatory basis to her relationship with B.

[ 134 ] When a marriage breaks down, the dissolution of a legal regime of partnership of acquests results in partition of the value of the acquests between the spouses to whom the regime applies. Partition is of relatively little consequence to couples whose main assets are part of the family patrimony, but it can acquire great significance where, for example, one of the spouses is actively involved in business ventures. Having made this clarification, I will now turn to s. 15(1) of the Charter , which is at the heart of this case. How it is interpreted and applied will determine the outcome of A’s constitutional claims and, as a result, the scope of her support and patrimonial rights in relation to B.

F. Development of the Jurisprudence on Section 15

(1) Underlying Values of Section 15: Equality, Dignity, Freedom and Personal Autonomy

[ 135 ] Section 15(1) of the Charter provides that “[e]very 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.” To understand the meaning and content of this guarantee as well as its impact on government action, it is necessary to clearly understand the nature of what McIntyre J. called “the broad range of values embraced by s. 15”: Andrews v. Law Society of British Columbia , [1989] 1 S.C.R. 143 , at p. 171 . The recognition of these values aids in the interpretation of s. 15(1) and in the overall assessment of the merits of claims that the right to equality has been violated: Law , at para. 54; Walsh , at para. 63 .

[ 136 ] In Andrews , the first case in which this Court considered the application of s. 15, McIntyre J. stated emphatically that the equality guarantee is intended to ensure the recognition of the equal worth of all human beings in Canadian society:

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 at law as human beings equally deserving of concern, respect and consideration. [p. 171]

(See also R. Dworkin, Taking Rights Seriously (1977), at pp. 272‑73.)

[ 137 ] According to McIntyre J., the establishment of formal equality does not suffice to meet the objectives underlying the adoption of the equality guarantee. Section 15 instead introduces a concept of substantive equality.

[ 138 ] As this Court has pointed out on several occasions, this value of substantive equality at the heart of s. 15 is closely tied to the concept of human dignity: Miron , at paras. 145‑46 ; Law , at paras. 52 and 54; Blencoe v. British Columbia (Human Rights Commission) , 2000 SCC 44 , [2000] 2 S.C.R. 307 , at para. 77 ; Gosselin v . Quebec (Attorney General) , 2002 SCC 84 , [2002] 4 S.C.R. 429 , at para. 20 . The innate and equal dignity of every individual is invariably an “essential value underlying the s. 15 equality guarantee”: Kapp , at para. 21 . Indeed, the Court has said that “the purpose of s. 15(1) is to prevent the violation of essential human dignity and freedom” ( Law , at para. 51) and to eliminate any possibility of a person being treated in substance as “less worthy” than others: Gosselin , at para. 22 . In other words:

This principle recognizes the dignity of each human being and each person’s freedom to develop his body and spirit as he or she desires, subject to such limitations as may be justified by the interests of the community as a whole. It recognizes that society is based on individuals who are different from each other, and that a free and democratic society must accommodate and respect these differences.

( Miron , at para. 145 )

[ 139 ] The principle of personal autonomy or self‑determination, to which self‑worth, self‑confidence and self‑respect are tied, is an integral part of the values of dignity and freedom that underlie the equality guarantee: Law , at para. 53; Gosselin , at para. 65 . Safeguarding personal autonomy implies the recognition of each individual’s right to make decisions regarding his or her own person, to control his or her bodily integrity and to pursue his or her own conception of a full and rewarding life free from government interference with fundamental personal choices: R. v. Big M Drug Mart Ltd. , [1985] 1 S.C.R. 295 , at p. 346, per Dickson J; R. v. Morgentaler , [1988] 1 S.C.R. 30 , at p. 164, per Wilson J.; Rodriguez v. British Columbia (Attorney General) , [1993] 3 S.C.R. 519 , at p. 554, per Lamer C.J., at pp. 587‑88, per Sopinka J.; Blencoe , at para. 77 , per Bastarache J.

[ 140 ] In the application of s. 15, promotion of the values of equality, dignity, freedom and autonomy requires “the remedying of discriminatory treatment” based on the personal characteristics enumerated in s. 15(1) or characteristics analogous to them: Law , at para. 52. Under the Charter , it is unfair to limit an individual’s full participation in society solely because the individual has one of these personal characteristics: Miron , at para. 146 ; Eldridge v. British Columbia (Attorney General) , [1997] 3 S.C.R. 624 , at paras. 72‑73 . Likewise, it is unacceptable to refuse on the basis of these characteristics to treat a person as a full member of society who deserves to realize his or her full human potential: Miron , at para. 146 ; Gosselin , at para. 23 .

[ 141 ] Since Andrews , this Court has endeavoured to identify types of situations involving discrimination contrary to s. 15(1) and to establish an analytical framework to help the courts deal with claims for application of the equality guarantee. Having a framework that can be used to identify discriminatory distinctions is of fundamental importance in light of the fact that the drawing of distinctions lies at the heart of legislative action: Andrews , at pp. 168‑69. Virtually all legislation distinguishes and makes categories. Without a coherent analytical framework, every distinction involving an enumerated or analogous ground of discrimination would be suspect and would need to be justified, even those relating to fundamental and essential measures, such as the prohibition against driving while intoxicated: Andrews , at pp. 181‑82. Such a framework makes it possible for a court to combat discrimination effectively without succumbing to the temptation to redefine legislative solutions that do not violate the s. 15 equality guarantee: Hodge v. Canada (Minister of Human Resources Development) , 2004 SCC 65 , [2004] 3 S.C.R. 357 , at para. 26 . Since the analytical framework developed by this Court has been discussed, reformulated and enriched many times over the past two decades, I must summarize the changes made to it in order to better describe its current form.

(2) Creation of an Analytical Framework

(a) Andrews

[ 142 ] In Andrews , the Court was considering an alleged infringement of s. 15(1) by a legislative provision under which Canadian citizenship was required for admission to the British Columbia Bar. The Court concluded that the provision was contrary to s. 15(1) because it barred a class of persons from certain types of employment solely on the ground that they were not Canadian citizens, and without consideration of their qualifications or of their individual attributes or merits. Although he dissented on the application of s. 1 to the case, McIntyre J. laid the foundations of the analytical framework that is indispensable when considering an alleged violation of s. 15(1) .

[ 143 ] McIntyre J. acknowledged at the outset that not every differentiation in the treatment of individuals under the law will result in inequality, “[n]or will a law necessarily be bad because it makes distinctions”: Andrews , at p. 167; see also pp. 164, 168 and 182. He thus refused to conclude that the Charter prohibits any form of inequality between humans. Rather, it is a tool for combating certain forms of inequality that are considered discriminatory. On the basis of s. 15, which states that the right to equality must apply “without discrimination”, McIntyre J. identified the existence of discrimination as the fundamental requirement for application of the protection afforded by s. 15(1) . Therefore, according to him, s. 15(1) is deemed to limit or qualify itself, since it condemns only distinctions that are “discriminatory”:

A complainant under s. 15(1) must show not only that he or she is not receiving equal treatment before and under the law or that the law has a differential impact on him or her in the protection or benefit accorded by law but, in addition, must show that the legislative impact of the law is discriminatory. [p. 182]

(See also Egan v. Canada , [1995] 2 S.C.R. 513, at paras. 33‑34 , per L’Heureux‑Dubé J., dissenting.)

[ 144 ] McIntyre J. then stated that discrimination under s. 15(1) is based on the grounds enumerated in that provision or grounds analogous to them. Determining whether a ground of distinction is an analogous ground and can therefore support a discrimination claim involves a contextual inquiry to determine whether it gives rise to questions of stereotyping, historical disadvantage or prejudice: Andrews , at pp. 180‑81; see also Corbiere v. Canada (Minister of Indian and Northern Affairs) , [1999] 2 S.C.R. 203 , at paras. 12‑13 . It can also be asked whether the complainant is a member of a discrete and insular minority, many of which will be disadvantaged, like some of the groups covered by the grounds enumerated in s. 15(1) . Thus, discrimination “epitomizes the worst effects of the denial of equality”: Andrews , at p. 172.

[ 145 ] McIntyre J. added that s. 15(1) limits prohibited distinctions or differentiations in treatment to those which involve prejudice or disadvantage. In Andrews , the members of the Court agreed that the legislative provision challenged by the complainant infringed s. 15 because it imposed a specific burden on non‑citizens. According to McIntyre J., “[n]on‑citizens, lawfully permanent residents of Canada, are . . . a good example of a ‘discrete and insular minority’ who come within the protection of s. 15”: Andrews , at p. 183. The Court was divided only on the application of s. 1, and not on this finding of a discriminatory infringement of the right to equality.

(b) From Andrews to Law

[ 146 ] As several authors have noted, McIntyre J.’s reasons left some room for uncertainty about what constitutes discriminatory treatment under s. 15: see, inter alia , P. W. Hogg, Constitutional Law of Canada (5th ed. Supp. (loose-leaf)), vol. 2, at p. 55‑26; L. B. Tremblay, “Promoting Equality and Combating Discrimination Through Affirmative Action: The Same Challenge? Questioning the Canadian Substantive Equality Paradigm” (2012), 60 Am. J. Comp. L. 181, at p. 185; B. J. Cameron, “A Work in Progress: The Supreme Court and the Charter ’s Equation of Rights and Limits”, in D. M. McAllister and A. M. Dodek, eds., The Charter at Twenty: Law and Practice 2002 (2002), 31, at p. 34; C. D. Bredt and A. M. Dodek, “Breaking the Law ’s Grip on Equality: A New Paradigm for Section 15” (2003), 20 S.C.L.R. (2d) 33 , at p. 56 ; D. G. Réaume, “Discrimination and Dignity” (2003), 63 La. L. Rev. 645, at pp. 652‑53.

[ 147 ] Taken out of context, some passages from the decision might suggest that every adverse distinction based on an enumerated or analogous ground is a form of discrimination prohibited by s. 15. However, McIntyre J. stated elsewhere in his reasons that “[d]istinctions based on personal characteristics attributed to an individual solely on the basis of association with a group will rarely escape the charge of discrimination, while those based on an individual’s merits and capacities will rarely be so classed”: Andrews , at pp. 174‑75 (emphasis added). He added that discrimination cannot be identified solely by determining whether the alleged ground is one of those enumerated in s. 15(1) or an analogous ground thereto: Andrews , at p. 182. McIntyre J. therefore seemed to leave open the question of how to identify situations in which an adverse distinction based on an enumerated or analogous ground is not discriminatory.

[ 148 ] The Court gave two separate answers in what has been called “the 1995 trilogy” of Miron , Egan and Thibaudeau v . Canada , [1995] 2 S.C.R. 627 . According to Lamer C.J. and La Forest, Gonthier and Major JJ., adverse differential treatment is discriminatory only if the alleged enumerated or analogous ground is irrelevant to the legislative goals or the values underlying the impugned provision. If the ground for excluding a group of persons from certain benefits arising from a law is relevant to the goals and values of that law, the exclusion will therefore not be discriminatory within the meaning of s. 15.

[ 149 ] According to Sopinka, Cory, McLachlin and Iacobucci JJ., and to some extent to L’Heureux‑Dubé J., differential treatment that disadvantages the complainant will be discriminatory only if it conflicts with the purpose of s. 15. They explained that the purpose of s. 15 is “to prevent the violation of human dignity and freedom through the imposition of limitations, disadvantages or burdens through the stereotypical application of presumed group characteristics, rather than on the basis of merit, capacity or circumstance”: Miron , at para. 140 , per McLachlin J. In other words, an adverse distinction based on an enumerated or analogous ground will be discriminatory only if it is “contrary to s. 15’s aim of protecting human dignity”: Egan , at para. 180 , per Cory and Iacobucci JJ. (dissenting). According to McLachlin J., requiring that a provision be found to violate the purpose of s. 15(1) before it can be characterized as discriminatory averts a trivialization of s. 15: Miron , at para. 131 .

(c) Synthesis in Law

[ 150 ] In 1999, Iacobucci J., writing for a unanimous Court, reaffirmed the substance of the approach taken by Cory J. in Egan and McLachlin J. in Miron . According to Iacobucci J., the Court had recognized “that the existence of a conflict between an impugned law and the purpose of s. 15(1) is essential in order to found a discrimination claim”: Law , at para. 41. In his view, “the goal of [ s. 15(1) is to assure] human dignity by the remedying of discriminatory treatment”: Law , at para. 52. He adopted an analytical approach based on a concept of substantive equality that will be violated only where adverse differential treatment by the government has a negative effect on the claimant’s human dignity. On this basis, Law proposed a three‑stage analytical framework to be used by the courts in ruling on a discrimination claim under s. 15(1) .

[ 151 ] First, the court must determine whether the law treats the claimant differently than others either in purpose or effect. To do so, the court must ask whether the impugned law draws a formal distinction between the claimant and others on the basis of one or more personal characteristics, or whether it fails to take into account the claimant’s already disadvantaged position within Canadian society, thereby giving rise to substantively differential treatment between the claimant and others on the basis of one or more personal characteristics. Bastarache J., discussing the nature of such differential treatment in a subsequent case, stated that “it is clear that Iacobucci J. [was] talking only about a detrimental purpose or effect, since it is nonsensical to think that a claimant might establish that a beneficial or benign purpose or effect infringes s. 15(1) ”: Gosselin , at para. 243 (emphasis in original).

[ 152 ] Second, the court must determine whether one or more of the grounds enumerated in s. 15(1), or analogous grounds, are the basis for the differential treatment.

[ 153 ] Third, the court must then determine whether the law in question has a purpose or effect that is discriminatory within the meaning of the equality guarantee by making the following inquiry:

Does the differential treatment discriminate, by imposing a burden upon or withholding a benefit from the claimant in a manner which reflects the stereotypical application of presumed group or personal characteristics, or which otherwise has the effect of perpetuating or promoting the view that the individual is less capable or worthy of recognition or value as a human being or as a member of Canadian society, equally deserving of concern, respect, and consideration?

( Law , at para. 88(3)(C))

[ 154 ] To resolve the third issue and thus determine whether the differential treatment discriminates in a substantive sense and brings the purpose of s. 15(1) into play, the court must undertake a full contextual inquiry concerning the circumstances of the claimant’s claim. This inquiry must be undertaken from the point of view of a reasonable person in circumstances similar to those of the claimant who takes the relevant context into account. Whereas the claimant must prove on a balance of probabilities that the impugned provision discriminates in a substantive sense, the court can take judicial notice of certain facts or matters but must be careful not to use judicial notice to recognize social phenomena that may not truly exist.

[ 155 ] Iacobucci J. identified four factors that are relevant to a contextual analysis of a discrimination claim. According to him, these factors may overlap, are not exhaustive and do not all have to be assessed in every case involving an alleged violation of s. 15(1) . They can nevertheless guide the court in determining whether the law in question discriminates in a substantive sense.

[ 156 ] The first contextual factor he identified is pre‑existing disadvantage experienced by the claimant or the group of which the claimant is a member. This factor suggests an inquiry into whether the individual or group might have experienced a historical disadvantage, vulnerability, stereotyping or prejudice. According to Iacobucci J., “[i]t is logical to conclude that, in most cases, further differential treatment will contribute to the perpetuation or promotion of their unfair social characterization, and will have a more severe impact upon [these individuals], since they are already vulnerable”: Law , at para. 63. However, the existence of a pre‑existing disadvantage does not give rise to a presumption that the adverse treatment of historically disadvantaged persons is necessarily discriminatory: Law , at para. 67.

[ 157 ] The second factor is the correspondence, or lack thereof, between the ground or grounds on which the discrimination claim is based and the actual needs, capacity or circumstances of the claimant or the affected group. Acceptance of the need for correspondence as a contextual factor is logical, since it will be more difficult to prove that a person’s dignity has been violated if the legislation in question takes into account the claimant’s actual circumstances, needs or capacity. In a subsequent decision, McLachlin C.J., writing for a majority of the Court, clarified the nature of this factor as follows:

The fact that some people may fall through a program’s cracks does not show that the law fails to consider the overall needs and circumstances of the group of individuals affected, or that distinctions contained in the law amount to discrimination in the substantive sense intended by s. 15(1).

( Gosselin , at para. 55 )

[ 158 ] The third factor is whether the impugned law has an ameliorative purpose or effect for certain members of society. Thus, a law that draws distinctions in order to alleviate certain inequalities affecting disadvantaged groups is less likely to be found to violate the dignity of more fortunate individuals to whom the measures do not apply.

[ 159 ] The fourth factor concerns the nature or scope of the benefit or interest which the claimant feels he or she has been denied. As Iacobucci J. explained, “[t]he more severe and localized the consequences of the legislation for the affected group, the more likely that the differential treatment responsible for these consequences is discriminatory within the meaning of s. 15(1) ”: Law , at para. 88(9)(D). Such consequences arise where the distinction restricts access to a fundamental social institution or impedes full membership in Canadian society.

(d) Application of the Equality Guarantee in Kapp and Withler

[ 160 ] Nearly 10 years after Law , the Court, in reasons written by McLachlin C.J. and Abella J., reviewed the synthesis proposed by Iacobucci J.: see Kapp . Almost three years after Kapp , the Court continued that review in Withler v. Canada (Attorney General) , 2011 SCC 12 , [2011] 1 S.C.R. 396 , a unanimous decision written, once again, by McLachlin C.J. and Abella J.

[ 161 ] In those two decisions, the Court observed that, despite the changes made to the s. 15 analysis over the years, the concept of substantive equality had remained central to the analytical framework for that provision: Kapp , at para. 15 . The Court also noted that, although the analytical framework adopted in Andrews had been enriched since that case, including by Iacobucci J. in Law , it had never been abandoned: Kapp , at para. 14 . The Court added that the purpose of s. 15(1) is “the elimination from the law of measures that impose or perpetuate substantial inequality”: Withler , at para. 40 .

[ 162 ] In Kapp , the Court reworked the three‑stage analytical framework from Law in light of the purpose of s. 15, namely to promote substantive equality, reshaping it into a two‑part test for showing discrimination under s. 15(1) . Where a violation of s. 15(1) is alleged, a court must ask the following questions: “(1) Does the law create a distinction based on an enumerated or analogous ground? (2) Does the distinction create a disadvantage by perpetuating prejudice or stereotyping?” ( Kapp , at para. 17 ). If the answer to each of these questions is yes, it can be concluded that the impugned legislative provision violates the equality guarantee in s. 15(1). The Court stated that this two‑part test was, “in substance, the same” as the test from Law and that Law had confirmed the approach to substantive equality set out in Andrews : Kapp , at paras. 17 and 24 .

[ 163 ] As the Court has acknowledged, the review of the analytical framework for s. 15(1) that began in Kapp and continued in Withler was also a response to criticism of the framework proposed in Law and the subsequent application of that framework, and in particular of the use of the violation of human dignity test, of the application of the contextual factors and of the comparison required by the framework.

[ 164 ] The Court began by stressing the importance of the value of human dignity. It stated that “[t]here can be no doubt that human dignity is an essential value underlying the s. 15 equality guarantee”: Kapp , at para. 21 . However, it noted that several difficulties had arisen from attempts to employ this concept as a legal test as proposed in Law . On the one hand, human dignity is a value that underlies all Charter rights: see R. v. Oakes , [1986] 1 S.C.R. 103 , at p. 136 . In this sense, its relevance is not limited to the equality guarantee: see, inter alia , D. Greschner, “The Purpose of Canadian Equality Rights” (2002), 6 Rev. Const. Stud. 291; D. Proulx, “Le concept de dignité et son usage en contexte de discrimination: deux Chartes, deux modèles”, [2003] R. du B. (numéro spécial) 485. On the other hand, as I mentioned above, dignity is not the only value underlying s. 15. Other values associated with it include freedom and personal autonomy.

[ 165 ] Rather than emphasizing the identification of a violation of the complainant’s dignity as an independent factor, the Court proposed that the analysis to determine whether a claimant has been discriminated against be focused on the context of the claim. At that time, the Court reaffirmed the importance of context and the relevance of the contextual factors set out in Law . Pre‑existing disadvantage and the nature of the affected interest are therefore factors to be applied to determine whether a distinction creates a disadvantage by perpetuating prejudice. Correspondence between the ground or grounds on which the claim is based and the actual circumstances of the claimant or the affected group is a factor to be applied to determine whether a distinction creates a disadvantage by stereotyping: Kapp , at para. 23 . Next, the Court attributed two main functions to the ameliorative factor referred to in Law . First, the ameliorative purpose or effect of a law may bring it within the purview of s. 15(2), which preserves the right of governments to implement specific programs aimed at helping disadvantaged groups without fear of challenge under s. 15(1) : Kapp , at para. 16 . Second, the Court explained alternatively that, “[w]here the impugned law is part of a larger benefits scheme . . . the ameliorative effect of the law on others and the multiplicity of interests it attempts to balance will also colour the discrimination analysis [under s. 15(1)]”: Withler , at para. 38 .

[ 166 ] According to the Court, regardless of whether the s. 15 analysis focuses on perpetuating disadvantage or stereotyping,

the analysis involves looking at the circumstances of members of the group and the negative impact of the law on them. The analysis is contextual, not formalistic, grounded in the actual situation of the group and the potential of the impugned law to worsen their situation.

( Withler , at para. 37 )

The Court thus noted that the contextual factors that are relevant at this stage of the analysis will vary with the nature of the case. “Just as there will be cases where each and every factor need not be canvassed, so too will there be cases where factors not contemplated in Law will be pertinent to the analysis”: Withler , at para. 66 . In both Kapp and Withler , the Court warned against an inflexible application of these factors that does not take into account their concrete effects in their larger social, political and legal contexts.

[ 167 ] The Court also recognized that it may be helpful at the stage of determining whether a distinction exists to compare the group of which the claimant is a member with other groups. However, it observed that a formalistic or artificial approach to comparison may prevent a court from adequately addressing the issue raised at the second stage of the analysis, namely whether the law has a purpose or effect that discriminates in a substantive sense: Withler , at paras. 62‑65 . Thus, it appears that a

mirror comparator group analysis may fail to capture substantive inequality, may become a search for sameness, may shortcut the second stage of the substantive equality analysis, and may be difficult to apply. In all these ways, such an approach may fail to identify — and, indeed, thwart the identification of — the discrimination at which s. 15 is aimed.

( Withler , at para. 60 )

[ 168 ] At this second stage, therefore, rather than limiting its analysis to a formalistic comparison of particular groups, the Court must endeavour to take the contextual factors relevant to the case into account: Kapp , at paras. 22‑23 . Although comparison may bolster the understanding of the context, “[t]he probative value of comparative evidence, viewed in [a] contextual sense, will depend on the circumstances”: Withler , at para. 65 .

[ 169 ] The Court thus acknowledged the general usefulness of comparison in determining whether a distinction exists and gaining a better contextual understanding of the claimant’s place within the legislative scheme at issue and within society. The Court nonetheless made the use of the comparative approach more flexible by emphasizing the need to assess the impact of the impugned scheme on substantive equality. In so doing, it moved away from the rigid comparative analytical approach based on the identification of comparator groups that had been adopted in some of its decisions: see, inter alia , Hodge , at para. 17 . Once a distinction is found to exist, therefore, the main question must always be the same: does the distinction create a disadvantage by perpetuating prejudice or stereotyping? In other words, if there is a distinction, is it discriminatory? The Court therefore stressed the importance of the factors of perpetuation of prejudice and stereotyping. While it did not make them the only factors, it determined that they were crucial to the identification of discrimination and to the application of the analytical framework for s. 15.

(e) Meaning and Scope of Kapp and Withler

[ 170 ] In Kapp and Withler , the Court reworked and provided important clarifications to the analytical framework for applying the s. 15(1) equality guarantee. However, some authors argue that those decisions did not eliminate all uncertainty concerning the concepts of disadvantage, prejudice and stereotyping in this framework: S. Moreau, “ R. v. Kapp : New Directions for Section 15” (2008‑2009), 40 Ottawa L. Rev. 283, at pp. 286 and 291‑92; J. Koshan and J. W. Hamilton, “Meaningless Mantra: Substantive Equality after Withler ” (2011), 16 Rev. Const. Stud. 31, at pp. 48‑51; Tremblay, at p. 188.

[ 171 ] In Kapp , the Court reiterated the fundamental principle that “[s]ection 15(1) is aimed at preventing discriminatory distinctions that impact adversely on members of groups identified by the grounds enumerated in s. 15 and analogous grounds”: para. 16; Alberta v. Hutterian Brethren of Wilson Colony , 2009 SCC 37 , [2009] 2 S.C.R. 567 , at para. 106 . The language used by the Court is clear: s. 15(1) prevents governments from establishing “ discriminatory distinctions that impact adversely” on members of groups on the basis of enumerated or analogous grounds (para. 106 (emphasis added)). Section 15 applies not only to laws enacted with discriminatory intent, but also, even if there is no such intent, to laws with discriminatory effects. Section 15(1) therefore does not prohibit distinctions that have an adverse impact unless they are discriminatory. In other words, the adverse impact or “disadvantage” must be discriminatory. What is a discriminatory disadvantage? As can be seen from the analytical framework as articulated in Kapp , a discriminatory disadvantage is as a general rule one that perpetuates prejudice or that stereotypes:

(1) Does the law create a distinction based on an enumerated or analogous ground?

(2) Does the distinction create a disadvantage by perpetuating prejudice or stereotyping ? [Emphasis added; para. 17.]

[ 172 ] In Ermineskin Indian Band and Nation v. Canada , 2009 SCC 9 , [2009] 1 S.C.R. 222 , Rothstein J. applied this analytical framework to dismiss a claim that the right to equality had been infringed in the context of the relationship between Aboriginal communities and the federal government. Rothstein J. summed up the Court’s position as follows:

This Court’s equality jurisprudence makes it clear that not all distinctions are discriminatory. Differential treatment of different groups is not in and of itself a violation of s. 15(1) . As this Court stated in Andrews v. Law Society of British Columbia , [1989] 1 S.C.R. 143, at p. 182 (restated in R. v. Kapp , 2008 SCC 41 , [2008] 2 S.C.R. 483, at para. 28 ), a complainant must show “not only that he or she is not receiving equal treatment before and under the law or that the law has a differential impact on him or her in the protection or benefit accorded by law but, in addition, must show that the legislative impact of the law is discriminatory ” (emphasis added). The analysis, as established in Andrews , consists of two questions: first, does the law create a distinction based on an enumerated or analogous ground; and second, does the distinction create a disadvantage by perpetuating prejudice or stereotyping. [para. 188.]

[ 173 ] This analytical framework has been reaffirmed by the Court in other recent decisions: Hutterian , at para. 106; A.C. v. Manitoba (Director of Child and Family Services) , 2009 SCC 30 , [2009] 2 S.C.R. 181 , at paras. 109 and 150 . The Court further explained the nature of the framework in Withler , in which it stated that there are usually two ways for a claimant to prove that a law containing an adverse distinction based on an enumerated or analogous ground “discriminates in a substantive sense”. On the one hand, the claimant can show that the impugned law perpetuates prejudice against members of a group. On the other hand, the claimant can show that the disadvantage imposed by the law is based on a stereotype. If either of these things is shown, the impugned law will be found to violate s. 15(1) : Withler , at paras. 34‑36 . I would add that there will no doubt be cases in which prejudice and stereotyping are both involved, and reinforce one another.

[ 174 ] It was stated in Kapp and reiterated in Withler that a discriminatory distinction is an adverse distinction that perpetuates prejudice or that stereotypes. I will now discuss the uncertainty that results from certain passages from Kapp and Withler , after which I will consider the meaning and scope of the concepts of disadvantage, prejudice and stereotyping.

[ 175 ] I note that in certain passages from Kapp and Withler , the Court used varying terminology that strayed from the language used to describe the recommended test. In particular, some of these passages might suggest that a violation of the right to equality can be established simply by proof of a disadvantage based on an enumerated or analogous ground, without having to establish that the disadvantage is discriminatory by showing that it results from the perpetuation of prejudice or from stereotyping: see, inter alia , Kapp , at para. 25 ; Withler , at paras. 35, 37, 65 and 71 .

[ 176 ] It would be wrong to ascribe such a meaning to the passages in question. The words “discriminates by perpetuating disadvantage or prejudice” in para. 71 of Withler (emphasis added) actually refer only to the importance of pre‑existing disadvantage as a contextual factor for the purpose of identifying prejudice in the sense of circumstances in which certain individuals are not recognized at law as human beings fully deserving of concern, respect and consideration: Andrews , at p. 171. In Law , Iacobucci J. explained the relevance of this link between disadvantage and prejudice as follows:

As has been consistently recognized throughout this Court’s jurisprudence, probably the most compelling factor favouring a conclusion that differential treatment imposed by legislation is truly discriminatory will be, where it exists, pre‑existing disadvantage, vulnerability, stereotyping, or prejudice experienced by the individual or group: see, e.g., Andrews , supra , at pp. 151‑53, per Wilson J., p. 183, per McIntyre J., pp. 195‑97, per La Forest J.; Turpin , supra , at pp. 1331‑33; Swain , supra , at p. 992, per Lamer C.J.; Miron , supra , at paras. 147‑48 , per McLachlin J.; Eaton , supra , at para. 66 . These factors are relevant because, to the extent that the claimant is already subject to unfair circumstances or treatment in society by virtue of personal characteristics or circumstances, persons like him or her have often not been given equal concern, respect, and consideration. It is logical to conclude that, in most cases, further differential treatment will contribute to the perpetuation or promotion of their unfair social characterization, and will have a more severe impact upon them, since they are already vulnerable. [para. 63]

[ 177 ] The existence of a pre‑existing or historical disadvantage will thus make it easier to prove prejudice or a stereotype. However, the existence or perpetuation of a disadvantage cannot in itself make a distinction discriminatory. The following comment by McLachlin C.J. from a recent case more clearly summarizes the need for a link between disadvantage and prejudice, on the one hand, and between disadvantage and stereotype, on the other:

Laws and government acts that perpetuate disadvantage and prejudice, or that single out individuals or groups for adverse treatment on the basis of stereotypes, violate s. 15(1) and are invalid, subject to justification under s. 1 of the Charter : Kapp ; Withler v. Canada (Attorney General) , 2011 SCC 12 , [2011] 1 S.C.R. 396. [Emphasis added.]

( Alberta (Aboriginal Affairs and Northern Development) v. Cunningham , 2011 SCC 37 , [2011] 2 S.C.R. 670, at para. 39 )

[ 178 ] The interpretation according to which there cannot as a rule be substantive discrimination, despite the existence of a disadvantage, without prejudice or stereotyping is consistent with this Court’s case law on the meaning and application of s. 15(1) . As McLachlin C.J. has stated, “not every adverse distinction made on the basis of an enumerated or analogous ground constitutes discrimination”: Gosselin , at para. 21 . More recently, Rothstein J. stated that, “even if [there] is a disadvantage, the legislation will violate s. 15(1) only if that disadvantage is one that is discriminatory, that is, if it perpetuates prejudice or stereotyping”: Ermineskin , at para. 192.

[ 179 ] This position is also consistent with the approach taken by this Court to substantive equality, the promotion of which “entails the promotion of a society in which all are secure in the knowledge that they are recognized at law as human beings equally deserving of concern, respect and consideration”: Kapp , at para. 15 , quoting Andrews , at p. 171; see also Tremblay, at pp. 189‑92. In the context of s. 15(1) , substantive equality is promoted by eliminating discrimination. The central question is not whether one person receives less than another, but whether one person obtains less than another as a result of prejudice or stereotyping. This is the essence of the wrong or injustice that s. 15(1) is intended to prevent. As Professor Moreau puts it,

[n]o plausible theory of equality maintains that what is objectionable about unequal treatment is the mere fact that some individuals end up with more or less than others. Rather, such theories hold that unequal treatment is objectionable when, and to the extent that, this treatment is unfair.

(“The Wrongs of Unequal Treatment” (2004), 54 U.T.L.J. 291, at p. 293)

[ 180 ] Thus, substantive equality is not denied solely because a disadvantage is imposed. Rather, it is denied by the imposition of a disadvantage that is unfair or objectionable, which is most often the case if the disadvantage perpetuates prejudice or stereotypes:

We think of discrimination not just as any sort of differential treatment but as a particular kind of differential treatment: to be discriminated against is not just to be denied something that others have but to be denied it in a way that is objectionable or unfair.

(S. Moreau, “The Promise of Law v. Canada ” (2007), 57 U.T.L.J. 415, at p. 426)

[ 181 ] Finally, I would like to make one more point about the principle of creation of a “disadvantage by perpetuating prejudice”. The use of the word “perpetuating” might seem to suggest that there can be discrimination within the meaning of s. 15(1) only where the prejudice has a historical origin: see, inter alia , Koshan and Hamilton, at p. 51.

[ 182 ] But this view is incorrect. Although it can be helpful, in order to establish that an impugned law imposes a disadvantage by perpetuating prejudice, to show that certain individuals or classes of persons have historically been victims of prejudice, it is not necessary to do so. As Iacobucci J. explained in Law , at paras. 65‑67, the historical contextual factors of vulnerability, past exposure to prejudice or stereotyping and pre‑existing disadvantage are useful, but if they do not apply, this does not necessarily mean that the legislative provision at issue is currently free of prejudice. Moreover, a proper assessment of a disadvantageous law’s impact on substantive equality will in most cases require evidence of discrimination focused on the adverse effects as of the date of the claim, as opposed to the date the impugned law came into force: see, inter alia , Réaume, at p. 687. As well, historical prejudices can change; some disappear, while new ones may emerge. The concept of immutability on which my colleagues Deschamps and Abella JJ. rely in their respective reasons, on the basis in particular of Corbiere , is not synonymous with eternity. Although this concept of immutability may underline the fact that certain factors of discrimination will exist for a long time, it cannot be employed without taking the extreme diversity of those factors and of societal circumstances into account. It does not mean that the factors of discrimination can never change or disappear, especially where they are related to customs or social behaviour that could change, as in the case of the attitudes of Quebec society with respect to de facto unions.

[ 183 ] Nor does recourse to these changeable contextual factors in analyzing a specific allegation of infringement of the right to equality mean that a ground of discrimination that is accepted or rejected in a specific case cannot be relied on in another situation. Context is critical, and it must be taken into account each time, since such factors are recognized “in the context of the place of the group in the entire social, political and legal fabric of our society”: R. v. Turpin , [1989] 1 S.C.R. 1296 , at p. 1332, per Wilson J., referring to her reasons in Andrews , at p. 152. The contextual nature of these factors means that they can change along with their social context, the one that has given rise to them.

[ 184 ] In Corbiere , the majority of the Court stressed the meaning and limits of the concept of immutability of the markers of discrimination. The presence of such a marker is not necessarily proof of discrimination. According to McLachlin and Bastarache JJ., the decision to categorize a characteristic as an analogous ground instead tells the court that it should consider the situation in light of s. 15. In this regard, McLachlin and Bastarache JJ. made the following observation in Corbiere : “To say that a ground of distinction is an analogous ground is merely to identify a type of decision making that is suspect because it often leads to discrimination and denial of substantive equality. Like distinctions made on enumerated grounds, distinctions made on analogous grounds may well not be discriminatory” (para. 8).

(3) Synthesis of the Analytical Framework

[ 185 ] In light of these reasons, and subject to my comments to the effect that prejudice or stereotyping is a crucial, although not the only, factor to be considered, a court analyzing the validity of a claim that s. 15(1) has been infringed must address the following questions: (1) Does the law create a distinction based on an enumerated or analogous ground? (2) Does the distinction create a disadvantage by perpetuating prejudice or stereotyping?

[ 186 ] The claimant must therefore prove on a balance of probabilities (a) that the law creates an adverse distinction based on an enumerated or analogous ground and (b) that the disadvantage is discriminatory because (i) it perpetuates prejudice or (ii) it stereotypes. Because of their fundamental importance to the application of s. 15, I will now review the key concepts of “disadvantage”, “prejudice” and “stereotyping” in order to more precisely set out the legal framework applicable to their use.

(a) Adverse Distinction Based on an Enumerated or Analogous Ground

[ 187 ] Right away in Andrews , McIntyre J. adopted a broad definition of an adverse distinction based on an enumerated or analogous ground that could be characterized as discriminatory. According to him, it was

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. [p. 174]

[ 188 ] In Law , Iacobucci J. reiterated this definition and specified that limiting access to advantages may create a disadvantage where the law fails “to take into account the claimant’s already disadvantaged position within Canadian society resulting in substantively differential treatment between the claimant and others on the basis of one or more personal characteristics”: para. 39. For example, a failure, as in Eldridge , to take account of the fact that some deaf persons cannot receive government health services of adequate quality without the aid of an interpreter creates an adverse distinction based on an enumerated ground, namely physical disability.

[ 189 ] Thus, the claimant can show that the impugned law creates a distinction directly by imposing limitations or disadvantages on the basis of an enumerated or analogous ground: see, inter alia , Miron , at para. 131 ; Lavoie v. Canada , 2002 SCC 23 , [2002] 1 S.C.R. 769 , at para. 52 ; Trociuk v. British Columbia (Attorney General) , 2003 SCC 34 , [2003] 1 S.C.R. 835 , at para. 10 . The same is true where the law restricts access to a fundamental social institution ( Law , at para. 74) or imposes obligations that are not imposed on others ( Withler , at para. 62 ). A claimant can also show that a law creates a distinction indirectly where, “although the law purports to treat everyone the same, it has a disproportionately negative impact on a group or individual that can be identified by factors relating to enumerated or analogous grounds”: Withler , at para. 64 . At this stage, comparisons, if any, can help to demonstrate the existence of an adverse distinction.

[ 190 ] Once such a distinction is established, the court must determine whether it is based on an enumerated or analogous ground. These grounds stand as “constant markers of suspect decision making or potential discrimination”: Corbiere , at para. 8 ; see also Lavoie , at paras. 2 and 41 . As I mentioned above, although a disadvantageous law will be suspect if they are present, it will not automatically be discriminatory. As this Court has pointed out, these grounds correspond to personal characteristics that cannot be changed or can be changed only at unacceptable cost to the claimant’s personal identity: Corbiere , at para. 13 ; Withler , at para. 33 .

[ 191 ] If the court finds that the government action being challenged creates an adverse distinction based on an enumerated or analogous ground, it must then consider the context and the facts of the case to determine whether the distinction is discriminatory because it violates the right to substantive equality by perpetuating prejudice or stereotyping: Withler , at para. 34 . The presence of relevant contextual factors will make it easier to determine whether such violations have occurred. I repeat that at this second stage, comparison between the claimant and other persons, although not indispensable, may bolster the understanding of the context of the discrimination claim: Withler , para. 65 .

(b) Perpetuating Prejudice

[ 192 ] The first way that substantive inequality — discrimination — may be established is by showing that the impugned disadvantageous law, in purpose or effect, perpetuates prejudice against members of a group on the basis of personal characteristics covered by s. 15(1): Withler , at para. 35 . Such a law will be found to be discriminatory if it “has the effect of perpetuating or promoting the view that the individual is less capable, or less worthy of recognition or value as a human being or as a member of Canadian society”: Law , at para. 51. In my view, this inquiry is of particular importance, as it is most likely to result in a finding of the types of discrimination to which s. 15 applies. It provides a framework to enable courts to consider such discrimination without lapsing totally into subjectivism. I do not rule out the theoretical possibility that there are forms of exclusion for which this analytical framework would be ill-suited. In practice, however, I feel that it would be hard to identify them unless all that was required for s. 15 to apply was a finding of disadvantages related to prohibited grounds and unless the inquiry into discrimination per se was dispensed with. This is another possible conception of the right to equality guaranteed by s. 15, but it is not the one this Court has adopted since Andrews .

[ 193 ] An adverse distinction therefore discriminates by perpetuating prejudice if it denotes an attitude or view concerning a person that is at first glance negative and that is based on one or more of the personal characteristics enumerated in s. 15(1) or on characteristics analogous to them. An adverse distinction can also be inconsistent with s. 15, even if there is no discriminatory intent whatsoever, if it has a discriminatory effect. Since equality is an expression of the values of a society in which all are secure in the knowledge that they are recognized at law as human beings equally entitled to respect, the perpetuation of such a negative view constitutes a denial of substantive equality.

[ 194 ] Thus, if the government either directly or indirectly disadvantages certain persons who share one of these personal characteristics that are immutable, or changeable only at unacceptable cost, it may be that a negative view is thereby being expressed either consciously or unconsciously. The government can treat individuals or groups differently, however. For example, it is accepted that it can confer advantages or impose disadvantages based on individual merit or capacity. Under s. 15(2), it can also implement specific programs to help disadvantaged groups. But the government is prohibited from showing certain individuals greater respect and consideration simply because they share an enumerated or analogous personal characteristic.

[ 195 ] Denise G. Réaume has described more precisely the meaning of the concept of prejudice and the nature of the harm that results from the expression or perpetuation of prejudice:

A legislative distinction based on prejudice denies a class of persons a benefit out of animus or contempt. It directly connotes a belief in their inferiority, a denial of equal moral status. Legislated prejudice denies a benefit for the sake of causing harm to those denied. It thus treats members of a group as loci of intrinsic negative value, rather than intrinsic moral worth. Such treatment not only deprives them of the concrete benefit at issue, but also, through doing so, treats them as unworthy of basic human respect. . . .

Prejudice works through the attribution of negative worth to personal characteristics that are important aspects of identity; it thus constitutes an assault on the sense of self of its victims. Personal identity has both an individual and a social dimension. The kinds of characteristics that people regard as important to their sense of self tend to be, at the same time, characteristics by which they define themselves as individuals and through which they identify as members of a group. This group affiliation is as important to human identity as any purely individual understanding of the self. We develop a sense of self only through our interactions with others, and our most intimate and formative interactions are frequently with people who share a cultural or ethnic identity that distinguishes them from other such clusters of people in society. And we know from our social and political history that it has tended to be precisely this aspect of identity that has often been targeted for contempt — individuals have been denied respect through use of a characteristic identifying them as part of a group that is devalued. [Emphasis in original; pp. 679‑80.]

[ 196 ] As I mentioned above, the devaluing of individuals need not be intentional to be considered an infringement of s. 15(1) . Although an intention that reflects prejudice on the part of the government and its officials can sometimes be identified, such an intention does not exist in every situation involving discrimination — far from it.

[ 197 ] For instance, a government might make laws that unintentionally convey a negative social image of certain members of society. This situation could arise if the government favours certain individuals at the expense of others because the others share an enumerated or analogous characteristic. Such laws would express or perpetuate prejudice against certain individuals by establishing a hierarchy of worth based on prohibited grounds of discrimination, such as sex or sexual orientation. The identification of such prejudice will require a contextual inquiry, which might take account, among other things, of the disadvantages suffered by groups defined by a common personal characteristic.

[ 198 ] Moreover, rules that are seemingly neutral (because they do not draw obvious distinctions) may also treat certain individuals like second‑class citizens whose aspirations are not equally deserving of consideration. As Réaume writes, “[p]ublic institutions and programs built, even unwittingly, in the image of a dominant group convey the message that others are not equally entitled to participate in society and its enterprises, and are not equally members of its institutions”: p. 686. In such a case, a disadvantage resulting from exclusion constitutes an expression and perpetuation of prejudice against certain persons.

[ 199 ] For example, in Eldridge , this Court found that the lack of interpretation services for deaf persons in the public health care system meant that such persons were treated as less worthy even though the government had not intended to devalue them. The disadvantage in that case was not based on stereotyping. Rather, it conveyed a devalued image of deaf persons by failing to recognize them as human beings who deserved to participate fully in Canadian society. The government was therefore imposing a disadvantage on them that constituted an expression of prejudice. The following passage from Eldridge illustrates how a law that does not stereotype can nonetheless be discriminatory if it conveys prejudice by denying certain individuals full participation in a fundamental aspect of life in society because of a disability:

The principal object of certain of the prohibited grounds is the elimination of discrimination by the attribution of untrue characteristics based on stereotypical attitudes relating to immutable conditions such as race or sex. In the case of disability, this is one of the objectives. The other equally important objective seeks to take into account the true characteristics of this group which act as headwinds to the enjoyment of society’s benefits and to accommodate them. Exclusion from the mainstream of society results from the construction of a society based solely on “mainstream” attributes to which disabled persons will never be able to gain access. Whether it is the impossibility of success at a written test for a blind person, or the need for ramp access to a library, the discrimination does not lie in the attribution of untrue characteristics to the disabled individual. The blind person cannot see and the person in a wheelchair needs a ramp. Rather, it is the failure to make reasonable accommodation, to fine‑tune society so that its structures and assumptions do not result in the relegation and banishment of disabled persons from participation, which results in discrimination against them. The discrimination inquiry which uses “the attribution of stereotypical characteristics” reasoning as commonly understood is simply inappropriate here. It may be seen rather as a case of reverse stereotyping which, by not allowing for the condition of a disabled individual, ignores his or her disability and forces the individual to sink or swim within the mainstream environment. It is recognition of the actual characteristics, and reasonable accommodation of these characteristics which is the central purpose of s. 15(1) in relation to disability. [para. 65]

(Quoting Eaton v. Brant County Board of Education , [1997] 1 S.C.R. 241, at para. 67 .)

[ 200 ] As I mentioned above, a court enquiring into the expression or perpetuation of prejudice can consider, among other things, the nature or scope of the benefit or interest which the claimant feels he or she has been denied. Does the distinction restrict access to a fundamental social institution or impede full membership in Canadian society? If the answer is yes, this could indicate that the government action expresses, or has the effect of perpetuating, prejudice against — i.e., a lower or demeaning opinion of — certain persons. Eldridge is one example of such a situation, as I have stated.

(c) Stereotyping

[ 201 ] In the analytical approach I am recommending here, the second way that substantive inequality — discrimination — may be established is “by showing that the disadvantage imposed by the law is based on a stereotype that does not correspond to the actual circumstances and characteristics of the claimant or claimant group”: Withler , at para. 36 . Such a law will be discriminatory because it is premised upon personal traits or circumstances that do not relate to individual needs, capacities or merits: Law , at para. 53. Laws premised on an inaccurate characterization of an individual or group on grounds that are unacceptable under s. 15(1) thus become arbitrary themselves: see, inter alia , Moreau, “The Wrongs of Unequal Treatment”, at p. 298.

[ 202 ] The following comments by Réaume contain an interesting description of the nature of negative stereotypes and their impact on the right to equality:

Stereotypes are inaccurate generalizations about the characteristics or attributes of members of a group that can usually be traced back to a time when social relations were based more overtly on contempt for the moral worth of the group. . . . Negative characteristics, such as lack of intelligence, laziness, being fit for some pursuits rather than others, predisposition to criminality, avarice, vice, etc., which are in fact distributed throughout the human race, are falsely attributed predominantly to members of a particular group. It is then the negative characteristic that becomes the focus of contempt. Nevertheless, inaccurate assumptions and stereotypes about the capacities, needs, or desires of members of a particular group can carry forward ancient connotations of second class status, even if the legislators did not intend that meaning. The overt hostility may have come to be washed out of the picture with the passage of time or the “normalization” of such attitudes, but the implication that those to whom the stereotype applies are less worthy than others remains.

Once this construction of a group has set in, others are likely to treat members of that group disadvantageously out of an honest belief that this merely reflects their just deserts or even simply because that is how everyone treats them, without ever thinking about the insult involved. They may even understand their conduct, as with certain traditional sexist practices, as a positive effort to accommodate the “natural weaknesses” of the stereotyped group. However, neither the absence of contempt as a subjective matter nor well‑meaning paternalism prevents the use of stereotype from violating dignity. To be denied access to benefits or opportunities available to others on the basis of the false view that because of certain attributes members of one’s group are less worthy of those benefits or less capable of taking up those opportunities can scarcely fail to be experienced as demeaning because it is demeaning. The message such legislation sends is that members of this group are inferior or less capable, and such a message is likely, in turn, to reinforce social attitudes attributing false inferiority to the group. [Emphasis in original; pp. 681‑82.]

[ 203 ] Continuing with a contextual approach, it may be helpful in determining whether stereotypes exist to consider the question of the correspondence, or lack thereof, between the grounds on which the claim is based and the actual need, capacity or circumstances of the claimant or the affected group. For example, in M. v. H. , the identification and rejection of certain stereotypes led the Court to declare a law under which support remedies were available only to opposite‑sex spouses to be invalid. The law in question conveyed the negative stereotype that persons of the same sex were incapable of forming intimate relationships involving economic interdependence similar to those of opposite‑sex couples, without regard to their actual circumstances. As a result, the impugned law violated s. 15(1) .

(d) Summary

[ 204 ] In accordance with the general analytical framework for the application of s. 15(1) of the Charter , there are thus two ways for a claimant to show that a law that draws a distinction based on an enumerated or analogous ground is discriminatory. On the one hand, the claimant can show that the impugned disadvantageous law perpetuates prejudice against members of a group. On the other hand, the claimant can prove that the disadvantage imposed by the law is based on a stereotype. Two comments are in order in this regard.

[ 205 ] First, because either one of these facts can on its own support a finding that the impugned law infringes s. 15(1) , a distinction need not be based on a stereotype to be discriminatory: Gosselin , at para. 116 , per L’Heureux‑Dubé J.; Lavoie , at para. 52 , per Bastarache J. A disadvantageous law can also be found to be discriminatory on the basis that it expresses or perpetuates prejudice. The Court has thus explicitly acknowledged the inadequacy of a uniquely stereotype‑based approach that has been criticized by several authors: see, inter alia , M. Young, “Blissed Out: Section 15 at Twenty”, in S. McIntyre and S. Rodgers, eds., Diminishing Returns: Inequality and the Canadian Charter of Rights and Freedoms (2006), 45, at p. 59; G. Brodsky, “Case Comment: Gosselin v. Quebec (Attorney General) : Autonomy with a Vengeance” (2003), 15 C.J.W.L. 194, at p. 212; M. Young, “Unequal to the Task: ‘Kapp’ing the Substantive Potential of Section 15”, in S. Rodgers and S. McIntyre, eds., The Supreme Court of Canada and Social Justice: Commitment, Retrenchment or Retreat (2010), 183, at pp. 204‑9.

[ 206 ] Second, the existence of these two approaches, which make it possible to prove that the impugned law infringes s. 15(1), also guides the use of the contextual factor of correspondence between the ground or grounds of discrimination on which the claim is based and the actual circumstances of the claimant or the affected group. It now seems clearer that this factor can be used to determine whether the distinction creates a disadvantage by stereotyping: Kapp , at para. 23 . However, the Court also recognizes that the correspondence factor may not be sufficient to support a finding of expression or perpetuation of prejudice. Although it is true that prejudice and stereotyping are frequently linked, a claimant can also show that the impugned law expresses or perpetuates prejudice by emphasizing other contextual factors unrelated to that of correspondence. The Court thus acknowledges that the correspondence factor is only one of many factors that can be used to establish substantive inequality. Indeed, in the past, some authors deplored an approach that was overly dependent on the use of the correspondence factor at the expense of other contextual factors: see, inter alia , B. Ryder, C. C. Faria and E. Lawrence, “What’s Law Good For? An Empirical Overview of Charter Equality Rights Decisions” (2004), 24 S.C.L.R. (2d) 103 , at pp. 120‑25 .

G. Walsh — Precedential Value of the Decision

(1) Nature of the Case

[ 207 ] At this point, the evolution of the Court’s case law on the right to equality raises the issue of the scope and application of an important case on the application of equality rights of de facto or common law spouses. In Walsh , which was decided several years before Kapp and Withler , this Court was dealing with an alleged violation of s. 15(1) by a Nova Scotia statute concerning the matrimonial rights of married spouses. The case involved a challenge to the validity of Nova Scotia’s Matrimonial Property Act , R.S.N.S. 1989, c. 275 (“ MPA ”), under s. 15(1) of the Charter . The MPA established the consequences of marriage breakdown by creating a presumption that matrimonial property was to be divided equally between the former spouses. The MPA regulated the division of the property acquired by one of the spouses either before or during the marriage and established guidelines to assist in determining whether that property continued to be the separate property of each spouse or became common property of the parties. However, it did not provide for spousal support on the breakdown of a marriage. Section 2 (g) of the MPA defined the term “spouse” by referring only to a man and a woman who were married to each other. It therefore had the effect of excluding unmarried opposite‑sex couples who lived together.

[ 208 ] The claimant in Walsh argued that, by limiting the presumption of equal division to married couples, the MPA discriminated on the basis of an analogous ground, namely marital status. The Court therefore had to decide whether the failure to include unmarried opposite‑sex couples in the ambit of the MPA violated s. 15(1) .

[ 209 ] Bastarache J., writing for a majority of the Court, found that the distinction between common law spouses and married spouses was not discriminatory and therefore did not violate s. 15(1) . Gonthier J. added a few comments on the contractual nature of marriage and its fundamental place in society. L’Heureux‑Dubé J., dissenting, found that married couples and unmarried couples were functionally identical. As a result, she concluded that the exclusion of unmarried couples from the ambit of the MPA was discriminatory and that the discrimination could not be justified under s. 1 of the Charter . Given the importance of that case, I will summarize the bases for the Court’s decision. I will then discuss the relevance of Bastarache J.’s reasons to the appeals now before this Court.

(2) Bases for Bastarache J.’s Reasons

[ 210 ] Bastarache J. analyzed the claimant’s position by applying the three‑stage framework developed by Iacobucci J. a few years earlier in Law . He began by considering whether the MPA imposed differential treatment between the claimant and others. He then asked whether the differential treatment was based on an enumerated or analogous ground. Finally, he considered whether the law in question had a discriminatory purpose or effect for the purposes of the equality guarantee. That would be the case if the law imposed a burden upon or withheld a benefit from the claimant in a manner which reflected the stereotypical application of presumed group or personal characteristics, or if it otherwise had the effect of perpetuating or promoting the view that the claimant was less deserving of respect than other members of Canadian society. Thus, Bastarache J.’s analysis was based on the promotion of substantive equality that was also the basis for Law and for this Court’s case law since Andrews . His approach represented a continuation of that case law.

[ 211 ] Bastarache J. began by acknowledging that the MPA imposed differential treatment within the meaning of s. 15(1) because it applied only to persons who were legally married and excluded persons in a common law relationship. That differential treatment was based on the analogous ground of marital status, as the Attorney General of Nova Scotia had conceded. Bastarache J. therefore proceeded to the third stage of the analytical framework from Law and asked “whether a reasonable heterosexual unmarried cohabiting person, taking into account all of the relevant contextual factors, would find the MPA ’s failure to include him or her in its ambit has the effect of demeaning his or her dignity”: Walsh , at para. 38 . At this point in the analysis, he undertook a contextual analysis of the claimant’s argument.

[ 212 ] Bastarache J. acknowledged that unmarried spouses had experienced historical disadvantage, social prejudice, and stereotyping of various kinds. However, he found that the version of the MPA then in force properly accommodated the claimant’s circumstances. It reflected the differences between common law relationships and marriage and respected the fundamental autonomy and dignity of common law spouses. In Bastarache J.’s view, despite the functional similarities between common law and married spouses, there was a fundamental difference between the two groups.

[ 213 ] On this point, Bastarache J. noted that the MPA deemed all marriages to be economic partnerships and thus imposed a significant alteration to the former status quo of married persons’ proprietary rights and obligations. Those statutorily created restrictions, obligations and rights arose at the time of the marriage and continued throughout the duration of the marriage until separation or death. According to Bastarache J., “[t]he decision to marry, which necessarily requires the consent of each spouse, encapsulates within it the spouses’ consent to be bound by the proprietary regime that the MPA creates”: Walsh , at para. 48 . This was a fundamental difference between married couples and unmarried couples; the former had chosen to be bound by the MPA , while the latter had not given their consent to be so bound.

[ 214 ] Moreover, common law spouses who were unwilling to marry but wanted to modify their proprietary rights and obligations had various alternatives for clearly expressing their agreement. They could own property jointly or enter into domestic contracts that could be enforced pursuant to the Maintenance and Custody Act , R.S.N.S. 1989, c. 160 (“ MCA ”), s. 52(1) , and the Maintenance Enforcement Act , S.N.S. 1994‑95, c. 6 (“ MEA ”), s. 2 (e). Bastarache J. also took note of the recent Law Reform (2000) Act , S.N.S. 2000, c. 29 (“ LRA ”), as a contextual consideration. Under the LRA , common law partners who decided to register their partnerships under the Vital Statistics Act , R.S.N.S. 1989, c. 494, thereby became subject to the MPA . There were therefore several options available to common law spouses who wished to avoid the application of the general principle that people who do not take consensual action maintain the right to deal with their property as they see fit. On this basis, Bastarache J. stated that the legislation corresponded to the free choice of the individuals involved and to their actual situation:

The MPA , then, can be viewed as creating a shared property regime that is tailored to persons who have taken a mutual positive step to invoke it. Conversely, it excludes from its ambit those persons who have not taken such a step. This requirement of consensus, be it through marriage or registration of a domestic partnership, enhances rather than diminishes respect for the autonomy and self‑determination of unmarried cohabitants and their ability to live in relationships of their own design. As Iacobucci J. phrased it in Law , at para. 102, “[t]he law functions not by the device of stereotype, but by distinctions corresponding to the actual situation of individuals it affects.”

( Walsh , at para. 50 )

[ 215 ] Bastarache J. thus found that it was not stereotyping to believe that common law couples had chosen to avoid the institution of marriage: Walsh , at para. 43 . Because this group was highly heterogeneous, there was no basis for arguing that even though common law spouses had not expressed a mutual intention to alter their proprietary regime, they had nevertheless implicitly chosen to be bound by the MPA . Accordingly, there was no constitutional requirement that the legislature extend the protections of the MPA to couples who had not expressed their consent to restrict their ability to deal with their own property during the relationship or to share their assets and liabilities should the relationship break down. The legislature’s decision to respect the freedom of choice of common law spouses was not unconstitutional.

[ 216 ] Furthermore, in addition to not being based on negative stereotypes, the exclusion of unmarried couples from the MPA did not promote or perpetuate the idea that they were less capable, or less worthy of respect or value as members of Canadian society. Bastarache J. concluded that the MPA was not discriminatory in light of the values — such as dignity, liberty and autonomy — that underlie the Charter , including s. 15 thereof. Rather, it maintained the liberty of all spouses to make fundamental choices in their lives, thereby respecting the fundamental personal autonomy and dignity of common law spouses. According to Bastarache J., even if the freedom to marry can sometimes be illusory, s. 15(1) did not justify eliminating an individual’s freedom of choice or imposing on common law spouses a regime designed for persons who had made an unequivocal commitment to form an equal partnership as provided for in the MPA : Walsh , at paras. 57, 62 and 63 . Bastarache J. distinguished that case from the Court’s decisions in Miron and M. v. H. , and concluded that the MPA was not discriminatory and therefore did not conflict with s. 15(1) ’s purpose of ensuring substantive equality.

(3) Kapp and Withler and Their Impact on Walsh

[ 217 ] In my opinion, Bastarache J. would have reached the same conclusion if his analysis had been based on the reworked analytical framework from Kapp and Withler . Although the MPA imposed differential treatment based on an analogous ground, that distinction did not create a disadvantage by perpetuating prejudice or stereotyping. Bastarache J.’s analysis of the MPA is therefore consistent with the decisions rendered after Walsh and with the principle of substantive equality, which “insists on going behind the facade of similarities and differences [and] asks not only what characteristics the different treatment is predicated upon, but also whether those characteristics are relevant considerations under the circumstances”: Withler , at para. 39 . As I mentioned above, this analysis was based on the wish to promote substantive equality that was also the basis for Law , despite the conceptual difficulties and problems of application that led to Kapp and Withler .

[ 218 ] In Walsh , the MPA did not have the effect of favouring marriage at the expense of cohabitation or of denying the worth of common law relationships. Nor did it draw distinctions that caused a disadvantage by perpetuating prejudice. According to Bastarache J., the Nova Scotia legislature was not favouring one form of relationship over another. It was merely defining the legal content of relationships and providing that any individuals in a conjugal relationship could, without changing their marital status, make a consensual choice to avail themselves of rights, obligations and restrictions analogous to the ones contained in the MPA . As a result, the law did not favour one form of relationship over another or express any prejudice against common law relationships.

[ 219 ] In Nova Scotia, as Bastarache J. demonstrated, the legislature had defined the content and consequences of various forms of relationship but had not favoured one form over another. Marriage and the registration of a common law partnership triggered the mutual rights, obligations and restrictions set out in the MPA , including the presumption of equal division of property should the relationship break down. Common law couples could also transform their relationships into economic partnerships as contemplated in the MPA by entering into domestic contracts that could be enforced under the MCA and the MEA . They could also purchase property jointly. The legislation did not establish an unacceptable hierarchy among the various forms of conjugal relationships. By expressing a consensual choice or intention, spouses could opt in to the regime of their choice to which the rights and obligations established by the legislature applied. Walsh was thus based on a principle of freedom to choose between different marital statuses that had different consequences for spouses, and that principle did not in this context infringe the constitutional equality guarantee. The principle in question continues to be valid in the circumstances of the case at bar despite the subsequent developments in the case law.

[ 220 ] In this regard, Walsh differed significantly from Miron . In the Insurance Act , R.S.O. 1980, c. 218, the Ontario legislature had not defined the content of relationships (that is, the relations between the members of a couple). Instead, it had favoured marriage over common law relationships by limiting benefits under an automobile insurance plan on the basis of marital status to those who were married.

[ 221 ] The common law spouses in Miron had therefore been excluded from certain provisions relating to automobile insurance because they were not married, a ground that was in all probability irrelevant to automobile insurance. As Bastarache J. pointed out in Walsh , “[t]he marital status of the couple should have had no bearing on the availability of the benefit”: para. 53. In the legislation at issue in Miron , the Ontario legislature had not defined the legal content of the spouses’ relationship with one another. Rather, it had given one class of couples a privilege that expressed or perpetuated prejudice in favour of marriage and against common law relationships.

[ 222 ] Bastarache J. then observed that the MPA was not based on stereotypes about common law spouses. To support that conclusion, he considered the contextual factor of correspondence and found that it was not stereotyping to believe that common law couples had chosen to avoid the institution of marriage. The common law spouses in that case had not expressed a consensual intention to change their legal property relationship in any of the ways established by the government, one of which would have been to enter into a domestic contract as a common law couple. Thus, in distinguishing between married and unmarried spouses, the MPA was not based on a stereotype that did not correspond to the actual circumstances and characteristics of common law spouses.

[ 223 ] Walsh can be distinguished in a similar way from M. v. H. , in which the Ontario legislature had limited certain provisions of the Family Law Act , R.S.O. 1990, c. F.3 (“ FLA ”), to heterosexual common law couples. More specifically, the FLA had given opposite‑sex common law spouses, but not same‑sex common law spouses, the right to seek support if their relationship broke down. The possibility of obtaining support had thus been extended “beyond married persons to include individuals in conjugal opposite‑sex relationships of some permanence”: M. v. H. , at para. 2 ; see also para. 52. This differential treatment of same‑sex and opposite‑sex common law spouses was based on the analogous ground of sexual orientation.

[ 224 ] In the FLA , the legislature had defined a common law relationship as a conjugal relationship of some permanence. This form of relationship entitled the spouses to support if they ceased living together. Individuals who decided to live in a common law relationship were therefore covered by provisions of the FLA that entitled them to apply to a court for support. Although same‑sex common law spouses could choose to enter into a relationship similar to the one defined in the FLA , they were denied the benefit of that support remedy. A majority of the Court held that the exclusion of same‑sex common law spouses “implies that they are judged to be incapable of forming intimate relationships of economic interdependence, without regard to their actual circumstances”: M. v. H. , at para. 3 . Since “[b]eing in a same‑sex relationship does not mean that it is an impermanent or a non‑conjugal relationship” ( M. v. H. , at para. 70 ), the legislature was imposing differential treatment that was found to be discriminatory because it created a disadvantage as a result of stereotyping. It was impossible for same‑sex spouses to opt in to Ontario’s legal regime. The exclusion of same‑sex couples from the FLA was based on an inaccurate and stereotypical characterization of their actual circumstances and therefore violated their dignity.

[ 225 ] The opposite was true in Walsh , in which the legislature had provided that the presumption of equal division of property was conditional on the expression of a consensual intention (through marriage, a domestic contract or, later, the registration of a common law partnership). Since common law spouses, unlike married spouses, had not expressed such an intention, their exclusion from the MPA was not based on an inaccurate and stereotypical characterization of their actual circumstances and did not violate their dignity.

[ 226 ] In short, it seems to me that Bastarache J. would have reached the same conclusion in Walsh even if the current analytical framework had applied. Although the MPA imposed differential treatment based on an analogous ground, that distinction did not create a disadvantage by expressing or perpetuating prejudice or by stereotyping. Accordingly, subject to the differences between the MPA and the Quebec scheme, Bastarache J.’s analysis can properly serve as a precedent for our assessment of the infringement of s. 15(1) alleged by A in the instant case. It is true that the Court’s analysis concerned not the obligation of support, but the Nova Scotia legislation’s presumption of equal division of family assets. However, Bastarache J.’s comments on the sources of the distinctions between the various forms of relationships and the consequences of those distinctions remain relevant. At any rate, even without Walsh , the same principles applied under the current analytical framework for s. 15(1) would lead to the conclusions I propose with respect to the Quebec legislation being challenged by A.

H. Preliminary Comments on the Approach Proposed by the Court of Appeal

[ 227 ] Before actually considering A’s constitutional challenge, I believe a clarification is required with regard to the judgment of Dutil J.A. in the Court of Appeal. In her reasons, Dutil J.A. found that the main issues raised by A’s appeal concerned [ translation ] “the obligation of support” on the one hand and “the division of property” upon the breakdown of a relationship on the other. That distinction was central to her reasons and served as the framework for her reasoning. The same distinction is central to Deschamps J.’s reasons. Regarding the discrimination that allegedly results from the provisions on partition of property (family patrimony, protection of the family residence, partnership of acquests and compensatory allowance), Dutil J.A. considered herself bound by the reasoning in Walsh and concluded on that basis that those provisions are not discriminatory. Deschamps J. concludes that the exclusion of de facto spouses from the provisions in question is discriminatory, but that it is justified under s. 1 of the Charter .

[ 228 ] Dutil J.A. argued that, unlike in the case of the provisions on the partition of property, the obligation of support is not contractual in origin but [ translation ] “exists to meet basic needs and represents an aspect of social solidarity”: para. 68. As a result, she found that art. 585 C.C.Q. , which deals with the obligation of support, had to be assessed separately from the rest of the impugned provisions, that the reasoning in Walsh was not applicable to art. 585 and that s. 15(1) of the Charter had to be considered in relation to art. 585 alone. In my opinion, this is incorrect.

[ 229 ] I find that the distinction drawn by Dutil J.A. between the “partition of property” and the “obligation of support” is inappropriate, as it disregards the character of an “economic partnership” that the Quebec legislature has established for marriage and the civil union. It also disregards the fact that this partnership is structured around a mandatory primary regime that has both patrimonial and extrapatrimonial aspects and that the primary regime establishes the obligation of support as an effect of marriage and of the civil union. In this sense, the obligation of support is tied to the other effects of marriage and of the civil union, such as the obligation to contribute to household expenses, rights and obligations with respect to the family residence, and the creation of a family patrimony. It forms an integral and indissociable part of the set of measures that constitute Quebec’s primary regime.

[ 230 ] This distinction also overlooks the fact that each of the impugned provisions shapes the spouses’ private patrimonial relationship and that a number of them rebalance the distribution of property between the spouses in some way, including through the payment of certain amounts or the granting of rights of use of or ownership in certain property. As well, by increasing the patrimony of the less wealthy spouse, each of these measures can enhance that spouse’s autonomy and reduce the potential of his or her becoming dependent on government assistance.

[ 231 ] In Quebec law, the obligation of support is one of the mandatory effects of marriage (or of a civil union) that the spouses may not renounce in a marriage contract. As an expression of the duty of succour that each spouse owes the other where needed, the obligation of support, like the spouses’ obligation to contribute towards household expenses in proportion to their respective means, is part of the primary regime that the spouses accept when they choose to marry. This obligation, in the form of the duty of succour, lasts as long as the spouses remain married or in a civil union.

[ 232 ] However, both the Civil Code and the Divorce Act provide that such an obligation may be imposed even after the bonds of civil union or marriage have been dissolved. Thus, in Bracklow , the Court considered the Divorce Act and enquired into the basis for the obligation of support after the marriage bond is dissolved. The Court noted first that support may be awarded to compensate a spouse who has been economically disadvantaged during the marriage: Bracklow , at paras. 36, 39 and 49 . In this regard, support is similar to a compensatory allowance, since its objective will be to compensate one of the spouses for losses incurred as a result of the marriage and its breakdown.

[ 233 ] However, support can also be non‑compensatory, and based on the “mere fact that a person who formerly enjoyed intra‑spousal entitlement to support now finds herself or himself without it”: Bracklow , at para. 41 . This basis for the obligation of support “postulates each of the parties to the marriage agreeing, as independent individuals, to marriage and all that it entails — including the potential obligation of mutual support” and “recognizes the artificiality of assuming that all separating couples can move cleanly from the mutual support status of marriage to the absolute independence status of single life, indicating the potential necessity to continue support, even after the marital ‘break’”: Bracklow , at paras. 30‑31 .

[ 234 ] As the Court explained, this form of post‑divorce support obligation is based on two important factors. First, it is based on the fact that an obligation of support, known as the duty of succour in the civil law, exists during marriage and may remain necessary after a marriage breaks down. Second, it can be seen as one of the many consequences to which individuals agree when they choose to marry. Thus, the choice to get married results in the application of certain mandatory effects codified in the Civil Code of Québec , such as the claim that arises for the value of the family patrimony, and of the effects set out in the Divorce Act , which provides for an obligation of support following the breakdown of a marriage. In both cases, the obligation of support is therefore based on consent to the marriage or civil union.

[ 235 ] Accordingly, these appeals cannot be disposed of on the basis of the distinction made by Dutil J.A. between the partition of property and the obligation of support. Indeed, Beauregard J.A. seemed to acknowledge this in his concurring reasons in admitting that, were it not for Walsh , the Court of Appeal’s reasoning on the discriminatory nature of the support obligation provision would apply equally to the other impugned provisions, such as those on the family patrimony: paras. 192 and 194. The main issue raised by these appeals is not whether the exclusion of de facto spouses from the obligation of support is discriminatory, but whether their exclusion from the entire statutory framework imposed on married and civil union spouses is discriminatory under s. 15(1) of the Charter . For the reasons that follow, I find that it is not.

I. A’s Discrimination Claim

[ 236 ] In these appeals, as I mentioned above, A is challenging the constitutional validity of the provisions of the Civil Code of Québec dealing with the family residence (arts. 401 et seq. ), the family patrimony (arts. 414 et seq. ), the compensatory allowance (arts. 427 et seq. ), the partnership of acquests (arts. 432 et seq. ) and the obligation of spousal support ( art. 585 ) on the basis of the fact that they apply to private legal relationships of married and civil union spouses but not to those of de facto spouses. To prove that she has been discriminated against, A must show on a balance of probabilities that the provisions in question create an adverse distinction based on an enumerated or analogous ground and that the disadvantage is discriminatory because it perpetuates prejudice or stereotypes.

[ 237 ] Although the burden of proving that the impugned provisions discriminate in a substantive sense lies on A, the Court can of course take judicial notice of certain facts. If it does so, it must follow the rules that permit, but define the limits of, judicial notice. As McLachlin J. explained in R. v. Williams , [1998] 1 S.C.R. 1128 , “[j]udicial notice is the acceptance of a fact without proof”: para. 54. In other words, it “dispenses with the need for proof of facts that are clearly uncontroversial or beyond reasonable dispute”: R. v. Find , 2001 SCC 32 , [2001] 1 S.C.R. 863 , at para. 48 .

[ 238 ] According to the principles laid down in Find , judicial notice applies only where facts are either

(1) so notorious or generally accepted as not to be the subject of debate among reasonable persons; or (2) capable of immediate and accurate demonstration by resort to readily accessible sources of indisputable accuracy: R. v. Potts (1982), 66 C.C.C. (2d) 219 (Ont. C.A.); J. Sopinka, S. N. Lederman and A. W. Bryant, The Law of Evidence in Canada (2nd ed. 1999), at p. 1055. [para. 48]

(See also R. v. Spence , 2005 SCC 71 , [2005] 3 S.C.R. 458, at para. 53 .)

[ 239 ] These categories of facts are obviously limited. A court must therefore be cautious when asked to take judicial notice of particular facts. In particular, it must refrain from taking judicial notice of social phenomena unless they are not the subject of reasonable dispute for the particular purpose for which they are to be used: Spence , at para. 65 . As Binnie J. explained in Spence , the closer a fact approaches the dispositive issue, or the closer it is to the centre of the controversy between the parties, the more stringently the court ought to verify it before taking judicial notice of it: paras. 60‑61.

[ 240 ] Aside from the facts of which the court may take judicial notice, the general rule of evidence remains the same: “We must decide [the] case on the evidence before us”: Gosselin , at para. 66 . This rule applies to cases involving allegations of infringements of the right to equality, including appeals to this Court.

(1) Adverse Distinction Based on an Enumerated or Analogous Ground

[ 241 ] In this appeal, the first matter to be proved by A presents no problem. The various articles of the Civil Code at issue here apply only to persons who are married or in a civil union. They govern the private legal relationships of such persons, but they do not apply to the relationships of de facto spouses. They therefore draw a distinction based on the analogous ground of marital status.

[ 242 ] That distinction may result in disadvantages for those who are excluded from the statutory framework applicable to a marriage or a civil union. Generally speaking, when de facto spouses separate, one of them will likely end up in a more precarious patrimonial situation than if the couple had been married or in a civil union. As a result, unless these de facto spouses have exactly the same earning capacity and exactly the same patrimony, regardless of the origin of that patrimony, one of them will be in a worse position after the relationship ends than would a married spouse in a similar patrimonial situation. A married spouse whose marriage breaks down would be entitled to partition under the regimes of family patrimony and partnership of acquests and might also be awarded the ownership of certain movables, the use of the family residence, a compensatory allowance and support. The de facto spouse, on the other hand, will not have the rights recognized by the relevant provisions of the Civil Code . In reality, however, each form of conjugal relationship is likely to have its share of disadvantages for one or the other of the spouses, depending on their personal circumstances at the time of the breakdown. The nature of these disadvantages will vary with the position of each of the spouses and the nature of the legal regime applicable to them.

[ 243 ] The provisions relating to the family patrimony, the family residence, the compensatory allowance, the partnership of acquests and the obligation of support therefore have the effect of creating a distinction based on an analogous ground, and that distinction can result in a disadvantage. What remains to be determined is whether the exclusion of de facto spouses from the framework applicable to marriage and civil unions discriminates in a substantive sense by violating the principle of substantive equality protected by s. 15(1) of the Charter . According to the general analytical framework in place since Withler and Kapp , there are two ways for A to prove this. First, she can show that the disadvantageous law perpetuates prejudice against de facto spouses. Second, she can show that the disadvantage imposed by the law is based on a stereotype. If either of these things is shown, the impugned provisions will be found to violate s. 15(1) .

(2) Perpetuating Prejudice

[ 244 ] The first way that substantive inequality may be established is by showing that the impugned disadvantageous law, in purpose or effect, perpetuates prejudice against members of a group on the basis of personal characteristics covered by s. 15(1): Withler , at para. 35 . Generally speaking, a law will perpetuate prejudice if it denotes a negative attitude or opinion concerning a person that is based on a personal characteristic enumerated in s. 15(1) or a characteristic analogous thereto. The same is true of provisions that attribute greater moral worth to certain persons at the expense of another group of persons on the basis of such a characteristic. Legislation that establishes a hierarchy among different individuals on the basis of their having or not having an enumerated or analogous characteristic would also perpetuate prejudice. As Professor Réaume explains, prejudice also works through the attribution of negative worth to personal characteristics that are important aspects of human identity: pp. 679‑80.

[ 245 ] Prejudice can sometimes be seen on the very face of the legislation, especially if the legislation reflects contempt for or hostility toward the group concerned or if its purpose is to inflict prejudice on a group of persons. In such circumstances, discrimination through the expression or perpetuation of prejudice will likely be intentional.

[ 246 ] It is generally accepted that de facto spouses in Quebec historically experienced disadvantageous treatment based on intentional prejudice. As I explained above, until the 1980 family law reform, disapproval of “concubinage” was expressed in legislation. This disapproval could be seen in particular in the Civil Code of Lower Canada , which effectively prohibited any financial arrangements between de facto spouses and also maintained a strict distinction between legitimate children, those born in wedlock, and “natural, incestuous or adulterine” children, those born out of wedlock. These measures were an attempt to prevent individuals from choosing to live in such unions, which at the time were considered [ translation ] “contrary to good morals”: Cossette, at p. 53. Moreover, the only way for de facto spouses to legitimate their natural children was to get married, where such a marriage was legally possible.

[ 247 ] McLachlin J. described the historical situation of de facto spouses as follows in Miron :

There is ample evidence that unmarried partners have often suffered social disadvantage and prejudice. Historically in our society, the unmarried partner has been regarded as less worthy than the married partner . The disadvantages inflicted on the unmarried partner have ranged from social ostracism through denial of status and benefits. [Emphasis added; para. 152.]

(See also Walsh , at para. 41 .)

[ 248 ] Thus, there was a period of Quebec history during which de facto spouses were subjected to both legislative hostility and social ostracism. In keeping with the principles stated in Law , this historical disadvantage means that particular care must be taken in reviewing any new legislation containing an adverse distinction with respect to certain members of the same group: paras. 63‑68. However, the recognition of historical disadvantage does not automatically lead to the conclusion that a distinction remains discriminatory today on the basis that it creates a disadvantage by perpetuating prejudice. It is also necessary that the disadvantage continue to exist today and that the new legislation confirm or perpetuate the legislative and social stigmatization of such spouses.

[ 249 ] On this point, I would point out that the limits on freedom of contract imposed on de facto spouses, like the distinction between legitimate and natural children, were eliminated in 1980. Furthermore, nothing in the evidence suggests that de facto spouses are now subject to public opprobrium or that they are otherwise subject to social ostracism. The expert reports filed by the parties tend to show the contrary. According to them, the de facto union has become a respected type of conjugality and is not judged unfavourably by Quebec society as a whole. Statistics from the latest census in 2011 confirm that it exists and is developing in Quebec society. Several of the married spouses with whom certain experts met in conducting sociological studies for the purposes of this litigation even regularly use language specific to de facto unions when describing themselves to third parties. The fact that they do so does not support a view that society continues to be suspicious or intolerant of de facto unions.

[ 250 ] Moreover, the legislature’s traditional hostility generally seems to have changed into acceptance of the de facto union. As I mentioned above, Quebec social legislation no longer draws distinctions between the various types of conjugality: see, inter alia , An Act respecting financial assistance for education expenses ; An Act respecting legal aid and the provision of certain other legal services ; Automobile Insurance Act . Rather, it applies uniformly to de facto , married and civil union spouses both in granting benefits and in imposing obligations where their relations with government institutions are concerned. As we have seen, the distinction continues to exist in the context of relations between the spouses themselves, within their conjugal relationship, where there is still a will to preserve the possibility of choosing between various types of conjugality.

[ 251 ] However, the fact that the form of union chosen by de facto spouses now generally seems to have the social and legislative acceptance it formerly lacked does not on its own lead to the conclusion that the law at issue in this case does not express prejudice. Prejudice does not have to be intentional for s. 15(1) of the Charter to be violated. If a law has the effect of favouring certain individuals at the expense of others by treating the latter as less worthy, or if it creates a hierarchy between them, even inadvertently, it will be considered to discriminate by expressing prejudice.

[ 252 ] In the case at bar, the articles of the Civil Code of Québec whose validity is challenged by A have the purpose and effect of regulating the private relationships of married spouses. They do this in two ways, as I explained above.

[ 253 ] First, as a result of their marriage, the spouses are subject to a mandatory primary regime that radically alters each spouse’s patrimonial rights. More specifically, the primary regime results in the formation of a partial economic union between the spouses. Second, where there is no marriage contract providing for separation as to property or for changes to the legal regime, the legal matrimonial regime of partnership of acquests also applies to the spouses as a result of their marriage. Like the primary regime, the regime of partnership of acquests significantly changes the rights of both spouses in relation to their patrimony. It expands the partial economic union already created by the primary regime. Upon separation or divorce, this matrimonial regime requires that the value of each spouse’s acquests be partitioned. The monetary impact of this regime can be significant: each spouse’s acquests include the proceeds of his or her work during the regime, and the fruits and income due or collected from all his or her private property or acquests during the regime.

[ 254 ] The Quebec legislature has imposed these regimes only on those who, by agreement with another person, have demonstrated that they wish to adhere to them. Their consent must be explicit, and must take the form of marriage or a civil union. The legislature did not view cohabitation on its own as an expression of such consent. As a result, the regimes do not apply automatically to de facto unions.

[ 255 ] Has the legislature in this way established a hierarchy between the various forms of conjugality? Has it expressed a preference for marriage and civil unions at the expense of de facto unions? Do the articles of the Civil Code of Québec being challenged in this case have the effect of sending a message or conveying a negative image or belief concerning de facto spouses?

[ 256 ] In my opinion, the answer to these questions must be no. Like the Nova Scotia legislature in enacting the MPA at issue in Walsh , the Quebec National Assembly has not favoured one form of union over another. This conclusion can be inferred if the questions of freedom of choice and autonomy of the will of the parties are correctly considered. The legislature has merely defined the legal content of the different forms of conjugal relationships. It has made consent the key to changing the spouses’ mutual patrimonial relationship. In this way, it has preserved the freedom of those who wish to organize their patrimonial relationships outside the mandatory statutory framework.

[ 257 ] This makes it easier to understand the real purpose of these appeals. It concerns the mutual rights and obligations of spouses in the various forms of conjugal relationships available to them in Quebec law. In Quebec family law, these rights and obligations are always available to everyone, but imposed on no one. Their application depends on an express mutual will of the spouses to bind themselves. This express, and not deemed, consent is the source of the obligation of support and of that of partition of spouses’ patrimonial interests. As we have seen, this consent is given in Quebec law by contracting marriage or a civil union, or entering into a cohabitation agreement. Participation in the protective regimes provided for by law depends necessarily on mutual consent. In this regard, the conclusion of a cohabitation agreement enables de facto spouses to create for themselves the legal relationship they consider necessary without having to modify the form of conjugality they have chosen for their life together.

[ 258 ] M. v. H. does not preclude this conclusion. In that case, the same‑sex spouses were excluded on the basis of their sexual orientation from a regime that required no form of express consent in order to apply. In the case of Quebec, consent is the ordinary rule. For de facto spouses, it can take legal forms that are well known and often straightforward, with which legal practitioners, if not the parties themselves, are familiar. If a spouse has concerns about the relative fragility of such agreements in the event of insolvency, agreements with respect to forms of co‑ownership are possible.

[ 259 ] In this context in which the existence of a set of rights and obligations depends on mutual consent in one of a variety of forms, it is hard to speak of discrimination against de facto spouses. In this regard, my colleague Abella J. is mistaken in saying that the choice of a form of conjugal relationship, especially that of marriage, is necessarily a mutual one. She criticizes what she describes as the “opt‑in” system of Quebec law, arguing that it leaves each member of the couple at the mercy of the other should one of them refuse to marry: para. 375.

[ 260 ] This represents an incomplete view of marriage and de facto unions in contemporary society. If we accept that an individual’s freedom to decide and personal autonomy are not purely illusory, his or her decision to continue living with a spouse who refuses to marry has the same value as that of a spouse who gives in to insistent demands to marry.

[ 261 ] In Quebec family law, as I mentioned above, choosing a de facto union permits spouses to opt out of the primary regime that is mandatory in the case of marriage or a civil union. By making this choice, they avoid entering into that regime and consequently assuming such obligations as that of support or the partition of the family patrimony. My colleague Abella J. adopts a position that would require these spouses to perform positive acts to opt out of a regime they did not intend to adopt. She would thus require them to exercise a freedom of choice whose validity and relevance she nonetheless denies in the context of opting for a particular form of conjugality.

[ 262 ] Moreover, the entire history of societal and legal changes that have led to the de facto union becoming an accepted form of conjugality in Quebec, far from being irrelevant to the analysis of an allegation that the right to equality has been infringed, is essential if we are to understand the constitutional issue before us and consider it in its context. This context can be understood only in light of the very widespread acceptance of the de facto union in Quebec society since the recognition of marital status as an analogous ground of discrimination: Turpin , at p. 1332. The resulting choice has become a key factor in the determination of the scope of the right at issue, and not only in the justification of a limit on that right: Lavoie , at paras. 47‑48 , per Bastarache J.

[ 263 ] That the legislature did not intend to favour marriage or the civil union is made even clearer by the fact that in Quebec law, de facto spouses may accept each of the effects of marriage set out in the impugned provisions. To do so, they must indicate their consent by expressing a clear intention.

[ 264 ] Thus, de facto spouses can enter into cohabitation agreements in which they can opt in to the rules on the partition of the family patrimony, provide for an obligation of support in the event that their relationship breaks down, grant a right of use of the family residence or reserve the right to claim a compensatory allowance. They can also establish the equivalent of a matrimonial regime and agree to have the rules of the regime of partnership of acquests apply to their relationship. Moreover, under statutory pension plans, de facto spouses can apply by mutual agreement for the partition of amounts accrued in one spouse’s name. Also, spouses who do not want to proceed by way of an agreement can purchase property as co‑owners, thus ensuring that they are entitled to partition the property when their union comes to an end. In the absence of such arrangements, a de facto spouse can always go to court to bring an action for unjust enrichment against his or her former spouse in circumstances in which such an action is available.

[ 265 ] To paraphrase Bastarache J. in Walsh , I believe that this requirement that a consensus between the spouses exist before any significant change is made to their rights of ownership “enhances rather than diminishes respect for the autonomy and self‑determination of unmarried cohabitants and their ability to live in relationships of their own design”: para. 50.

[ 266 ] In short, I do not consider it imperative that there be an identical framework for each form of union in order to remain true to the purpose of s. 15(1). In the instant case, the fact that there are different frameworks for private relationships between spouses does not indicate that prejudice is being expressed or perpetuated, but, rather, connotes respect for the various conceptions of conjugality. Thus, no hierarchy of worth is established between the different types of couples. Conversely, as we saw in Miron , differential treatment of spouses by the government in conferring financial benefits may be a sign that prejudice is being expressed against certain forms of conjugality. On this fundamental distinction, I will repeat the comments of Professor Goubau, who discusses the importance of maintaining the diversity of forms of conjugality in private law and refraining from imposing a single form of conjugal relationship on spouses:

[ translation ] Moreover, our reflection on the appropriateness of extending certain rights traditionally reserved for marriage (and now extended to civil unions) to de facto spouses must not confuse, as is done too often, the recognition of de facto spouses in social law with the idea that they should be treated in the same way as other spouses in private law, which, in my view, is the very negation of recognition. In public law and social law, the recognition of conjugality outside marriage, both heterosexual and homosexual, is now a fact. . . . To accept that unmarried couples can, for example, enjoy the benefits of a public pension plan just like married couples or have the same tax disadvantages is in fact to recognize that the private choices made by individuals have nothing to do with their status in society. Making the status of spouse available to all couples under social and public law regardless of the legal forms of their relationships is therefore a genuine way of recognizing the real diversity of conjugality in contemporary society.

On the other hand, to treat all couples in the same way when it comes, for example, to the obligation of support or the partition of the family patrimony, that is, to matters of private law, would amount to denying precisely what creates diversity among couples, namely the voluntary arrangement of the private effects of their conjugality. [Emphasis in original.]

(D. Goubau, “La conjugalité en droit privé: comment concilier ‘autonomie’ et ‘protection’?”, in Lafond and Lefebvre, 153, at p. 156)

[ 267 ] I conclude on completing this part of the analysis that the articles of the Civil Code of Québec whose constitutional validity is being challenged by A do not express or perpetuate prejudice against de facto spouses. On the contrary, it appears that, by respecting personal autonomy and the freedom of de facto spouses to organize their relationships on the basis of their needs, those provisions are consistent with two of the values underlying s. 15(1) of the Charter . They were enacted as part of a long and complex legislative process during which the Quebec National Assembly was concerned about keeping step with changes in society and about adapting family law to new types of conjugal relationships in a manner compatible with the freedom of spouses.

[ 268 ] At this point of my analysis, I must mention my reservations with respect to the position taken by my colleague Abella J. First of all, she does not recognize the role played by consent in the application of the rights and obligations that result from the various forms of conjugality. And it is odd that the opt-out solution she proposes for parties living in a de facto union would itself depend on this mutuality of consent and would not be available to parties who have chosen other forms of conjugal relationships. Next, she fails in practice to consider the social context of the de facto union in Quebec. Finally, her analysis would tend to reduce the review of alleged infringements of the right to equality to a requirement that adverse distinctions be found. There would no longer be an analytical framework to guide the courts in considering such matters, and this could affect the legitimacy of their decisions in this regard.

(3) Stereotyping

[ 269 ] There is a second way for A to prove substantive inequality, however. She can try to show that the disadvantage imposed by these legislative provisions is based on a stereotype that does not correspond to the actual circumstances or characteristics of de facto spouses: Withler , at para. 36 .

[ 270 ] This argument essentially focuses on the issue of the validity of the basic premise of Quebec family law, namely the exercise of autonomy of the will. In Quebec, the legislature has provided that the application of the provisions on the family patrimony, the compensatory allowance, the obligation of support, the family residence and partnership of acquests are conditional on the expression of a consensual intention (through marriage, a civil union or a cohabitation agreement). It thus refused to impose these measures on persons who have not expressed their consent to be bound by them and, in so doing, took the view that cohabitation alone does not amount to the expression of an intention to be so bound.

[ 271 ] If this premise is false and the decision to marry or not to marry does not imply consent to be bound by or excluded from the regimes in the Civil Code , the provisions challenged by A could well be based on an inaccurate characterization of the circumstances of Quebec couples. In other words, if autonomy of the will is merely wishful thinking and does not really exist in matrimonial matters, then the distinction made by the legislature does not correspond to the actual circumstances and characteristics of de facto spouses and therefore creates a disadvantage based on stereotyping.

[ 272 ] There is no evidence in the Court’s record in this case that would justify accepting the validity of such an argument and finding that the exclusion of de facto spouses from the primary regime and the regime of partnership of acquests is based on a stereotypical characterization of the actual circumstances of such spouses. More specifically, none of A’s evidence tends to show that the policy of freedom of choice, consensualism and autonomy of the will does not correspond to the reality of the persons in question.

[ 273 ] Nor can I take judicial notice of the fact that the choice of type of conjugality is not a deliberate and genuine choice that should have patrimonial consequences but necessarily results from the spouses’ ignorance of the consequences of their status. Such a fact is clearly controversial and not beyond reasonable dispute: Find , at paras. 48 and 60-61 .

[ 274 ] It is not unreasonable to believe that, in theory, individuals sometimes make uninformed choices and that some individuals may be unaware of the consequences of their choice of conjugal lifestyle. Nevertheless, to take judicial notice of the fact that the voluntary choice not to marry does not reflect an autonomous decision to avoid the legal regimes would be to exceed the limits of legitimate judicial notice, especially in relation to an issue at the centre of the controversy.

[ 275 ] In the case at bar, A has not established that it is stereotypical to believe that couples in a de facto union have chosen not to be bound by the regimes applicable to marriage and civil unions. The Quebec scheme, the effect of which is to respect each person’s freedom of choice to establish his or her own form of conjugality, and thus to participate or not to participate in the legislative regime of marriage or civil union with its distinct legal consequences, is not based on a stereotype.

[ 276 ] In this sense, recognition of the principle of autonomy of the will, which is one of the values underlying the equality guarantee in s. 15 of the Charter , means that the courts must respect choices made by individuals in the exercise of that autonomy. In this context, it will be up to the legislature to intervene if it believes that the consequences of such autonomous choices give rise to social problems that need to be remedied.

[ 277 ] In Quebec, the current legal framework for marriage and the other forms of conjugal relationships developed as a result of this type of intervention by the legislature, which intended to remedy problems caused by the evolution of marriage in Quebec and the longstanding preference of couples for the regime of separation of property over community matrimonial regimes.

[ 278 ] In the instant case, in the absence of an infringement of s. 15(1) of the Charter , the Court has no power to impose on the private relationships of de facto spouses a legal framework based on a social policy that differs from the policy adopted by the Quebec legislature. Only the legislature can intervene to change that legislative policy and remedy any problems encountered by de facto spouses.

[ 279 ] This type of legislative intervention has in fact occurred in certain provinces. Provincial legislatures have chosen to regulate the private relationships of common law spouses on the basis of their own provinces’ legislative objectives. Today, each province defines the effects of de facto unions or common law relationships differently, which is a mark of Canadian legal pluralism.

[ 280 ] For example, in all provinces except Quebec, and in the territories, cohabitation for a certain number of years gives rise to an obligation of support between common law spouses: see, inter alia , Family Law Act , R.S.O. 1990, c. F.3 ; Family Services Act , S.N.B. 1980, c. F‑2.2; The Family Maintenance Act , R.S.M. 1987, c. F20; Maintenance and Custody Act , R.S.N.S. 1989, c. 160; Family Relations Act , R.S.B.C. 1996, c. 128; Family Law Act , R.S.N.L. 1990, c. F‑2; The Family Maintenance Act, 1997 , S.S. 1997, c. F‑6.2; Family Law Act , R.S.P.E.I. 1988, c. F‑2.1; Family Law Act , S.N.W.T. 1997, c. 18; Domestic Relations Act , R.S.A. 2000, c. D‑14. Some provinces, such as Ontario, have imposed this policy to alleviate the burden on the public purse: see, inter alia , W. Holland, “Intimate Relationships in the New Millennium: The Assimilation of Marriage and Cohabitation?” (2000), 17 Can. J. Fam. L. 114, at p. 128. In British Columbia, in addition to the obligation of support, certain measures to protect the family residence apply to common law spouses: Family Relations Act . In Saskatchewan and Manitoba, common law relationships are, in addition to being subject to a support obligation and measures related to the family residence, subject to the division of family property: The Miscellaneous Statutes (Domestic Relations) Amendment Act, 2001 (No. 2) , S.S. 2001, c. 51; Common Law Partners’ Property and Related Amendments Act , S.M. 2002, c. 48. As we saw above, Nova Scotia’s legislation provides that common law partners can choose to register their partnerships and thus be governed by the legal framework applicable to marriage with respect to matrimonial property: Law Reform 2000 Act , S.N.S. 2000, c. 29.

(4) Conclusion

[ 281 ] I therefore conclude that, although arts. 401 to 430, 432, 433, 448 to 484 and 585 C.C.Q. draw a distinction based on marital status between de facto spouses and married or civil union spouses, they do not create a disadvantage by expressing or perpetuating prejudice or by stereotyping. The exclusion of de facto spouses from the scope of those provisions is not discriminatory within the meaning of s. 15(1) of the Charter and does not violate the constitutional right to equality. As a result, it is not necessary to proceed to the s. 1 stage of the Charter analysis.

V. Disposition

[ 282 ] I would allow the appeals of the Attorney General of Quebec and B and dismiss A’s appeal, without costs in all cases. I would answer the constitutional questions as follows:

1. Do arts. 401 to 430, 432, 433, 448 to 484 and 585 of the Civil Code of Québec , S.Q. 1991, c. 64, infringe s. 15(1) of the Canadian Charter of Rights and Freedoms ?

No.

2. If so, is the infringement a reasonable limit prescribed by law that can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

It is not necessary to answer this question.

The following are the reasons delivered by

[ 283 ] Abella J. (dissenting in result) — When spouses who are married or in civil unions separate or divorce in Quebec, they are guaranteed certain legal protections. They have the right to claim support from each other and an equal division of the family property. During separation, their use of the family home and household effects is also protected. These legal protections are not contractual by nature; they are statutorily imposed either presumptively or mandatorily. The spousal support and family property provisions in Quebec are aimed at recognizing and compensating for the roles assumed within the relationship and any resulting dependence and vulnerability on its dissolution.

[ 284 ] Many de facto spouses — the term used in Quebec for those who are neither married nor in a civil union — share the characteristics that led to these protections. They form long-standing relationships; they divide household responsibilities and develop a high degree of interdependence; and, critically, the economically dependent — and therefore vulnerable — spouse is faced with the same disadvantages when the union is dissolved. Yet de facto dependent spouses in Quebec can claim none of the economic protections that are available to those in marriages or civil unions. They have no rights or obligations towards each other under the Civil Code of Québec , S.Q. 1991, c. 64 (“ C.C.Q. ”). They are entitled to enter into a cohabitation contract to govern their obligations, and can claim child support where appropriate, but they have no right to claim spousal support, no right to divide the family patrimony, and are not governed by any matrimonial regime.

[ 285 ] The question before us is whether dependent de facto spouses in Quebec should be denied access to fundamental legal protections simply because their spousal relationship lacks the formality of a civil union or marriage. In my respectful view, the total exclusion of de facto spouses from the legal protections for both support and property given to spouses in formal unions is a violation of s. 15(1) of the Canadian Charter of Rights and Freedoms and is not justified under s. 1 . As the history of modern family law demonstrates, fairness requires that we look at the content of the relationship’s social package, not at how it is wrapped.

Facts

[ 286 ] Mr. B and Ms. A met in Ms. A’s home country in 1992. At the time, she was 17 years old, while Mr. B was 32 and the owner of a lucrative business. The couple saw each other sporadically until 1995, travelling around the world together at times. In early 1995, Ms. A came to live in Canada. They broke up soon after, but saw each other over Christmas and in early 1996, when Ms. A became pregnant with their first child. She gave birth to two other children with Mr. B, in 1999 and 2001.

[ 287 ] The couple discussed marriage on at least two occasions. In 1996, Ms. A asked Mr. B to marry her, but he refused, on the basis that he did not believe in the institution of marriage. He said that he could possibly envision getting married someday, but only to make a long-standing relationship official. On January 1, 2000, the topic came up again, though the parties presented different accounts as to whether they agreed to marry. In any event, neither Ms. A nor Mr. B followed up on these plans.

[ 288 ] In 2001, the parties discussed separating. They agreed to live together for six months, in an attempt to reconcile, but in October 2001, Ms. A ended the relationship. By the time they stopped living under the same roof in 2002, they had cohabited for seven years. During the relationship, Ms. A had attempted to start a career as a model, but she largely did not work outside of the home and often accompanied Mr. B on his travels.

[ 289 ] In February 2002, Ms. A began proceedings seeking custody of the children, spousal support, a lump sum support payment, and use of the family home. She challenged the constitutionality of certain provisions of the Civil Code , claiming access to the same protections as married spouses with respect to support, the family patrimony, the presumptive partnership of acquests regime, and the compensatory allowance. Her constitutional challenges were dismissed by the application judge in the Superior Court of Quebec. The Court of Appeal overturned the decision on support, concluding that the spousal support provision ( art. 585 C.C.Q. ) was unconstitutional. It considered itself bound, however, by this Court’s decision in Nova Scotia (Attorney General) v. Walsh , 2002 SCC 83, [2002] 4 S.C.R. 325, and therefore did not interfere with the property aspect of the decision.

[ 290 ] Ms. A appealed the Court of Appeal’s conclusion that the division of property provision was constitutional. Mr. B and the Attorney General of Quebec also appealed, on the conclusion that the spousal support provisions were unconstitutional.

Analysis

[ 291 ] The current legislative scheme in Quebec leaves an economically vulnerable spouse excluded from mandatory support and property division regimes simply because he or she was not in a formally created union. The issue in this appeal is whether this exclusion violates s. 15 of the Charter .

[ 292 ] It is helpful to review the reasons motivating the development of the modern approach to spousal support and family property regimes. While the ways in which these regimes have been implemented vary between provinces, the social rationale for the regimes is common across the country.

[ 293 ] In Quebec, spouses in marriages or civil unions have an obligation of support during the marriage ( art. 585 C.C.Q. ), which continues following separation, allowing a court to require one spouse to pay support payments to the other (arts. 507, 511 and 521.17 C.C.Q. ). Following a divorce, the obligation of support under the Code ends, and s. 15.2 of the federal Divorce Act , R.S.C. 1985, c. 3 (2nd Supp.), governs support orders.

[ 294 ] Article 585 of the Code has a clearly protective purpose, since it extends the maintenance obligation not only to married or civil union spouses, but also to “ relatives in the direct line in the first degree” such as children. The mechanism is adjustable, and support is determined based on “the needs and means of the parties, their circumstances and, as the case may be, the time needed by the creditor of support to acquire sufficient autonomy” (art. 587 C.C.Q. ; see Jean Pineau and Marie Pratte, La famille (2006), at pp. 133 and 782 ). Moreover, the provisions in the Code make it clear that the obligation of support is fundamental. It is of public order and cannot be renounced, ceded or alienated by the dependent spouse, since it is indispensable to his or her survival ( Québec (Procureure générale) v. B.T. , 2005 QCCA 748, [2005] R.D.F. 709 (C.A.)).

[ 295 ] In other words, Quebec explicitly subordinated a contractual theory of support to a protective one based on mutual obligation, since its law does not allow a couple in a formally recognized union to contract out of the Civil Code ’s mandatory support provision.

[ 296 ] Throughout Canada, provincial and federal law reform commissions were predominantly concerned with the impact of separation and divorce on the economically vulnerable spouse, usually the wife (see British Columbia’s Royal Commission on Family and Children’s Law, Family Maintenance (1975), at p. 7; Ontario’s Ministry of the Attorney General, Family Law Reform (1976), at p. 1; Manitoba Law Reform Commission, Reports on Family Law , Part I — The Support Obligation , Report #23 (1976), at p. 19). The Law Reform Commission of Canada concluded in its 1975 working paper 12, Maintenance on Divorce , that the right to support — and the obligation to pay it — did not rest on the legal status of either husband or wife, but on the reality of the dependence or vulnerability that the spousal relationship had created:

Financial rights and obligations based upon marriage should be legal results that follow from the internal arrangements made by the spouses in line with their priorities, circumstances and interests rather than being imposed according to traditional legal preconceptions of the sexually determined roles of each spouse. [Emphasis deleted; p. 17.]

[ 297 ] In other words, the right to support was not created by marriage per se , but by the “reasonable needs” of a vulnerable spouse ( Maintenance on Divorce , at p. 18). Its purpose was protective: “. . . to enable a former spouse who has incurred a financial disability as a result of marriage to become self-sufficient again in the shortest possible time” (p. 17).

[ 298 ] In elaborating on the rationale underlying spousal support in Moge v. Moge , [1992] 3 S.C.R. 813, this Court held that spousal maintenance “seeks to recognize and account for both the economic disadvantages incurred by the spouse who makes such sacrifices and the economic advantages conferred upon the other spouse” (p. 864). And in Bracklow v. Bracklow , [1999] 1 S.C.R. 420, McLachlin J. confirmed that the mutual obligation of support is protective: it stresses the interdependencies created by marriage, and “recognizes the reality that when people cohabit over a period of time in a family relationship, their affairs may become intermingled and impossible to disentangle neatly” (para. 31). Need is central to this theory, which, as discussed later in these reasons, is conceptually applicable as much to de facto relationships as to marriages and civil unions.

[ 299 ] Bracklow echoes the Law Reform Commission of Canada in noting that the spousal support obligation does not stem from “ the bare fact of marriage, so much as the relationship that is established and the expectations that may reasonably flow from it” (para. 44 (emphasis deleted)). Notably, it also rejects the paramountcy of the “clean-break” theory of support (para. 32), whereby each spouse is entitled at marriage breakdown to “what the individuals contracted for” (para. 29).

[ 300 ] A concern about the disproportionate number of women who experienced poverty when they separated was also at play in the development of the law. The goal of addressing this imbalance was clear in the work of the law reform commissions. It was also accepted by this Court in Moge , which noted that while support obligations were framed in gender-neutral terms, the reality is that “in many if not most marriages, the wife still remains the economically disadvantaged partner” (pp. 849-50). Justice L’Heureux-Dubé explained what spousal support was intended to remedy as follows:

Women have tended to suffer economic disadvantages and hardships from marriage or its breakdown because of the traditional division of labour within that institution. . . . [O]nce the marriage dissolves, the kinds of non‑monetary contributions made by the wife may result in significant market disabilities. The sacrifices she has made at home catch up with her and the balance shifts in favour of the husband who has remained in the work force and focused his attention outside the home. In effect, she is left with a diminished earning capacity and may have conferred upon her husband an embellished one. [Emphasis added; pp. 861-62.]

[ 301 ] The law dealing with division of family property has also come to be conceptualized in recent years on a protective basis rather than a contractual one. Quebec has legislation making the division of family property between married and civil union spouses equitable. Its legislation dealing with property division for those spouses states that during a marriage or civil union, or during a period of separation, spouses are prohibited from alienating or leasing certain property, including the family residence, without the other’s consent (arts. 401 and 404 C.C.Q. ). The Code also establishes the family patrimony, a core of family property that must in principle be shared equally by the spouses on divorce. The provisions creating the family patrimony are of public order, so they cannot be contracted out of, except after separation, divorce or the death of the other spouse (arts. 414 to 416, 419 and 423 C.C.Q. ).

[ 302 ] Except for the family patrimony, spouses in marriages or civil unions are able to choose the matrimonial regime governing their property. The default regime is the partnership of acquests, under which spouses each have control of their own property during the relationship, but most property acquired during the union is divided equally on its dissolution (art. 432 C.C.Q. ). Alternatively, spouses can elect the separation as to property regime, under which spouses hold their property separately both before and after the union (art. 486 C.C.Q. ), or a community regime, where property is controlled jointly during the union (art. 492 C.C.Q. ).

[ 303 ] Finally, on separation or divorce, spouses in marriages or civil unions can claim a compensatory allowance, a court-ordered payment that compensates one spouse for his or her contribution to the enrichment of the other (art. 427 C.C.Q. ).

[ 304 ] The legislative progression in Quebec from the “community” matrimonial regime, to the partnership of acquests, to the compensatory allowance, and finally to the family patrimony, has been elegantly unfolded by my colleague LeBel J. Two important strands from this history merit particular attention.

[ 305 ] The first is that the goal of better protecting economically vulnerable spouses can be seen as motivating each successive stage of reform in Quebec. With many spouses opting for separation of property under the initial community regime, as Prof. Alain Roy explains, [ translation ] “the breakdown of the conjugal relationship brought to light how very vulnerable wives were financially. . . . From this perspective, the legislature had to establish mechanisms to protect wives” (“Le régime législatif de l’union civile: entre symbolisme et anachronisme”), in Pierre-Claude Lafond and Brigitte Lefebvre, eds., L’union civile: nouveaux modèles de conjugalité et de parentalité au 21 e siècle (2003), 165, at p. 170).

[ 306 ] In recommending the partnership of acquests as the new default matrimonial regime, the Civil Code Revision Office in 1968 noted that the “freedom and independence” provided to those spouses who chose separation of property “sometimes proves extremely burdensome for one of the consorts and, in certain cases, even results in real injustice” ( Report on Matrimonial Regimes (1968), at p. 9). The subsequent compensatory allowance and family patrimony regimes targeted the same injustices, which had lingered in spite of the partnership of acquests (for judicial discussion of the respective goals of these regimes, see this Court’s M. (M.E.) v. L. (P.) , [1992] 1 S.C.R. 183, and Droit de la famille — 977 , [1991] R.J.Q. 904 (C.A.), at p. 908).

[ 307 ] The second strand is that, far from being designed to reflect the actual choices made by married spouses, these measures subordinated those choices to the agenda of protection. In crafting the partnership of acquests regime, which became the presumptive matrimonial regime for all married couples in 1970, the National Assembly abandoned wording that would have presumed that the equal sharing of property it created was an implicit choice by the couple. The Law Reform Commission of Canada applauded this decision, noting that “we cannot assume, in all cases, that the spouses have not made a marriage contract because they consider the legal regime to be the one which suits them best” ( Studies on Family Property Law , Research Paper: Matrimonial Regimes in Québec (1975), at p. 61). Significantly too, the subsequent provision on compensatory allowance was made part of public order, applying mandatorily to all married couples regardless of matrimonial regime. As a result, couples who had chosen the separation of property regime were largely denied their freedom to contract with respect to contributions made during the marriage.

[ 308 ] The same was true under the family patrimony regime, which was also made part of public order. In introducing the new legislation, Monique Gagnon-Tremblay, the Minister responsible for the Status of Women, while noting that a mandatory division of the family patrimony negates freedom of choice, nonetheless said that choice would have to defer to the more important social goal of remedying a legal and social barrier to equality:

[ translation ] It was necessary to rethink a legal and social mechanism that tends to reproduce inequality.

In a word, it seemed to us that to refuse to introduce a family patrimony on the basis that this new institution is incompatible with certain matrimonial regimes spouses might want to choose would amount to giving greater importance to legal models than to the imperatives of social change . Were it to do so, the legislature would deprive itself of an essential lever for social change. [Emphasis added.]

(National Assembly, Journal des débats , vol. 30, No. 125, 2nd Sess., 33rd Leg., June 8, 1989, at p. 6489)

The opposition critic at the time, Louise Harel, provided crucial support to the measure, noting specifically that it represented a move from formal to substantive equality within marriage:

[ translation ] In our society, for an entire generation, the institution of marriage represented something that was contrary to the principle of equality. Then a legal equality was introduced. And, Mr. President, I cannot place too much emphasis on the fact that, if we are going to work together to pass this bill, it is precisely because that formal legal equality is not enough. It is precisely to open a new legal path to the full economic and social equality of women . [Emphasis added.]

(National Assembly, Journal des débats , vol. 30, No. 125, 2nd Sess., 33rd Leg., June 8, 1989, p. 6497)

[ 309 ] Thus the mandatory nature of both the compensatory allowance and family patrimony regimes highlights the preeminent significance Quebec has accorded to concerns for the protection of vulnerable spouses over other values such as contractual freedom or choice.

[ 310 ] These concerns were harmoniously echoed across Canada. The case of Rathwell v. Rathwell , [1978] 2 S.C.R. 436, contains useful insight into the shift to a protective understanding of the division of family property. Dickson J. noted that matrimonial property disputes were “bedevilled by conflicting doctrine and a continuing struggle between the ‘justice and equity’ school . . . and the ‘intent’ school” (p. 442). Intent had ruled the day in the prior case of Murdoch v. Murdoch , [1975] 1 S.C.R. 423, which confirmed the century-old approach to matrimonial property: the wife had been denied any share in the property held in her husband’s name because of her inability to prove common intent to vest in her a beneficial interest in the property.

[ 311 ] In Rathwell , Dickson J. relied instead on the doctrine of constructive trust, which requires no common intent, to award Helen Rathwell a share of the matrimonial property. Common intent, Dickson J. explained, could rarely be found, and to look for it was to misapprehend the way most couples approach their relationship:

. . . the plain fact is that there rarely is agreement [on the disposition of matrimonial property in the event of divorce] because the parties do not turn their minds to the eventuality of separation and divorce.

. . . There is rarely implied agreement or common intention, apart from the general intention of building a life together. It is not in the nature of things for young married people to contemplate the break-up of their marriage and the division, in that event, of assets acquired by common effort during wedlock. [pp. 444 and 447-48]

The emergence of a constructive trust to resolve matrimonial property disputes, Dickson J. explained,

reflects a diminishing preoccupation with the formalities of real property law and individual property rights and the substitution of an attitude more in keeping with the realities of contemporary family life. . . . The state of legal title may merely reflect conformity with regulatory requirements . . .; it may, on the other hand, be a matter of utmost indifference to the spouses as to which name appears on the title, so long as happy marriage subsists . . . . The state of title may be entirely fortuitous . . . . [Emphasis added; p. 456.]

[ 312 ] This brings us to the status of unmarried spouses in Canada. Historically, they were stigmatized. The children of unmarried relationships, for example, were deemed “illegitimate” and unable to inherit on intestacy. But as social attitudes changed, so did the approaches of legislatures and courts, which came to accept conjugal relationships outside a formal marital framework.

[ 313 ] This change reflected an enhanced understanding of what constitutes a “family”. In a 1993 report in which it recommended the extension of both spousal support and division of property regimes to unmarried spouses, the Ontario Law Reform Commission noted that

throughout much of the [ Family Law Act , R.S.O. 1990, c. F.3], “family” is equated with “marriage”. . . . [T]he Act . . . provides little room for other forms of relationship that embody the fundamental elements of intimacy, mutual economic interdependence, and living together in a “close personal relationship that is of primary importance in both persons’ lives”, which we see as the essence of the concept of “family”.

( Report on the Rights and Responsibilities of Cohabitants under the Family Law Act (1993), at p. 1)

The Report found that in

common-law relationships, which embody virtually all the characteristics of marriage . . . the need to protect the interests of both parties and to ensure equality and fairness in the event of the breakdown of the relationship is the same as in marriages.

Common-law spouses pool their resources and make joint economic plans, they provide each other financial and emotional support, and they raise children. [pp. 2 and 27]

[ 314 ] This Court launched the possibility of an equitable division of property for unmarried spouses in Pettkus v. Becker , [1980] 2 S.C.R. 834, where it extended the availability of a constructive trust to the division of property between separated common law spouses. Significantly, Dickson J. found that there were no grounds for a distinction in property division between a marriage and a less formally recognized long-term relationship, and that being unmarried was no barrier to the claimant obtaining the fruits of her contribution to a common law partnership.

[ 315 ] Subsequently, in M. v. H. , [1999] 2 S.C.R. 3, this Court found that the exclusion of same-sex couples from statutory support benefits violated s. 15(1) of the Charter . In developing its analysis, the Court noted that the various features that characterize a conjugal relationship could be found in same-sex relationships. Those features could be “present in varying degrees and not all are necessary for the relationship to be found to be conjugal” (para. 59). A conjugal relationship, in other words, is not a binary question: married or unmarried, opposite-sex or same-sex, economically dependent or economically independent. This decision highlighted an increasing willingness to look past the relationship’s formal wrapping and into its content. It is the nature of the relationship that is paramount, not what it is called.

[ 316 ] As attitudes shifted and the functional similarity between many unmarried relationships and marriages was accepted, this Court expanded protection for unmarried spouses. In Miron v. Trudel , [1995] 2 S.C.R. 418, the Court found that marital status was an analogous ground under s. 15(1) of the Charter , because while in theory an individual is free to choose whether to marry, there are, in reality, a number of factors that may place the decision out of his or her control. McLachlin J. described the impediments to choice as including:

The law; the reluctance of one’s partner to marry; financial, religious or social constraints — these factors and others commonly function to prevent partners who otherwise operate as a family unit from formally marrying. In short, marital status often lies beyond the individual’s effective control. [para. 153]

[ 317 ] The recognition of marital status as an analogous ground was a recognition of the complex and mutual nature of the decision to marry and the myriad factors at play in that decision. It was also an acknowledgment that the decision to live together as unmarried spouses may not in fact be a choice at all.

[ 318 ] I would make one further observation about the narrative of the treatment of de facto or common law unions in Canada. As part of the acceptance of marital status as an analogous ground in Miron , McLachlin J. recognized that unmarried spouses have faced historical disadvantage stemming from societal prejudice. She acknowledged that this disadvantage has faded as attitudes have changed, but nonetheless concluded that “the historical disadvantage associated with this group cannot be denied” (para. 152), a significant reminder that there has rarely been, in our lifetime, a bright line demarcating the successful evolution of an historically disadvantaged group into a barrier-free reality. The fact that society appears to have attenuated overtly discriminatory attitudes it once held towards a group does not mean that there is no continuing discriminatory conduct, however benignly or unconsciously motivated.

[ 319 ] This is the history that animates our s. 15 inquiry. In R. v. Kapp , 2008 SCC 41, [2008] 2 S.C.R. 483, this Court reaffirmed its commitment to the test that was set out in Andrews v. Law Society of British Columbia , [1989] 1 S.C.R. 143, whereby s. 15 was seen as an anti-discrimination provision. Building on the human rights decisions in Ontario Human Rights Commission v. Simpsons-Sears Ltd. , [1985] 2 S.C.R. 536, at p. 551, and Canadian National Railway Co. v. Canada (Canadian Human Rights Commission) , [1987] 1 S.C.R. 1114 (“ Action Travail ”), McIntyre J., in Andrews , noted that

the main consideration must be the impact of the law on the individual or the group concerned. Recognizing that there will always be an infinite variety of personal characteristics, capacities, entitlements and merits among those subject to a law, there must be accorded, as nearly as may be possible, an equality of benefit and protection and no more of the restrictions, penalties or burdens imposed upon one than another. [Emphasis added; p. 165.]

He identified the purpose of the equality provision and anti-discrimination law in general, as being to eliminate the exclusionary barriers faced by individuals in the enumerated or analogous groups in gaining meaningful access to what is generally available.

[ 320 ] Andrews involved a British citizen who challenged the citizenship requirement for admission to the British Columbia bar under the Barristers and Solicitors Act , R.S.B.C. 1979, c. 26. He succeeded because the requirement was found to have drawn a distinction based on an enumerated ground — national origin — and because that distinction resulted in an additional burden on non-citizens, namely a delay of several years before they could become entitled to practise law. The government’s justification was evaluated under s. 1 of the Charter .

[ 321 ] McIntyre J.’s approach in Andrews had several important features. First, it stipulated that “[t]he analysis of discrimination . . . must take place within the context of the enumerated grounds and those analogous to them” (p. 180).

[ 322 ] Second, the words “without discrimination” require more than a mere distinction in the treatment given to different groups or individuals. Instead, McIntyre J. found that those words were a form of qualifier built into s. 15 which limits the distinctions forbidden by the section to “those which involve prejudice or disadvantage ” (p. 181 (emphasis added)). McIntyre J.’s definition of discrimination contains the following statement about what constitutes “disadvantage”:

. . . 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 . [Emphasis added; p. 174.]

[ 323 ] In sum, the claimant’s burden under the Andrews test is to show that the government has made a distinction based on an enumerated or analogous ground and that the distinction’s impact on the individual or group perpetuates disadvantage. If this has been demonstrated, the burden shifts to the government to justify the reasonableness of the distinction under s. 1 . As McIntyre J. explained, “any justification, any consideration of the reasonableness of the enactment; indeed, any consideration of factors which could justify the discrimination and support the constitutionality of the impugned enactment would take place under s. 1 ” (p. 182).

[ 324 ] Kapp , and later Withler v. Canada (Attorney General) , 2011 SCC 12, [2011] 1 S.C.R. 396, restated these principles as follows: (1) Does the law create a distinction based on an enumerated or analogous ground? (2) Does the distinction create a disadvantage by perpetuating prejudice or stereotyping? ( Kapp , at para. 17 ; Withler , at para. 30 ). As the Court stated in Withler :

The focus of the inquiry is on the actual impact of the impugned law, taking full account of social, political, economic and historical factors concerning the group. [para. 39]

[ 325 ] In referring to prejudice and stereotyping in the second step of the Kapp reformulation of the Andrews test, the Court was not purporting to create a new s. 15 test. Withler is clear that “[a]t the end of the day there is only one question : Does the challenged law violate the norm of substantive equality in s. 15(1) of the Charter ?” (para. 2 (emphasis added)). Prejudice and stereotyping are two of the indicia that may help answer that question; they are not discrete elements of the test which the claimant is obliged to demonstrate, as Professor Sophia Moreau explains:

Such a narrow interpretation will likely have the unfortunate effect of blinding us to other ways in which individuals and groups, that have suffered serious and long-standing disadvantage, can be discriminated against. This would include cases, for instance, that do not involve either overt prejudice or false stereotyping, but do involve oppression or unfair dominance of one group by another, or involve a denial to one group of goods that seem basic or necessary for full participation in Canadian society.

(“ R. v. Kapp : New Directions for Section 15” (2008-2009), 40 Ottawa L. Rev. 283, at p. 292)

[ 326 ] Prejudice is the holding of pejorative attitudes based on strongly held views about the appropriate capacities or limits of individuals or the groups of which they are a member. Stereotyping, like prejudice, is a disadvantaging attitude, but one that attributes characteristics to members of a group regardless of their actual capacities. Attitudes of prejudice and stereotyping can undoubtedly lead to discriminatory conduct, and discriminatory conduct in turn can reinforce these negative attitudes, since “the very exclusion of the disadvantaged group . . . fosters the belief, both within and outside the group, that the exclusion is the result of ‘natural’ forces, for example, that women ‘just can’t do the job’” ( Action Travail , at p. 1139). As Walter Tarnopolsky observed:

. . . it is the overt act and not the thought which is prohibited and, as a natural consequence thereof, in many cases action could be contrary to human rights legislation even in the absence of a discriminatory intent, if its effect is discriminatory. [Emphasis in original.]

( Discrimination and The Law in Canada (1982), at p. 86)

[ 327 ] We must be careful not to treat Kapp and Withler as establishing an additional requirement on s. 15 claimants to prove that a distinction will perpetuate prejudicial or stereotypical attitudes towards them. Such an approach improperly focuses attention on whether a discriminatory attitude exists, not a discriminatory impact, contrary to Andrews , Kapp and Withler . In explaining prejudice in Withler , the Court said: “Without attempting to limit the factors that may be useful in assessing a claim of discrimination, it can be said that where the discriminatory effect is said to be the perpetuation of disadvantage or prejudice, evidence that goes to establishing a claimant’s historical position of disadvantage or to demonstrating existing prejudice against the claimant group, as well as the nature of the interest that is affected, will be considered” (para. 38).

[ 328 ] It is the discriminatory conduct that s. 15 seeks to prevent, not the underlying attitude or motive, as Dickson C.J. explained in Action Travail :

It is not a question of whether this discrimination is motivated by an intentional desire to obstruct someone’s potential, or whether it is the accidental by-product of innocently motivated practices or systems. If the barrier is affecting certain groups in a disproportionately negative way, it is a signal that the practices that lead to this adverse impact may be discriminatory. [p. 1139, citing the Report of the Commission on Equality in Employment (1984).]

This was reiterated in Withler , where the Court said: “Whether the s. 15 analysis focuses on perpetuating disadvantage or stereotyping, the analysis involves looking at the circumstances of members of the group and the negative impact of the law on them” (para. 37 (emphasis added)).

[ 329 ] That was the lesson learned from the former “dignity” test from Law v. Canada (Minister of Employment and Immigration) , [1999] 1 S.C.R. 497, which required claimants to establish that the impugned law had “the effect of perpetuating or promoting the view that the individual is less capable, or less worthy of recognition or value as a human being or as a member of Canadian society” (para. 51 (emphasis added)). In Kapp , this Court recognized that “dignity” was an underlying objective of the whole Charter , not a discrete and additional component of the equality test that the claimant had the burden of proving:

. . . human dignity is an abstract and subjective notion that . . . cannot only become confusing and difficult to apply; it has also proven to be an additional burden on equality claimants . . . . [Emphasis in original; para. 22.]

Similarly, prejudice and stereotyping are neither separate elements of the Andrews test, nor categories into which a claim of discrimination must fit. A claimant need not prove that a law promotes negative attitudes , a largely unquantifiable burden.

[ 330 ] Requiring claimants, therefore, to prove that a distinction perpetuates negative attitudes about them imposes a largely irrelevant, not to mention ineffable burden.

[ 331 ] Kapp and Withler guide us, as a result, to a flexible and contextual inquiry into whether a distinction has the effect of perpetuating arbitrary disadvantage on the claimant because of his or her membership in an enumerated or analogous group. As Withler makes clear, the contextual factors will vary from case to case — there is no “rigid template”:

The particular contextual factors relevant to the substantive equality inquiry at the second step [of the Andrews test] will vary with the nature of the case. A rigid template risks consideration of irrelevant matters on the one hand, or overlooking relevant considerations on the other : Kapp . Factors such as those developed in Law — pre-existing disadvantage, correspondence with actual characteristics, impact on other groups and the nature of the interest affected — may be helpful. However, they need not be expressly canvassed in every case in order to fully and properly determine whether a particular distinction is discriminatory . . . . [Emphasis added; para. 66.]

[ 332 ] The root of s. 15 is our awareness that certain groups have been historically discriminated against, and that the perpetuation of such discrimination should be curtailed. If the state conduct widens the gap between the historically disadvantaged group and the rest of society rather than narrowing it, then it is discriminatory. As the U.S. Supreme Court warned in Griggs v. Duke Power Co. , 401 U.S. 424 (1971):

. . . practices, procedures, or tests neutral on their face, and even neutral in terms of intent, cannot be maintained if they operate to “freeze” the status quo of prior discriminatory employment practices. [p. 430]

[ 333 ] An emphasis at this stage on whether the claimant group’s exclusion was well motivated or reasonable is inconsistent with this substantive equality approach to s. 15(1) since it redirects the analysis from the impact of the distinction on the affected individual or group to the legislature’s intent or purpose . As McIntyre J. warned in Andrews , an approach to s. 15(1) based on assessing the “reasonableness” of the legislative distinction would be a “radical departure from the analytical approach to the Charter ”, under which “virtually no role would be left for s. 1 ” (pp. 181-82). It would also effectively turn the s. 15(1) analysis into a review of whether the legislature had a “rational basis” for excluding a group from a statutory benefit. This reduces the test for discrimination to “a prohibition on intentional discrimination based on irrational stereotyping” (Sheila McIntyre, “Deference and Dominance: Equality Without Substance”, in Sheila McIntyre and Sanda Rodgers, eds., Diminishing Returns: Inequality and the Canadian Charter of Rights and Freedoms (2006), 95, at p. 104). Assessment of legislative purpose is an important part of a Charter analysis, but it is conducted under s. 1 once the burden has shifted to the state to justify the reasonableness of the infringement.

[ 334 ] This crucial distinction between the s. 15 analysis and the s. 1 justificatory step of the equality test brings us to another legal issue of particular importance in this case: the proper stage in the analysis to address the effect of the choice not to marry. In Miron , the fact that marital status is not a real choice was the basis for designating marital status as an analogous ground under s. 15(1) . McLachlin J. accepted that the choice to marry is constrained by a number of factors. Her reasons, already briefly referred to, bear more fulsome repetition:

In theory, the individual is free to choose whether to marry or not to marry. In practice, however, the reality may be otherwise. The sanction of the union by the state through civil marriage cannot always be obtained. The law; the reluctance of one’s partner to marry; financial, religious or social constraints — these factors and others commonly function to prevent partners who otherwise operate as a family unit from formally marrying. In short, marital status often lies beyond the individual’s effective control. In this respect, marital status is not unlike citizenship, recognized as an analogous ground in Andrews : the individual exercises limited but not exclusive control over the designation. [para. 153]

[ 335 ] Any discussion of the reasonableness of distinctions based on this ground, or justifications for such distinctions, must take place under s. 1 . To focus on the “choice” to marry at the s. 15(1) stage is not only contrary to the approach in Andrews , it is completely inconsistent with Miron and undermines the recognition of marital status as an analogous ground . By definition, analogous grounds are “personal characteristic[s] that [are] immutable or changeable only at unacceptable cost to personal identity” ( Corbiere v. Canada (Minister of Indian and Northern Affairs) , [1999] 2 S.C.R. 203, at para. 13 ). This Court has firmly rejected the context-dependency of analogous grounds: they are not deemed immutable in some legislative contexts and a matter of choice in others. Rather, they stand as “constant marker[s] of potential legislative discrimination” ( Corbiere , at para. 10 ). Having accepted marital status as an analogous ground, it is contradictory to find not only that de facto spouses do have a choice about their marital status, but that it is that very choice that excludes them from the protection of s. 15(1) to which Miron said they were entitled.

[ 336 ] Moreover, this Court has repeatedly rejected arguments that choice protects a distinction from a finding of discrimination. In Brooks v. Canada Safeway Ltd. , [1989] 1 S.C.R. 1219, the employer argued that a different amount of compensation for women who took time off from work while pregnant was not discriminatory because “pregnancy is a voluntary state and, like other forms of voluntary leave, it should not be compensated” (p. 1236). Dickson C.J. refused to accept that pregnancy was a choice, noting that an emphasis on choice would be “against one of the purposes of anti-discrimination legislation . . . the removal of unfair disadvantages which have been imposed on individuals or groups in society” (p. 1238). In other words, not only was pregnancy not a “true choice”, but choice was irrelevant to the question of discrimination.

[ 337 ] In Lavoie v. Canada , 2002 SCC 23, [2002] 1 S.C.R. 769, the Court was faced with a question of discrimination on the grounds of citizenship. The claimants challenged a provision of the Public Service Employment Act , R.S.C. 1985, c. P-33, that gave the Public Service Commission the discretion to prefer Canadian citizens in open competitions for employment. Bastarache J., for the majority, expressly rejected the argument, relied on by Arbour J. in her separate reasons, that the claimants could have chosen to obtain Canadian citizenship. In their own reasons, which agreed with Bastarache J. on this point, McLachlin C.J. and L’Heureux‑Dubé J. were even clearer in rejecting choice as justifying discriminatory treatment:

. . . the fact that a person could avoid discrimination by modifying his or her behaviour does not negate the discriminatory effect. If it were otherwise, an employer who denied women employment in his factory on the ground that he did not wish to establish female changing facilities could contend that the real cause of the discriminatory effect is the woman’s “choice” not to use men’s changing facilities. The very act of forcing some people to make such a choice violates human dignity, and is therefore inherently discriminatory. The law of discrimination thus far has not required applicants to demonstrate that they could not have avoided the discriminatory effect in order to establish a denial of equality under s. 15(1). [para. 5]

[ 338 ] Having reviewed the distinct analytical approaches to the s. 15(1) analysis and s. 1 justificatory step and the significance of this Court’s finding that marital status is an analogous ground, I turn to the final legal issue that commands the Court’s attention: the applicability of this Court’s decision in Walsh . Because the equality analysis under s. 15(1) of the Charter has evolved substantially in the decade since Walsh was decided, I would, with respect, decline to follow Walsh . Two aspects of the majority’s decision in Walsh are, in fact, manifestly contrary to the substantive equality analysis developed in Kapp and Withler , namely its approach to the issue of choice and its reliance on the heterogeneity of common law relationships.

[ 339 ] Walsh , similarly to the case before us, considered the role of freedom of choice in a s. 15 application dealing with the exclusion of common law spouses from a family property regime. The majority judgment in Walsh found that “people who marry can be said to freely accept mutual rights and obligations” while common law spouses cannot. In turn, it found that common law spouses “are free to take steps to deal with their personal property” privately (para. 55). Walsh was determinatively applied by the trial judge in the case before us, and in part by the Court of Appeal.

[ 340 ] As noted, in Walsh freedom of choice was key to the s. 15(1) analysis. Although the Walsh majority accepted that some common law spouses would suffer disadvantage under the Matrimonial Property Act , R.S.N.S. 1989, c. 275, it held that because they could choose to marry, their exclusion from the legislative scheme was not an issue that fell within the protection of s. 15(1):

While there is no denying that inequities may exist in certain unmarried cohabiting relationships and that those inequities may result in unfairness between the parties on relationship breakdown, there is no constitutional requirement that the state extend the protections of the MPA to those persons. The issue here is whether making a meaningful choice matters, and whether unmarried persons are prevented from taking advantage of the benefits of the MPA in an unconstitutional way. [Emphasis added; para. 57.]

The majority in Walsh accepted that marital status is an analogous ground, but justified distinctions within this ground by pointing to an individual’s “choice” to marry. This contradicts the approach to substantive equality under s. 15(1), where any argument concerning the reasonableness of the legislation is considered under s. 1 . Contrary to this approach, the majority of the Court in Walsh collapsed the justification into the s. 15 analysis, leaving the claimants to justify what should analytically have been part of the government’s burden.

[ 341 ] The majority in Walsh went on to find that despite the adverse impact suffered by some unmarried spouses under the Matrimonial Property Act , the claimant had failed to satisfy the dignity test. Again, the majority emphasized the importance of the claimant’s choice, concluding that a legislative regime that respected the personal autonomy and freedom of choice of an individual enhanced rather than detracted from their dignity. This reliance on dignity is the analytic approach this Court eschewed in Kapp when it dropped “dignity” as a required component in the s. 15(1) analysis because it had become an undue evidentiary burden for claimants.

[ 342 ] In Walsh , the majority’s focus on choice rather than on the impact of the distinction on members of the group also paid insufficient attention to the requirement for a true substantive equality analysis, affirmed in Kapp and Withler . In contrast to formal equality, which assumes an “autonomous, self-interested and self-determined” individual, substantive equality looks not only at the choices that are available to individuals, but at “the social and economic environments in which [they] pla[y] out” (Margot Young, “Unequal to the Task: ‘Kapp’ing the Substantive Potential of Section 15”, in Sanda Rodgers and Sheila McIntyre, eds., The Supreme Court of Canada and Social Justice: Commitment, Retrenchment or Retreat (2010), 183, at pp. 190-91 and 196).

[ 343 ] This is not to suggest that the issue of choice is entirely irrelevant to a claim under s. 15(1). It may be an important factor in determining whether a ground of discrimination qualifies as an analogous ground. In addition, it may factor into the s. 1 analysis. Examining choice at the s. 1 stage instead of integrating it into the discrimination analysis as the majority did in Walsh properly places the onus on the government to justify the exclusion based on freedom of choice, rather than compromising the s. 15(1) analysis. It is not the claimant’s burden to disprove the legislative purpose for the exclusion, but the government’s to demonstrate it under s. 1 .

[ 344 ] Walsh is also at odds with the substantive equality analysis of Kapp and Withler in its emphasis on the heterogeneity of common law relationships . The Walsh majority found this heterogeneity to be a pertinent distinction between married spouses and unmarried spouses. The majority accepted that some common law spouses suffered adverse impact, but emphasized that “many persons in circumstances similar to those of the parties, that is, opposite sex individuals in conjugal relationships of some permanence, have chosen to avoid the institution of marriage and the legal consequences that flow from it” (para. 43). The majority relied on the fact that not all common law spouses suffered discrimination as a basis for rejecting the s. 15(1) claim.

[ 345 ] The importance the Walsh majority placed on the heterogeneity of unmarried relationships resulted from its use of the then operative comparator group analysis. The majority assessed the discrimination claim by comparing two groups: married heterosexual cohabitants and unmarried heterosexual cohabitants. Although the majority in Walsh found that the “functional similarities” between married and common law spouses may be substantial, it held that “it would be wrong to ignore the significant heterogeneity that exists within the claimant’s comparator group [i.e. unmarried heterosexual cohabitants]” (para. 39).

[ 346 ] The majority’s reasoning in Walsh illustrates the problems with comparator groups that the subsequent decision in Withler sought to address, namely that “a mirror comparator group analysis may fail to capture substantive inequality, may become a search for sameness, may shortcut the second stage of the substantive equality analysis, and may be difficult to apply” (para. 60). In Walsh , the fact that the comparator group of married spouses was not perfectly mirrored by the group of unmarried spouses, based on the heterogeneity of the latter, short-circuited the analysis of the actual adverse impact experienced by a significant proportion of unmarried spouses.

[ 347 ] For all these reasons, and with great respect, I think, unlike the Court of Appeal, we can appropriately proceed to the application of the s. 15(1) test in this case untethered from Walsh .

Application

[ 348 ] The first step in s. 15(1) is to identify the distinction at issue and determine whether it is based on an enumerated or analogous ground. This is easily demonstrated in this case. The exclusion of de facto spouses from the economic protections for formal spousal unions is a distinction based on marital status, an analogous ground.

[ 349 ] We must then consider whether the distinction is discriminatory. That it imposes a disadvantage is clear, in my view: the law excludes economically vulnerable and dependent de facto spouses from protections considered so fundamental to the welfare of vulnerable married or civil union spouses that one of those protections is presumptive, and the rest are of public order, explicitly overriding the couple’s freedom of contract or choice. The disadvantage this exclusion perpetuates is an historic one: it continues to deny de facto spouses access to economic remedies they have always been deprived of, remedies the National Assembly considered indispensable for the protection of married and civil union spouses.

[ 350 ] There is little doubt that some de facto couples are in relationships that are functionally similar to formally recognized spousal relationships. When introducing family law reforms in 1976, the Ontario Ministry of the Attorney General acknowledged that the functional characteristics of unmarried relationships justified some protection:

When a man and woman have been living together in a relationship of some permanence, their lives take on the same financial characteristics as a legally recognized marriage. Often the couple both contribute to household expenses. One may be just as dependent on the other for certain tasks as married persons are.

( Family Law Reform , at p. 18)

[ 351 ] On the same note, the British Columbia Law Institute commented that people in “relationship[s] that resembl[e] marriage may suffer economic prejudice when the relationship ends” and “are also in need of protection” ( Report on Recognition of Spousal and Family Status (1998), at p. 7). The Law Reform Commission of Nova Scotia noted that, due to the functional similarities, unmarried relationships “deserve to be treated similarly by the law” as marriages ( Final Report: Reform of the Law Dealing with Matrimonial Property in Nova Scotia (1997), at p. 21). And the Law Reform Commission of Saskatchewan tied these functional similarities to the goals of matrimonial property legislation, commenting that

[t]he realities of marriage, not the legal status it creates, justified [matrimonial property] legislation. If long-term common law relationships are functionally similar, the mechanism by which the status is created is less important than the fact that the status entails social expectations that are usually associated with marriage.

(Discussion Paper, “Common Law Relationships Under the Matrimonial Property Act ”, July 1997 (online), at p. 12)

[ 352 ] This understanding of the functional similarity of de facto unions to marriages, it should be stressed, is shared in Quebec, where the Civil Code Revision Office, in proposing changes to the regime governing de facto spouses in 1978, accepted these functional similarities, commenting that “ [d]e facto unions, though perhaps more tenuous, are often as stable as marriages” ( Report on the Québec Civil Code (1978), vol. II — Commentaries , t. 1, at p. 113).

[ 353 ] This Court has also readily recognized that some de facto spouses share the functional characteristics of those in formal marriages. In the context of spousal support obligations, Cory and Iacobucci JJ. held in M. v. H. that unmarried same-sex couples, unmarried opposite-sex couples, and married couples, can all fulfill the “generally accepted characteristics of a conjugal relationship [which] include shared shelter, sexual and personal behaviour, services, social activities, economic support and children, as well as the societal perception of the couple” (para. 59). And in Pettkus v. Becker , Dickson J. acknowledged that there was “no basis for any distinction, in dividing property and assets, between marital relationships and those more informal relationships which subsist for a lengthy period” and that the unmarried parties “lived as man and wife” (p. 850).

[ 354 ] Even if there is a range of need or vulnerability among de facto spouses, as there must inevitably be, this Court has held that heterogeneity within a claimant group does not defeat a claim of discrimination. In Janzen v. Platy Enterprises Ltd. , [1989] 1 S.C.R. 1252, Dickson C.J., as he had in Brooks , squarely rejected the idea that for a claim of discrimination to succeed, all members of a group had to receive uniform treatment from the impugned law:

While the concept of discrimination is rooted in the notion of treating an individual as part of a group rather than on the basis of the individual’s personal characteristics, discrimination does not require uniform treatment of all members of a particular group. It is sufficient that ascribing to an individual a group characteristic is one factor in the treatment of that individual. If a finding of discrimination required that every individual in the affected group be treated identically, legislative protection against discrimination would be of little or no value. It is rare that a discriminatory action is so bluntly expressed as to treat all members of the relevant group identically. In nearly every instance of discrimination the discriminatory action is composed of various ingredients with the result that some members of the pertinent group are not adversely affected, at least in a direct sense, by the discriminatory action. To deny a finding of discrimination in the circumstances of this appeal is to deny the existence of discrimination in any situation where discriminatory practices are less than perfectly inclusive . It is to argue, for example, that an employer who will only hire a woman if she has twice the qualifications required of a man is not guilty of sex discrimination if, despite this policy, the employer nevertheless manages to hire some women. [Emphasis added; pp. 1288-89.]

[ 355 ] Although Janzen and Brooks were decided in the human rights context, they were applied in the Charter context in Nova Scotia (Workers’ Compensation Board) v. Martin , 2003 SCC 54, [2003] 2 S.C.R. 504, where Gonthier J. held that “[t]his Court has long recognized that differential treatment can occur . . . despite the fact that not all persons belonging to the relevant group are equally mistreated” (para. 76). In other words, even if only some members of an enumerated or analogous group suffer discrimination by virtue of their membership in that group, the distinction and adverse impact can still constitute discrimination.

[ 356 ] The National Assembly enacted economic safeguards for spouses in formal unions based on the need to protect them from the economic consequences of their assumed roles. Since many spouses in de facto couples exhibit the same functional characteristics as those in formal unions, with the same potential for one partner to be left economically vulnerable or disadvantaged when the relationship ends, their exclusion from similar protections perpetuates historic disadvantage against them based on their marital status.

[ 357 ] There is no need to look for an attitude of prejudice motivating, or created by, the exclusion of de facto couples from the presumptive statutory protections. Nor need we consider whether the exclusions promote the view that the individual is less capable or worthy of recognition as a human being or citizen — which, as discussed in Kapp , would be difficult to prove. There is no doubt that attitudes have changed towards de facto unions in Quebec, but what is relevant is not the attitudinal progress towards them, but the continuation of their discriminatory treatment .

[ 358 ] This brings us to the s. 1 analysis. The application judge found that the purpose of the exclusion of de facto spouses from the presumptive statutory protections was to preserve their freedom to choose to be outside the legal regimes governing marriage and civil unions. The Court of Appeal questioned whether, in the context of spousal support, freedom of choice can qualify as a pressing and substantial objective. Quebec has implemented mandatory provisions relating to support, they noted, thereby denying spouses in marriages or civil unions any freedom of choice in the interests of protecting and compensating economically vulnerable spouses. De facto spouses, on the other hand, have been excluded from similar protection based on the very freedom of choice Quebec has decided is irrelevant for formally recognized spousal relationships. However, since the objective of preserving freedom of choice was not vigorously challenged by the parties before this Court, I would accept it for the purposes of the s. 1 analysis.

[ 359 ] At the rational connection stage, the government does not face a heavy burden. It must show “that it is reasonable to suppose that the limit may further the goal, not that it will do so” ( Alberta v. Hutterian Brethren of Wilson Colony , 2009 SCC 37, [2009] 2 S.C.R. 567, at para. 48 ). While I find the connection tenuous, I cannot say that excluding de facto spouses from the support and division of property protections is wholly unconnected to the goal of allowing couples the freedom to be outside the legal regimes governing marriage and civil union.

[ 360 ] The critical stage in this case, in my view, is minimal impairment, under which “the government must show that the measures at issue impair the right . . . as little as reasonably possible in order to achieve the legislative objective” ( RJR-MacDonald Inc. v. Canada (Attorney General) , [1995] 3 S.C.R. 199, at para. 160 ). In my view, an outright exclusion of de facto spouses cannot be said to be minimally impairing of their equality rights. A presumptively protective scheme, on the other hand, with a right on the part of de facto spouses to opt out , is an example of an alternative that would provide economically vulnerable spouses with the protection they need, without in any way interfering with the legislative objective of giving freedom of choice to those de facto spouses who want to exercise it.

[ 361 ] This Court has generally been reluctant to defer to the legislature in the context of total exclusions from a legislative scheme. In Eldridge v. British Columbia (Attorney General) , [1997] 3 S.C.R. 624, the Court described the lack of funding for sign language interpretation as an approach that did not “reasonably balanc[e] the competing social demands which our society must address” (para. 93, citing McKinney v. University of Guelph , [1990] 3 S.C.R. 229, at p. 314). Similarly, in Tétreault-Gadoury v. Canada (Employment and Immigration Commission) , [1991] 2 S.C.R. 22, the Court held that, even allowing for “a healthy measure of flexibility . . ., the complete denial of unemployment benefits [was] not an acceptable method of achieving any of the government objectives” (p. 47). And in Vriend v. Alberta , [1998] 1 S.C.R. 493, in the context of the total exclusion of sexual orientation from human rights protections, the Court found that “the call for judicial deference [was] inappropriate” (para. 127).

[ 362 ] The antipathy towards complete exclusions is hardly surprising, since the government is required under s. 1 to “explain why a significantly less intrusive and equally effective measure was not chosen” ( RJR-MacDonald , at para. 160). This will be a difficult burden to meet when, as in this case, a group has been entirely left out of access to a remedial scheme.

[ 363 ] I concede that the exclusion of de facto spouses from spousal support and property regimes in Quebec was a carefully considered policy choice. As my colleague LeBel J. points out, it was discussed and reaffirmed during successive family law reforms from 1980 onwards. But the degree of legislative time, consultation and effort cannot act as a justificatory shield to guard against constitutional scrutiny. What is of utmost relevance is the resulting legislative choice. Neither the deliberative policy route — nor the popularity of its outcome — is a sufficient answer to the requirement of constitutional compliance.

[ 364 ] This Court concluded in M. v. H. that there should be deference to the policy choices that the legislature is “in a better position than the court to make, as in the case of difficult policy judgments regarding the claims of competing groups or the evaluation of complex and conflicting social science research” (para. 79). But it went on to find that the question of spousal support entitlement was not a question on which the legislature should be given deference:

[Since] no group will be disadvantaged by granting members of same-sex couples access to the spousal support scheme under the [ Family Law Act , R.S.O. 1990, c. F.3], the notion of deference to legislative choices in the sense of balancing claims of competing groups has no application . . . . [para. 126]

[ 365 ] The argument was made that de facto spouses have other mechanisms available to them that compensate for their exclusion from the support and division of property regimes, specifically the ability to sign a cohabitation agreement and the possibility of claiming for unjust enrichment. However, these contractual and statutory protections available to de facto spouses in Quebec fall far short of what married and civil union spouses obtain presumptively, both in their content and in their realistic availability to the most vulnerable de facto spouses. The Court in M. v. H. has already rejected the view that the availability of either of these options meant that there was no discrimination against same-sex couples who were excluded from spousal support. It held that “neither the common law equitable remedies nor the law of contract are adequate substitutes for the FLA ’s spousal support regime” and that “if these remedies were considered satisfactory there would have been no need for the spousal support regime” (para. 124). As shall be seen, these alternatives are, in my respectful view, equally inadequate as substitutes for division of property.

[ 366 ] With respect to cohabitation agreements, Iacobucci J. in M. v. H. found that they did not provide an adequate justification for the exclusion of same-sex couples from the statute (see para. 124). A contract requires positive action on the part of the spouses. That means that “[t]hose who want to resolve support issues before the relationship breaks down are forced either to expend resources to devise a suitable contractual arrangement or risk being left without a remedy in law” (para. 122). De facto spouses face the same alternative with respect to division of property: since they are excluded from presumptive statutory regimes, they must either expend resources to create contractual protections or accept the risk of being unprotected. Iacobucci J. noted further that contracts provided inferior protection than a statute, such as against bankruptcy (para. 123).

[ 367 ] As to the ability of de facto spouses to claim unjust enrichment, M. v. H. also firmly rejected this as a viable alternative for spousal support. The Court found that unjust enrichment addressed different interests than a support order and that such a claim was “more onerous [to] claimants [and] available under far narrower circumstances” (para. 120). These comments are a full answer to the notion that the availability of a claim for unjust enrichment is an equitable substitute for spousal support.

[ 368 ] A claim for unjust enrichment is equally inadequate as a substitute for a statutorily presumptive division of property. The greatest difference between unjust enrichment and the presumptive or mandatory division of property in the Code is the burden placed on the claimant. While the partnership of acquests and the family patrimony presume equal sharing, unjust enrichment requires the claimant to establish his or her contribution before the court will order any corresponding compensation.

[ 369 ] Critical on this point, in my view, is the legislative history of the family property provisions. As discussed above, the family patrimony was adopted in response to the perceived weaknesses of the compensatory allowance, which allows a spouse to claim compensation for their demonstrated contributions to the enrichment of the patrimony of the other spouse (art. 427 C.C.Q. ). In other words, the compensatory allowance fulfills a very similar role to unjust enrichment. However, the National Assembly specifically decided that the compensatory allowance was an insufficient remedy, because it imposed too high a burden on the claimant. As a result, it established the family patrimony, which was mandatory, did not require proof of contribution, and presumed equal sharing. In light of the fact that the legislature itself did not consider an unjust enrichment-type remedy to be sufficient protection in marital spousal relationships, unjust enrichment is an inadequate alternative remedy for de facto spouses under s. 1 .

[ 370 ] As this Court noted in Martin , we can look to the measures taken by the rest of Canada in considering whether there are alternative, less infringing options available (para. 112). Every other province has extended spousal support to unmarried spouses. They have drawn different borders by setting minimum periods of cohabitation before couples are subject to their regimes, and have preserved freedom of choice by allowing couples to opt out. Saskatchewan, Manitoba, British Columbia, Nunavut and the Northwest Territories have also extended statutory division of property to unmarried spouses (with British Columbia’s law to that effect not yet in force). In spite of the lack of a uniform position on division of property, however, and regardless of the various thresholds that the rest of Canada has drawn for unmarried spouses, the existence of these alternatives to total exclusion is instructive.

[ 371 ] Quebec is, of course, in no way obliged to mimic any other province’s treatment of de facto spouses. Quebec not only has a separate system of private law from the rest of Canada, it also has unique historical and societal values which it has a right to express through its legislation. The fact of these other regimes, however, can be helpful in determining that there is a less impairing way to fulfill the objective of preserving freedom of choice.

[ 372 ] The current opt in protections may well be adequate for some de facto spouses who enter their relationships with sufficient financial security, legal information, and the deliberate intent to avoid the consequences of a more formal union. But their ability to exercise freedom of choice can be equally protected under a protective regime with an opt out mechanism. The needs of the economically vulnerable, however, require presumptive protection no less in de facto unions than in more formal ones.

[ 373 ] Professor Hélène Belleau’s expert report notes that the de facto spouses in her sample most likely to be aware that they did not benefit from the same legal protections as married or civil union spouses were those familiar with law through their profession, their spouse’s profession, or a prior separation with a previous spouse. Professor Belleau explains that beyond this group who have had some personal or professional experience with law, most spouses rarely consider, or are ignorant of, the law surrounding de facto unions:

[ translation ] On the whole, it was clear that the respondents were mistaken about the rights and obligations applicable to marriage, which they associated more generally with conjugal life, and therefore also with couples living in de facto unions. . . .

The majority of de facto spouses and of married spouses think that couples who have been living together in de facto unions for several years, or where they have children, have the same rights and obligations in the case of a breakdown. . . .

Aside from this misunderstanding, it was also observed that couples rarely discuss legal questions, in particular because such questions are incompatible with the notion of being in love. To discuss the legal questions that circumscribe the conjugal relationship inevitably leads one to foresee the possibility of an eventual breakup. In the context of a proposed marriage or a de facto relationship, these questions are not really compatible with the notion of being in love. . . .

For the majority of de facto spouses and of married spouses, legal questions are not included in the reflection that takes place when deciding whether to get married. At any rate, they believe that they have the same rights and obligations as married spouses. [Joint Record, vol. 8, at pp. 70-71]

Similarly, in “Controversy Over Couples in Canada: The Evolution of Marriage and Other Adult Interdependent Relationships” (2003), 29 Queen’s L.J. 41, at p. 53, Nicholas Bala points out that,

while it is doubtless true that “some” cohabitants live together because they have consciously chosen not to assume the obligations of marriage, many cohabitants give little thought to their rights and obligations, or are ill-informed or understandably confused about exactly what rights common-law partners have.

[ 374 ] This echoes Dickson J.’s observations in Rathwell that for many spouses, issues of economic or legal rights arising from their relationship are not a preoccupation when the relationship is a happy one. Many couples — married or de facto — simply “do not turn their minds to the eventuality of separation and divorce” (p. 444). This lack of awareness of a great number of de facto spouses, confirmed by the evidence, speaks to the relative merit of a system of presumptive protection, under which they would be protected whether aware of their legal rights or not, while leaving de facto spouses who wish to do so the freedom to choose not to be protected.

[ 375 ] A further weakness of the current opt-in system is its failure to recognize that the choice to formally marry is a mutual decision. One member of a couple can decide to refuse to marry or enter a civil union and thereby deprive the other of the benefit of needed spousal support when the relationship ends. In her dissenting reasons in Walsh , L’Heureux‑Dubé J. observed that “[t]his results in a situation where one of the parties to the cohabitation relationship preserves his or her autonomy at the expense of the other: ‘The flip side of one person’s autonomy is often another’s exploitation’” (para. 152, citing W. H. Holland, “Marriage and Cohabitation — Has the Time Come to Bridge the Gap?”, in Special Lectures of the Law Society of Upper Canada 1993 — Family Law: Roles, Fairness and Equality (1994), 369, at p. 380). The case before us resonates with this observation: Ms. A consistently wanted to marry, but Mr. B refused, depriving Ms. A of access to the possibility of spousal support at the end of the relationship.

[ 376 ] At the end of the day, the methodology for remedying the s. 15 breach lies with the Quebec legislature. The Quebec scheme currently gives de facto spouses the choice of entering into a contract to enshrine certain protections, or marrying and receiving all the protections provided by law, or remaining unbound by any mutual rights or obligations. None of these choices is compromised by a presumptively protective scheme of some sort. It is entirely possible for Quebec to design a regime that retains all of these choices. Spouses who are aware of their legal rights, and choose not to marry so they can avoid Quebec’s support and property regimes, would be free to choose to remove themselves from a presumptively protective regime. Changing the default situation of the couple, however, so that spousal support and division of property protection of some kind applies to them, would protect those spouses for whom the choices are illusory and who are left economically vulnerable at the dissolution of their relationship.

[ 377 ] In view of the conclusion that the provisions are not minimally impairing since other mechanisms for preserving choice are available, it is unnecessary, strictly speaking, to consider the final step of Oakes . Nonetheless, there seems to me to be some value in clarifying why the deleterious impact of the exclusion is more pronounced than its salutary effects. The harm of excluding all de facto spouses from the protection of the spousal support and family property regimes is clearly profound. Hallée J. at the Superior Court found, based on census figures and expert reports, that the number of de facto unions in Quebec continues to rise, representing 34.6% of all Quebec unions in 2006. These exclusions thus impact over a third of Quebec couples.

[ 378 ] Being excluded requires potentially vulnerable de facto spouses, unlike potentially vulnerable spouses in formal unions, to expend time, effort and money to try to obtain some financial assistance. If the vulnerable spouse fails to take these steps, either through a lack of knowledge or resources, or because of the limits on his or her options imposed by an uncooperative partner, he or she will remain unprotected. The outcome for such a spouse in the event of a separation can be, as it is for economically dependent spouses in formal unions, catastrophic. The difference is that economically dependent spouses in formal unions have automatic access to the possibility of financial remedies. De facto spouses have no such access.

[ 379 ] The salutary impact of the exclusion, on the other hand, is the preservation of de facto spouses’ freedom to choose not to be in a formal union. Leaving aside the trenchant observation of McLachlin J. in Miron about whether such choices are realistically genuine, this freedom would be equally protected under a presumptive scheme. Those for whom a de facto union is truly a chosen means to preserve economic independence can still achieve this result by opting out. Since this salutary effect can be achieved without in any way impairing a de facto spouse’s freedom of choice, it cannot be said to outweigh the serious harm for economically vulnerable de facto spouses that results from their exclusion from the spousal support and family property regimes.

[ 380 ] Because the distinction in excluding de facto spouses from the protective support regime in art. 585 and the division of property provisions in arts. 401 to 430, 432, 433 and 448 to 484 of the Civil Code cannot be justified, these articles are unconstitutional.

[ 381 ] I would therefore allow Ms. A’s appeal in part and dismiss the appeals of the Attorney General of Quebec and Mr. B.

English version of the reasons of Deschamps, Cromwell and Karakatsanis JJ. delivered by

[ 382 ] Deschamps J. (dissenting in part in result) — I agree with Abella J. that the Quebec legislature has infringed the guaranteed right to equality by excluding de facto spouses from all the measures adopted to protect persons who are married or in civil unions should their family relationships break down. Unlike my colleague, however, I do not view all these measures as a package. Nor can I endorse the position of my colleague LeBel J. that the majority of the protective measures constitute a mandatory primary regime with a single dominant objective. Although support and the measures relating to patrimonial property have some of the same functions and objectives, they cannot and must not be confused with one another. The needs they address and how the legislature has dealt with them in the past warrant their being considered separately. My analysis leads me to conclude that only the exclusion from support is not justified under s. 1 of the Canadian Charter of Rights and Freedoms (“ Charter ”).

[ 383 ] Aside from support, the measures in question are the matrimonial regime, the compensatory allowance, the family residence and the family patrimony. The effect of these various measures, which are the result of successive actions by the Quebec legislature, has been to gradually increase the protection provided to married and civil union spouses. However, art. 585 of the Civil Code of Québec , S.Q. 1991, c. 64 (“ C.C.Q. ”), establishes that the right to support granted to persons in need who are part of the family unit is distinct in that it does not have a compensatory function and does not depend on the consent of the debtor of support. This provision is different from the provisions on partition of property, which address a greater variety of needs, including the needs to protect vulnerable spouses, to compensate for contributions made by the parties while living together and to recognize the economic union formed by married and civil union spouses. I agree with the Court of Appeal’s conclusion that the exclusion of de facto spouses from the protection of support cannot be treated the same way as their exclusion from that of the other measures ( 2010 QCCA 1978 , [2010] R.J.Q. 2259). For the reasons that follow, I would dismiss the appeals.

[ 384 ] In my analysis on the right of de facto spouses to equality, I will not be relying directly on Nova Scotia (Attorney General) v. Walsh , 2002 SCC 83 , [2002] 4 S.C.R. 325. I find that certain aspects of that decision have not survived the recent decisions of this Court in R. v. Kapp , 2008 SCC 41 , [2008] 2 S.C.R. 483, and Withler v. Canada (Attorney General) , 2011 SCC 12 , [2011] 1 S.C.R. 396, particularly the fact that, according to it, the possibility for the parties of marrying to benefit from the measures relating to patrimonial property was one of the factors to be applied in determining whether the right to equality was infringed rather than in determining whether the infringement was justified under s. 1 of the Charter . In my opinion, the fact that the parties’ freedom of choice was invoked at the infringement stage of the analysis can only be attributed to the application of the test established in Law v. Canada (Minister of Employment and Immigration) , [1999] 1 S.C.R. 497. The autonomy of the will of the parties becomes relevant only at the justification stage, since this is the objective being pursued by the Quebec legislature.

[ 385 ] My colleagues LeBel and Abella JJ. do not take issue with the recognition of marital status as an analogous ground for the purposes of the analysis under s. 15 of the Charter . LeBel J. finds that the distinction at issue is not discriminatory. With respect, I agree instead with Abella J.’s analysis of s. 15 of the Charter . The exclusion of de facto spouses from the protections provided for in the C.C. Q. perpetuates a historical disadvantage ( Withler , at paras. 3, 35, 37 and 54 ). The Court has recognized the fact of being unmarried as an analogous ground because, historically, unmarried persons were considered to have adopted a lifestyle less worthy of respect than that of married persons. For this reason, they were excluded from the social protections. Even though society’s perception of de facto spouses has changed in recent decades and there is no indication that the Quebec legislature intended to stigmatize them, the denial of the benefits in question perpetuates the disadvantage such people have historically experienced ( Miron v. Trudel , [1995] 2 S.C.R. 418, at para. 152 ). The Attorney General of Quebec therefore had to justify this distinction.

I. Justification

[ 386 ] From a functional perspective, all the impugned measures have the effect of protecting married and civil union spouses who are in need following a separation. However, since support and the other protections do not have all the same bases, I find that the Court of Appeal was correct to distinguish support.

[ 387 ] In Bracklow v. Bracklow , [1999] 1 S.C.R. 420, at para. 15 , McLachlin J. (as she then was), writing for the Court, recognized that there are three possible bases for support: the first is compensatory, the second contractual and the third non‑compensatory. Although the “autonomist” trend that has predominated for the last 50 years or so has emphasized the compensatory and contractual aspects, the non‑compensatory basis nevertheless continues to be relevant. McLachlin J.’s comments on the non‑compensatory basis for support in the context of marriage are general enough to extend to de facto spouses (para. 31):

The mutual obligation view of marriage also serves certain policy ends and social values. First, it recognizes the reality that when people cohabit over a period of time in a family relationship, their affairs may become intermingled and impossible to disentangle neatly. When this happens, it is not unfair to ask the partners to continue to support each other (although perhaps not indefinitely). Second, it recognizes the artificiality of assuming that all separating couples can move cleanly from the mutual support status of marriage to the absolute independence status of single life, indicating the potential necessity to continue support, even after the marital “break”. Finally, it places the primary burden of support for a needy partner who cannot attain post‑marital self‑sufficiency on the partners to the relationship, rather than on the state , recognizing the potential injustice of foisting a helpless former partner onto the public assistance rolls. [Emphasis added.]

[ 388 ] These comments, together with others she had made in Miron , make it clear that the non‑compensatory basis is just as valid for de facto spouses as for married and civil union spouses, since de facto spouses may find themselves in a position of vulnerability without having had the choice of getting married or not getting married ( Miron , at para. 153 ):

In theory, the individual is free to choose whether to marry or not to marry. In practice, however, the reality may be otherwise. The sanction of the union by the state through civil marriage cannot always be obtained. The law; the reluctance of one’s partner to marry; financial, religious or social constraints — these factors and others commonly function to prevent partners who otherwise operate as a family unit from formally marrying. In short, marital status often lies beyond the individual’s effective control. [Emphasis added.]

[ 389 ] The changes that have been made to the rules governing support in Quebec confirm that this measure is distinct. In the codification of 1866, the Quebec legislature provided that married persons had an obligation of succour to one another (art. 173 of the Civil Code of Lower Canada , obligation reiterated in art. 392 of the Civil Code of Québec ). At the time of the major reform of family law in 1980, the legislature considered it appropriate to make the obligation of support resulting from the creation of a family unit a separate requirement. That obligation was incorporated into the Civil Code of Québec (1980), in art. 633. When the Civil Code of Québec was enacted in 1991, the obligation was set out in art. 585, which is in Book Two on the family, under Title Three, “Obligation of Support”. That title is separate from the one dealing with the rules of marriage, namely Title One, “Marriage”. In 2002, the obligation in question was extended to civil union spouses.

[ 390 ] It can be seen from the legislative history as set out by LeBel J. that the autonomist trend had a strong influence on the reform of the rules governing cohabitation. However, that trend was not what led to the recognition of the obligation of support itself or to the decision to make it a separate obligation. Rather, art. 585 reflects the non‑compensatory basis, which is closely tied to the creation of the “family unit” referred to in Miron . This interpretation is reinforced by the observation that art. 585 includes not only married and civil union spouses but also children and relatives. In short, the family unit benefits from the right to support.

[ 391 ] The bases for the other disputed measures vary. Several of them can easily be linked to the autonomist movement, while others are motivated by a desire to protect the disadvantaged spouse and establish rules of fairness between spouses. The oldest of the measures, the legal regime of partnership of acquests ( art. 432 ), was adopted in 1970. As LeBel J. explains, it was introduced to counter a tendency among spouses to choose the regime of separation of property, which often left wives with nothing if the marriage broke down. The regime of partnership of acquests gives spouses control over the property they acquire during the marriage and provides a basis for recognizing, if the marriage breaks down and property is to be partitioned, that the marriage gave rise to an “economic union” that results in a presumptive claim to equal standards of living: Moge v. Moge , [1992] 3 S.C.R. 813, at p. 870. Two of the other measures were part of the major reform of family law that took place in 1980: the compensatory allowance (arts. 427 to 430) and the family residence (arts. 401 to 413). As can be seen from the wording of art. 427, the former has a compensatory basis and is intended to recognize the contribution one spouse may have made to the other spouse’s patrimony. The purpose of this measure is to protect the spouse who made the contribution, thereby ensuring that the spouses’ economic autonomy is recognized — one spouse does not work for the other as a volunteer. As for the provisions protecting the family residence, their fundamental purpose is to protect the family unit. The final measure, the family patrimony (arts. 414 to 426), dates back to 1989. I agree with LeBel J. that the legislature’s objectives were to “remedy the problems encountered by women who had married under the regime of separation of property, make up for the ineffectiveness of the compensatory allowance and redefine marriage” (para. 74). Thus, the purpose of these provisions is, first and foremost, to protect the contributions made by the spouses and establish a legislative framework for the economic partnership of the parties in relation to family property.

[ 392 ] Because of the diversity of legislative sources and the variety of objectives being pursued and means that have been adopted by the Quebec legislature, it is impossible to place the majority of the measures at issue under a single umbrella, that of the “protection of vulnerable persons”, or to conclude that these measures should form an inflexible unit described as a “primary” regime, which, I should add, is a concept to which the legislature did not refer. Moreover, the measures that protect the patrimony of spouses are not, like support, focused on the basic needs of the vulnerable spouse. Their purpose is to ensure autonomy and fairness for couples who have been able to, or wanted to, accumulate property.

[ 393 ] There is another reason, a pragmatic one, why a distinction must be drawn between the measures related to property and those related to support. Whereas a plan to live together takes shape gradually and can result in the creation of a relationship of interdependence over which one of the parties has little or no control, property such as the family residence or pension plans can be acquired only as a result of a conscious act. The process that leads to the acquisition of a right of ownership is different from the one that causes a spouse to become economically dependent. In short, I find that the Court of Appeal was correct to distinguish the right to support from the patrimonial rights.

II. Support

[ 394 ] I agree with Abella J. that, since the parties do not really dispute that the objective of promoting the autonomy of the parties is pressing and substantial, the Court need not discuss this stage of the justification analysis in detail. I also agree with her that there is a rational connection, given that even a tenuous connection will satisfy the constitutional requirement in this regard. Since the obligation of support is mandatory for married and civil union spouses, the exclusion of de facto spouses means that those who do not wish to be bound by such an obligation are able to avoid it. This is the confirmation of a situation that, although undesirable in cases in which interdependence is the cause of the vulnerability of the spouse in need, constitutes a form of autonomy nonetheless.

[ 395 ] However, I cannot agree that this measure meets the minimal impairment test. The affected interest is vital to persons who have been in a relationship of interdependence. I will take the liberty of adapting the following comment of Cory and Iacobucci JJ. in M. v. H. , [1999] 2 S.C.R. 3, at para. 72 , to the context of the case at bar:

. . . the interest protected by [support] is fundamental, namely the ability to meet basic financial needs following the breakdown of a relationship characterized by intimacy and economic dependence. Members of [unmarried] couples are entirely ignored by the statute, notwithstanding the undeniable importance to them of the benefits accorded by the statute.

[ 396 ] The rationale for awarding support on a non‑compensatory basis applies equally to persons who are married or in a civil union and to de facto spouses. If the legal justification for support is based on, among other things, the satisfaction of needs resulting from the breakdown of a relationship of interdependence created while the spouses lived together, it is difficult to see why a de facto spouse who may not have been free to choose to have the relationship with his or her spouse made official through marriage or a civil union, but who otherwise lives with the latter in a “family unit”, would not be entitled to support. For someone in such a position, the possibility the parties have, according to the Attorney General, of choosing to marry or to enter into a civil union does not really exist. As the majority of the Court recognized in Miron , it is possible for a couple to remain unmarried contrary to the ardent wish of one of its members, the vulnerable one. As McLachlin J. said, freedom of choice is, or may be, theoretical.

[ 397 ] The Quebec government’s decision to take care of persons in need by providing them with social assistance benefits is not likely to make up for the exclusion of de facto spouses from the protection of support. Minimalist assistance such as this is not an adequate response. Social assistance is intended to be a measure of last resort and is not a reasonable substitute for support from a spouse who can afford to pay it.

[ 398 ] In Walsh , since the Court was not dealing with the issue of exclusion from support, it did not enquire into either the bases for the obligation of support or the fact that a relationship of interdependence may develop and be imposed on one of the parties without his or her having made a personal choice in this regard. That case is therefore of no assistance on this aspect of the instant case.

[ 399 ] The concept of “mutual obligation” as the non‑compensatory basis for the obligation of support must guide legislators in seeking ways to promote the autonomy of the parties while interfering as little as reasonably possible with the right to support itself. A total exclusion from the right to support benefits only de facto spouses who want to avoid the obligation of support, and it impairs the interests of dependent and vulnerable former spouses to a disproportionate extent. The legislature could, for example, have imposed on the parties an obligation to resolve their separation fairly and imposed on the dissatisfied party the burden of proving that the conditions of separation are unfair. Such a requirement would respect the autonomy of the parties while preventing abuse. This is only one of a number of possible solutions, and I mention it only to illustrate the fact that the legislature has less intrusive means at its disposal. The finding that there is a total exclusion from support without any mitigation of the effects of that exclusion is sufficient for me to conclude, like the Court of Appeal, that this measure is not justified under s. 1 of the Charter .

III. Measures Related to Rights of Ownership

[ 400 ] In my opinion, I must accept, as in the case of support, that the objective of promoting the autonomy of the parties is pressing and substantial and that a rational connection has been established.

[ 401 ] Regarding the compensatory allowance, I find that the exclusion of de facto spouses from this measure represents a minimal impairment. On the one hand, the right is in effect patrimonial in nature. In this regard, I take into account my conclusion that a spouse who can afford to do so must pay support to a spouse who needs it. The negation of this protective measure therefore does not compromise the basic ability of the former spouse to survive the breakdown of the relationship of interdependence with as much dignity as possible. On the other hand, although the legislature did not eliminate this measure, the debate that preceded the enactment of the provisions on the family patrimony shows that it was dissatisfied with decisions in which courts had failed to give sufficient recognition to the spouses’ respective contributions. That debate should provide the courts with a sound basis for interpreting the Civil Code ’s provisions on unjust enrichment in a manner consistent with the concept of equality entrenched in the Charter . Although total exclusion was certainly not the only possible solution for the legislature, the means that are still available to the vulnerable party are sufficient to meet the minimal impairment and balance of convenience tests.

[ 402 ] As for the partnership of acquests — the legal matrimonial regime — I find that this measure is also justified. First, this right too is patrimonial in nature. Unlike support, this measure does not relate to the ability of vulnerable persons to meet their basic needs. Next, participation in the legal regime requires a positive action by the parties. It is not a state resulting solely from the passage of time like the state of dependency that can gradually take hold in the parties’ relationship. Although it is not a contract, the formalization of a union through a marriage or civil union ceremony nevertheless constitutes consent that is given at a specific time, most often before a relationship of interdependence develops. In seeking ways to promote the autonomy of the parties, it was difficult for the legislature to avoid providing for, in parallel with the conventional regimes and the legal regime, a “no regime” option. In addition, like married and civil union spouses who opt for the legal regime, de facto spouses, while living together, remain completely autonomous and retain full ownership of the property they acquire. Of course, some de facto spouses will not be concerned about maintaining a fair division of the property acquired during the time they live together. If one spouse has been unjustly enriched at the other’s expense, however, this could be rectified by an action for unjust enrichment, interpreted, as I mentioned above, generously and in a manner consistent with the Charter .

[ 403 ] My reasons for concluding that the exclusion of de facto spouses from the protection of the family patrimony is justified are similar to those set out above. First, the affected interest is once again patrimonial in nature. Next, property becomes part of the family patrimony because the parties have deliberately decided to acquire it. Neither a residence nor movable property becomes part of a party’s patrimony over time without concrete action being taken. Unlike the interdependence that sometimes steals into conjugal life, over which the parties have no real control, the acquisition of patrimonial property results from decisions regarding which the government is justified in respecting the autonomy of the parties. The rules on the family patrimony were not established without causing a stir in Quebec society. At the time when they were adopted, there were transitional measures allowing married spouses to opt out of the protection so that they could not complain that their property had been “expropriated”. In the future, individuals would be able to opt out of these rules by not marrying.

[ 404 ] I recognize that support and the division of assets are both measures that make it possible to ease the burden created by the breakdown of a relationship of economic interdependence. However, the division of patrimonial property on the basis of an economic union in which each party is entitled to an equal share is based first and foremost on a laudable objective pursued by the legislature. The government had the power to impose measures of patrimonial protection on a given group, and it was not obliged to impose them on everyone. Those who are excluded from the application of these measures nevertheless do not lack ways to form an economic union analogous to the one imposed on persons who are married or in a civil union; they can, for example, purchase their residence jointly. There are protections that apply to spouses even if they have not formalized their relationship by choosing a protected form of union. As I mentioned above, a vulnerable spouse who is in need could be awarded support. Thus, in assessing what the debtor of support can afford to pay, the court must take all the debtor’s resources, including patrimonial property, into account. To ensure that the needs of the spouse in need are met, the court does not necessarily have to make an award of patrimonial property.

[ 405 ] In light of the objective of promoting the autonomy of the parties, the positive actions the parties must take to acquire family property and the flexibility the courts have in assessing the resources available for the payment of support, I find that the exclusion of de facto spouses from the protection of the family patrimony satisfies the minimal impairment requirement.

[ 406 ] The disadvantages of this measure do not outweigh its advantages, since, although the parties do not have an automatic right, there are nevertheless other ways for them to obtain sufficient protection.

[ 407 ] One issue remains: that of the family residence. Although this protection was originally adopted as a separate measure from the protection of the family patrimony, there are now a number of ways in which these two measures overlap, so I do not intend to engage in a separate analysis. However, I would note that the courts have taken a flexible approach, exercising their incidental powers with regard to the family residence. Indeed, an order to that effect was made in the instant case.

[ 408 ] In summary, I conclude that the exclusion of de facto spouses from support is not justified, but that their exclusion from the patrimonial measures is justified.

[ 409 ] For these reasons, I would dismiss the appeals and affirm the decision of the Court of Appeal to suspend the declaration of constitutional invalidity of art. 585 C.C.Q. for a period of 12 months, without costs.

The following are the reasons delivered by

[ 410 ] The Chief Justice — One of the responsibilities of provincial legislatures across Canada is to provide laws to deal with disputes concerning support and property of couples in conjugal relationships. In the old days, the problem was seen as simple; most couples were married, and it sufficed — or was thought to suffice — to pass laws regulating what happened when married couples separated.

[ 411 ] No longer are matters so simple. Increasingly, in all parts of Canada, couples are choosing to live together without being married. The stigma that once attached to these relationships has faded. The law has recognized that married couples and unmarried couples are entitled to equal treatment, for example with respect to insurance regime benefits, and that treating married couples differently from unmarried couples may be discriminatory and violate the equality guarantee of s. 15 of the Canadian Charter of Rights and Freedoms : Miron v. Trudel , [1985] 2 S.C.R. 418 .

[ 412 ] The legislatures of different provinces have responded to this challenge in different ways. In many parts of Canada, the choice has been to apply to de facto spouses an attenuated version of the mandatory regime that applies to married couples, unless the de facto couple formally chooses to opt out. For example, some provinces apply to de facto spouses the spousal support aspects of the mandatory regime applicable to married spouses. The Province of Quebec has chosen a different approach. Its law contemplates two completely different and distinct legal regimes — one for married couples and couples in civil unions, and one for de facto spouses. Couples who choose to marry or to enter into a civil union are subject to a mandatory regime governing both property and support (the “mandatory regime”). Upon separation, the family patrimony is divided between the spouses, and one spouse may be ordered to make support payments to the other. Couples who choose not to marry or to enter into a civil union — and this category is much larger in Quebec than in other provinces — are not subject to the mandatory regime that applies to married and civil union couples, and are free to craft whatever arrangements suit them. Unless they provide otherwise, each partner holds his or her property as an individual, and cannot be ordered to divide it with the other partner on separation. Nor, subject to exceptions related to children, can one partner be ordered to pay support to the other partner.

[ 413 ] Underlying the Quebec policy is the desire to enhance the right of Quebec couples to choose the regime they prefer, the one that best suits their particular needs. The policy is aimed at enhancing their choice and autonomy. Instead of a single norm based on the mandatory regime, there is a distinct choice between two different regimes: the mandatory regime, providing for a division of property and spousal support upon the dissolution of marriage or civil union, and a regime of full autonomy, allowing de facto spouses complete freedom to provide for the consequences of a break-up. The evidence is clear that this dual regime approach enjoys wide popularity in Quebec; many couples deliberately choose not to marry or to enter into a civil union in order to avoid the mandatory regime. If a couple does not marry or enter into a civil union, they will not be required to share their property or pay spousal support if the relationship ends. No special opt-out agreement is required, unlike in other provinces. The Quebec approach is grounded in Quebec’s unique history and social situation, as my colleague LeBel J. explains.

[ 414 ] The issue in this case is whether the Province of Quebec can maintain the dual regime approach that its legislature has adopted. A argues that this approach is unconstitutional because it unjustifiably discriminates against de facto spouses by denying them access to the more protective mandatory regime that applies to married and civil union couples. Accordingly, she argues that the Quebec law must be struck down and replaced by a regime that treats married, civil union and de facto spouses the same with respect to property division and spousal support upon separation.

[ 415 ] I agree with LeBel J. that the Quebec dual regime approach is constitutional. Unlike LeBel J., I conclude — as do my colleagues Deschamps J. and Abella J. — that the law violates the equality guarantee in s. 15 of the Charter . However, I find that the limit on the equality right of de facto spouses imposed by the law is reasonable and justifiable in a free and democratic society. Quebec’s goal is to enhance the choice and autonomy of couples in conjugal relationships. This policy goal is important to Quebec. Treating de facto spouses differently from married and civil union spouses enhances this goal, and does so in a proportionate way. The fact that Quebec has chosen a different policy than other provinces in keeping with its own history and social values does not make the law unconstitutional.

I. Section 15: Does the Quebec Law Discriminate Against De Facto Spouses?

A. The Section 15 Analysis

[ 416 ] I agree with the s. 15 analysis set out in Abella J.’s reasons, which flows from the refinements to the s. 15 analysis that the Court made in R. v. Kapp , 2008 SCC 41 , [2008] 2 S.C.R. 483, and Withler v. Canada (Attorney General) , 2011 SCC 12 , [2011] 1 S.C.R. 396. I disagree, however, as to whether the legislative scheme is justified under s. 1 of the Charter .

[ 417 ] Section 15 of the Charter protects against discrimination on the basis of personal characteristics, the enumerated or analogous grounds. Marital status is such a ground. To constitute discrimination, the impugned law must have the purpose or effect “of perpetuating or promoting the view that the individual is less capable or worthy of recognition or value as a human being or as a member of Canadian society, equally deserving of concern, respect, and consideration”: Law v. Canada (Minister of Employment and Immigration) , [1999] 1 S.C.R. 497, at para. 88 (3)(C); see also Andrews v. Law Society of British Columbia , [1989] 1 S.C.R. 143 , at p. 171.

[ 418 ] Most recently, this Court has articulated the approach in terms of two steps: (1) Does the law create a distinction based on an enumerated or analogous ground? (2) Does the distinction create a disadvantage by perpetuating prejudice or false stereotyping?: Kapp , at para. 17 ; Withler , at para. 30 . While the promotion or the perpetuation of prejudice, on the one hand, and false stereotyping, on the other, are useful guides, what constitutes discrimination requires a contextual analysis, taking into account matters such as pre-existing disadvantage of the claimant group, the degree of correspondence between the differential treatment and the claimant group’s reality, the ameliorative impact or purpose of the law, and the nature of the interests affected: Withler , at para. 38 ; Kapp , at para. 19 .

[ 419 ] A few further points that touch on differences between my reasons and those of LeBel J. bear noting. First, the issue of whether the law is discriminatory must be considered 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 claimant”: Law , at para. 60.

[ 420 ] Second, a legal distinction can be discriminatory either in purpose or in effect. As a practical matter, legislatures seldom set out to discriminate on purpose; discrimination when it occurs is usually a matter of unintended effect.

[ 421 ] Finally, and related to this, it is important to maintain the analytical distinction between s. 15 and s. 1 . While the public policy basis for legislation has a limited relevance to the s. 15 analysis, it is central to the s. 1 inquiry: see Andrews , at pp. 177-78. This flows from the two-stage model of constitutional review inherent in the Charter . As Aharon Barak, former President of the Supreme Court of Israel, puts it:

. . . what is the case when the legal system has adopted a two-stage model [of constitutional review], such as in Germany, Canada, South Africa, and Israel? . . . Should public interest considerations be included in the first stage or the second or in both stages? Should public interest considerations affect the determination of the right’s scope, or should consideration of these interests be postponed to the stage of . . . the discussion regarding proportionality?

The proper location for public interest considerations is in the second stage of the constitutional review, as part of the discussion of the justification of the limitation on the constitutional right.

( Proportionality: Constitutional Rights and their Limitations (2012), at pp. 75-76)

B. Application of the Section 15 Framework to A’s Claim of Discrimination

[ 422 ] The first question is whether this Court’s decision in Nova Scotia (Attorney General) v. Walsh , 2002 SCC 83 , [2002] 4 S.C.R. 325, resolves this appeal. Like my colleague Abella J., I am of the view that Walsh does not bind this Court in the present case. Walsh involved different issues (division of property only) and was decided at an earlier point in our evolving appreciation of s. 15. More fundamentally, however, I agree with Abella J. that freedom of choice and individual autonomy, which were held in Walsh to negate a breach of s. 15, are better considered at the s. 1 stage of the analysis. Freedom of choice and autonomy are public interest considerations. They are relied on by Quebec to justify the obvious fact that its law may disadvantage some de facto spouses by denying claims to property division and support in circumstances where they may not have truly chosen to forego the protections of the mandatory regime, but rather have been unable to access them due to their partner’s refusal to marry. As discussed above, under s. 15 of the Charter , public policy considerations should be considered at the second stage of the constitutional analysis.

[ 423 ] In my view, the Quebec dual regime approach makes discriminatory distinctions that limit the s. 15 equality right of de facto spouses. All the elements of a s. 15 violation are present. The law denies de facto spouses protections available to married and civil union spouses. These distinctions are made on the basis of the analogous ground of marital status: Miron . The distinctions create a disadvantage: de facto spouses do not automatically benefit from a series of provisions that ensure an equitable division of property and continued financial support at the end of a relationship characterized by financial interdependence ( Civil Code of Québec , S.Q. 1991, c. 64, arts. 401 et seq. , 414 et seq. , 427 et seq. , 432 and 585). Finally, the disadvantage is discriminatory from the point of view of a reasonable person placed in circumstances similar to those of A. The law in fact shows less concern for people in A’s position than for married and civil union spouses on break-up of the relationship. As it applies to people in A’s situation, it perpetuates the effects of historical disadvantage rooted in prejudice and rests on a false stereotype of choice rather than on the reality of the claimant’s situation.

[ 424 ] LeBel J.’s review of the relevant legislative history demonstrates that the purpose and intent animating the impugned provisions are not discriminatory. The Quebec legislature did not view de facto spouses as inferior or second-class spouses.

[ 425 ] However, this does not end the inquiry. It is necessary to go on to ask whether the adverse distinctions made by the law against de facto spouses discriminate against them in effect, on the approach to s. 15 set out above.

[ 426 ] In its effect, the Quebec scheme denies separated de facto partners important protections that it accords to separated married and civil union partners, despite the fact that they may not have meaningfully exercised a choice of regime. It is reasonable to infer from this, subject to a full analysis of the relevant contextual factors, that the law that denies them these protections treats them as less deserving of concern, respect and consideration.

[ 427 ] A reasonable person in A’s situation would conclude that the law perpetuates pre-existing disadvantage. De facto spouses in Quebec suffer from significant pre-existing disadvantage. Until the enactment of the family law reform in 1980, the legislation actively discouraged and marginalized de facto spousal relationships, by prohibiting de facto spouses from contractually agreeing to obligations stemming from their relationships and by de-legitimizing their offspring. While the legislative animus that underlay those measures has disappeared, the present law continues to exclude de facto spouses from the protective schemes of Quebec family law.

[ 428 ] Equally, a reasonable person in A’s position would conclude that in denying her recourse to spousal support and equitable property division, the law relies on false stereotypes. The law assumes that de facto partners choose to forego the protections it offers to married and civil union partners. Yet people in A’s situation have not in fact chosen to forego the protections of the mandatory regime. A’s real choice was of a different nature: she could either remain in a de facto relationship with B, or walk away from it after having become accustomed to the lifestyle she shared with him. More broadly, the law rests on the assumption that de facto partners will provide for their needs by making their own agreements or arrangements for property and support. Again, for claimants in A’s situation, this assumption fails to accord with the reality of their situation.

[ 429 ] What then of context? Quebec’s argument is that when the denial of protection to de facto partners is considered in the context of the absence of prejudice against de facto partners in Quebec and the desire to enhance the choice and autonomy of Quebec couples, the denial does not promote the view that de facto partners are less worthy of concern and protection than married and civil union partners, and hence is not discriminatory.

[ 430 ] The first difficulty with this argument is that it does not consider the distinction from the perspective of a reasonable person in the claimant’s position , as the equality jurisprudence requires. It may be that some de facto spouses are not prejudiced by the law because they choose to be unmarried and make alternate agreements or arrangements that suit them. But from the perspective of a person in the claimant’s position, who has as a matter of fact been denied the right to choose, it is reasonable to view the law as treating her as less deserving of concern and consideration than married and civil union spouses.

[ 431 ] The second and related difficulty with Quebec’s argument is that it imports public interest considerations — the goal of maximizing choice and autonomy for conjugal partners as a whole — into the s. 15 analysis. Such interests, as I discussed above, should not be considered at the first stage of determining whether a right has been limited, but at the second stage of determining whether the limitation on the right is justified.

II. Is the Breach of Section 15 Justified Under Section 1 of the Charter ?

[ 432 ] The equality analysis under the Charter is a two-stage process. The first stage asks whether the law limits the right at issue. The second stage asks whether that limit is reasonable and justified in a free and democratic society ( s. 1 ).

[ 433 ] For the reasons just discussed, I conclude that the dual regime approach — which denies de facto spouses the protections of the mandatory regime accorded to married and civil union spouses — treats de facto spouses unequally and discriminates against them contrary to s. 15 of the Charter . The remaining question is whether this limit on the equality right of de facto spouses is justified under s. 1 of the Charter.

[ 434 ] The state bears the burden of establishing justification on a balance of probabilities. The state must demonstrate (1) a sufficiently important objective to justify an infringement of a Charter right, (2) a rational connection between that objective and the means chosen by the state, (3) that the means are minimally impairing of the right at issue, and (4) that the measure’s effects on the Charter -protected right are proportionate to the state objective: R. v. Oakes , [1986] 1 S.C.R. 103.

A. An Important Objective

[ 435 ] The objective of the distinction between de facto spouses and married or civil union couples made by the Quebec dual regime approach is to promote choice and autonomy for all Quebec spouses with respect to property division and support. Those who choose to marry choose the protections — but also the responsibilities — associated with that status. Those who choose not to marry avoid these state-imposed responsibilities and protections, and gain the opportunity to structure their relationship outside the confines of the mandatory regime applicable to married and civil union spouses.

[ 436 ] The legislature pursued this objective in response to rapidly changing attitudes in Quebec with respect to marriage, namely a rejection of the gender inequalities associated with the tradition of marriage, a shift away from the influence of the Church and the assertion of values linked to individualism: B. Moore, “Culture et droit de la famille: de l’institution à l’autonomie individuelle” (2009), 54 McGill L.J. 257, at p. 268. The legislator sought to accommodate the social rejection of the traditional control by the state and the Church over intimate relationships. When the family patrimony provisions were adopted in Quebec, the responsible Minister stated that a harmonization of the legislative treatment of marriage and of de facto spousal relationships [translation] “would not be without consequences, for what then would be the meaning of marriage or the value in the civil context of religious marriage, and what would be the form of union developed by those who do not want to be regulated ?”: National Assembly, Journal des débats , vol. 30, No. 125, 2nd Sess., 33rd Leg., June 8, 1989, p. 6487 (emphasis added).

[ 437 ] The objective of the law is sufficiently important to justify an infringement of the right to equality.

B. Rational Connection

[ 438 ] The distinction made by the law between married, civil union and de facto spouses is rationally connected to the state objective of preserving the autonomy and freedom of choice of Quebec spouses. Without this distinction, the clear choice between a regime of division of property and support on the one hand, and a regime of full autonomy on the other hand, would be absent. The Quebec approach only imposes state-mandated obligations on spouses who have made a conscious and active choice to accept those obligations. The requirement of an active choice to undertake obligations is consistent with the objective of enhancing autonomy.

C. Minimum Impairment

[ 439 ] The Oakes test requires that impugned provisions impair the infringed right “as little as is reasonably possible”: R. v. Edwards Books and Art Ltd. , [1986] 2 S.C.R. 713, at p. 772. This Court has recognized that the state must have a margin of appreciation in selecting the means to achieve its objective: Irwin Toy Ltd. v. Quebec (Attorney General) , [1989] 1 S.C.R. 927, at p. 999; Alberta v. Hutterian Brethren of Wilson Colony , 2009 SCC 37 , [2009] 2 S.C.R. 567, at paras. 35 and 53 ; Edwards Books and Art . The question is whether the impugned provisions fall within a range of reasonable alternatives: Lavoie v. Canada , 2002 SCC 23 , [2002] 1 S.C.R. 769, at para. 61 ; RJR-MacDonald Inc. v. Canada (Attorney General) , [1995] 3 S.C.R. 199, at para. 160 ; McKinney v. University of Guelph , [1990] 3 S.C.R. 229 , at pp. 285-86 ; Hutterian Brethren , at para. 37. This is particularly the case where the impugned measures “attempt to strike a balance between the claims of legitimate but competing social values”: McKinney , at p. 285; Hutterian Brethren , at para. 53.

[ 440 ] In addition, the minimum impairment test is informed by the values of federalism. “The uniformity of provincial laws that would be entailed by a stringent requirement of least drastic means is in conflict with the federal values of distinctiveness, diversity and experimentation”: P. W. Hogg, Constitutional Law of Canada (5th ed. Supp. (loose-leaf)), vol. 2, p. 38-39; see also R. v. Advance Cutting & Coring Ltd. , 2001 SCC 70 , [2001] 3 S.C.R. 209, at para. 275 . The test must not be applied in a manner that amounts to identifying the Canadian province that has adopted the “preferable” approach to a social issue and requiring that all other provinces follow suit.

[ 441 ] A argues that the Quebec dual regime approach does not minimally impair the equality right of de facto spouses. She argues that choice and autonomy can be respected without excluding de facto spouses entirely from the mandatory regime applicable to married and civil union spouses. She points to other provinces, where aspects of the mandatory regime apply to de facto spouses, unless they have formally opted out of that regime. Under such schemes, de facto spouses are denied protection only if they have agreed to that result. Unlike A, they are not effectively left unprotected because their partner did not consent to marriage. She also argues that a scheme that allowed for judicial intervention where property division and/or support are warranted would be less impairing of her equality right than the current scheme.

[ 442 ] There is no doubt that schemes can be conceived — and indeed have been adopted in other provinces — that impair the equality right of de facto spouses to a lesser degree than the Quebec scheme. However — and this is the important point — such approaches would be less effective in promoting the goals of the Quebec scheme of maximizing choice and autonomy for couples in Quebec. The question at the minimum impairment stage is whether the limit imposed by the law goes too far in relation to the goal the legislature seeks to achieve. “Less drastic means which do not actually achieve the government’s objective are not considered at this stage”: Hutterian Brethren , at para. 54.

[ 443 ] A presumptive scheme that applied the mandatory regime to all spouses, subject to the right to opt out, would automatically sweep in all couples. Even if de facto spouses were given the opportunity to opt out, this scheme would offer a narrower conception of choice than does Quebec’s current approach. Indeed, opting out would require agreement and positive action on the part of de facto spouses. The Quebec scheme, by contrast, allows couples to avoid state-imposed obligations simply by not marrying. The state-free zone created by the Quebec scheme is thus broader than under a presumptive regime.

[ 444 ] The Quebec scheme has the benefit of giving spouses the opportunity to perform a cost-benefit analysis of staying in a de facto relationship that does not confer any rights upon them, but that correlatively does not impose any legal obligations on them : R. Leckey, “Chosen Discrimination” (2002), 18 S.C.L.R. (2d) 445, at p. 458 . The legislature has chosen to avoid mandatory protective provisions that important segments of the population may view as paternalistic, by instead allowing spouses to weigh the consequences of their choices and to make decisions accordingly.

[ 445 ] What then of the absence of judicial recourse? Quebec is the only province that provides for no court intervention whatsoever to ensure that de facto spouses exercised a meaningful choice to forego legal protection of their interests in the event of a breakdown of the relationship. Permitting judges to intervene and make orders for property and support for de facto spouses would obviously be less impairing of their equality right than the Quebec regime. However, again there would be a trade-off in diminished choice and autonomy. The Quebec scheme leaves it up to partners to choose whether to opt into the mandatory regime and leaves them the discretion to manage their independence if they do not opt in. Allowing judges to make orders would limit those choices, and result in individuals who thought they were free to structure their affairs finding themselves bound by judicially imposed obligations.

[ 446 ] Finally, it is suggested that the law does not meet the minimal impairment requirement because it affects support as well as property. For the reasons just discussed, allowing judges to award support would undermine the legislative goal of maximizing choice and autonomy. A judge, not the parties, would decide. The question at this stage of the analysis is whether the legislative goal could be achieved in a way that impacts the right less, not whether the legislative goal should be altered. Moreover, the protective effects of support and property division are intertwined and cannot be readily separated.

[ 447 ] For these reasons, I conclude that the Quebec law falls within a range of reasonable alternatives for maximizing choice and autonomy in the matter of family assets and support.

D. Proportionality

[ 448 ] Ultimately, the infringement of a protected right must be proportionate to the benefits of pursuing the state objective, having regard to the impact of the law on the exercise of the right and the broader public benefits it seeks to achieve.

[ 449 ] The impact of the Quebec scheme on the exercise and enjoyment of the equality right is significant. However, the discriminatory effects of the exclusion of de facto spouses from the mandatory regime are attenuated in the modern era, as compared to earlier points in Quebec’s history. The impugned provisions do not appear to perpetuate animus against de facto spouses. All parties to this appeal agreed that the de facto spousal relationship is a popular form of relationship in Quebec. There is no longer any stigma attached to de facto spousal relationships. Many spouses in Quebec appreciate and take advantage of the ability to structure their relationship outside the traditional strictures of marriage. The impugned provisions enhance the freedom of choice and autonomy of many spouses as well as their ability to give personal meaning to their relationship. Against this must be weighed the cost of infringing the equality right of people like A, who have not been able to make a meaningful choice. Critics can say and have said that the situation of women like A suggests that the legislation achieves only a formalistic autonomy and an illusory freedom. However, the question for this Court is whether the unfortunate dilemma faced by women such as A is disproportionate to the overall benefits of the legislation, so as to make it unconstitutional. Having regard to the need to allow legislatures a margin of appreciation on difficult social issues and the need to be sensitive to the constitutional responsibility of each province to legislate for its population, the answer to this question is no.

III. Conclusion

[ 450 ] I would allow the appeals of B and the Attorney General of Quebec and find the Quebec scheme to be constitutional. I would dismiss A’s appeal and would not award costs.

Appeals of the Attorney General of Quebec and B allowed, appeal of A dismissed, Deschamps , Cromwell and Karakatsanis JJ. dissenting in part in the result, Abella J. dissenting in the result.

Solicitors for the appellant/respondent the Attorney General of Quebec: Bernard, Roy & Associés, Montréal.

Solicitors for the appellant/respondent A: Borden Ladner Gervais, Montréal.

Solicitors for the appellant/respondent B: Norton Rose Canada, Montréal; Suzanne H. Pringle, Montréal.

Solicitor for the intervener the Attorney General of New Brunswick: Attorney General of New Brunswick, Fredericton.

Solicitor for the intervener the Attorney General of Alberta: Attorney General of Alberta, Edmonton.

Solicitors for the intervener Fédération des associations de familles monoparentales et recomposées du Québec: Garneau, Verdon, Michaud, Samson, Québec.

Solicitors for the intervener the Women’s Legal Education and Action Fund: Martha McCarthy & Company, Toronto; O’Hanlon, Sanders, Teixeira, Montréal.