Canada (Attorney General) v. Mossop

Marceau J.A. noted that the Tribunal recognized the fundamental question that was put to it as being whether the term "family status" includes a homosexual relationship. In his opinion, the question that the Tribunal defined as fundamental would not solve the real issue that had to be addressed. Marceau J.A. was of the view that a more specific analysis was required than one based on the general attributes of a family group.

Marceau J.A. first held that any error of law by a Canadian Human Rights Tribunal is reviewable on an application pursuant to s. 28 of the Federal Court Act , as the CHRA did not include any privative clause.

Marceau J.A. also held that the Tribunal erred in interpreting the term "family status" in the CHRA as including a homosexual relationship between two individuals. In so concluding, Marceau J.A. examined the propositions on which the Tribunal based its reasoning and noted that only a legal approach could lead to a proper understanding of the term "family status". In this respect, he stated, at p. 35:

To these serious difficulties I have with the propositions adopted by the Tribunal, I will add my concern with an approach that simply forgets that the word "family" is not used in isolation in the Act, but rather coupled with the word "status". A status, to me, is primarily a legal concept which refers to the particular position of a person with respect to his or her rights and limitations as a result of his or her being member of some legally recognized and regulated group. I fail to see how any approach other than a legal one could lead to a proper understanding of what is meant by the phrase "family status". Even if we were to accept that two homosexual lovers can constitute "sociologically speaking" a sort of family, it is certainly not one which is now recognized by law as giving its members special rights and obligations.

He added that the CHRA was amended in 1983, to express in English what the French version was already saying, "so that the English version must be taken to express the notion underlying the words used in French". He then concluded as follows, at p. 36:

So, the reasoning of the Tribunal simply does not appear to me acceptable. The Tribunal had no authority to reject the generally understood meaning given to the word "family" and to adopt in its stead, through a consciously ad hoc approach, a meaning ill-adapted to the context in which the word appears and obviously not in conformity with what was intended when the word was introduced, as shown by the legislative history of the amendment.

In his analysis of the real issue underlying the complaint, Marceau noted that sexual orientation was the real ground of discrimination involved. He added, at pp. 37-38:

But could it not be said, at this point, that although sexual orientation is not one of the enumerated grounds of discrimination in the Act, according to two cases, Veysey v. Canada (Commissioner of the Correctional Services), [1990] 1 F.C. 321 (T.D.) (affirmed on other grounds by the Court of Appeal on May 31, 1990, Court file A-557-89), and Brown v. B.C. (Min. of Health) (1990), 42 B.C.L.R. (2d) 294 (S.C.), it is a ground protected from discrimination under section 15 of the Charter , so that the conclusion reached by the Tribunal would be validated by being the only application of "family status" consonant with the Charter .

My reaction is that I do not see the Charter as capable of being used as a kind of ipso facto legislative amendment machine requiring its doctrine to be incorporated in the human rights legislation by stretching the meaning of terms beyond their boundaries.

Stone J.A. agreed with the result proposed by Marceau J.A., and also with his reasons except for a few differences which follow.

Firstly, according to Stone J.A. it is of considerable significance, in deciding the correctness of the decision of the Tribunal, to look at the objective of Parliament in adding "family status" as a prohibited ground of discrimination. He stated, at p. 40:

Until that amendment was adopted on July 1, 1983 the original English version of the Act included only "marital status" whereas the original French version included only "situation de famille". The amendment appears to have been introduced to resolve a discrepancy between the two versions.

Relying on the Minutes of Proceedings and Evidence of the Standing Committee on Justice and Legal Affairs, Stone J.A. noted that, at the time of the amendment, the then Minister of Justice made it clear that the government decided not to include in the Act "sexual orientation" as a prohibited ground of discrimination. In the opinion of Stone J.A., the term "family status", as used in the CHRA , does not include discrimination based on sexual orientation.

Secondly, Stone J.A. expressed the view that they were not asked to decide whether the term "family status" includes or excludes common law relationships, only briefly noting that a common law relationship is one that exists between two persons of the opposite sex.

Finally, as to the Charter issue, Stone J.A. noted that none of the parties had sought to demonstrate that any provision of the CHRA was in conflict with the Charter . The argument advanced before the court was that the Charter mandated the expression "family status" to include same-sex couples. Stone J.A.'s response to this argument was unequivocal (at p. 43):

While accepting that human rights legislation should be interpreted, as much as possible, in a manner consistent with the provisions of the Charter and its interpretation, I cannot accept that the Charter should operate so as to mandate the courts to ascribe to a statutory term a meaning which it was not intended to possess. If the statutory term, construed as I think it should be construed, is thought to conflict with the provisions of the Charter then the constitutional validity of that term must be put in issue for the Charter to play a role in resolving the dispute. Having already decided that the term "family status", as it is used in the Act, does not import sexual orientation as a prohibited ground of discrimination, I am unable to see how the Charter can alter the construction of that term.

IV - Issues

1.Did the Federal Court of Appeal err when it held that any error of law by a Canadian Human Rights Tribunal is reviewable on an application pursuant to s. 28 of the Federal Court Act ?

2.Did the Federal Court of Appeal err when it held that the term "family status' in the CHRA did not include a homosexual relationship between two individuals?

V - Analysis

The first question put to the Court, that concerning the Federal Court of Appeal's jurisdiction to review any error of law by the Canadian Human Rights Tribunal, is in my opinion quite straightforward. The appeal before the Federal Court of Appeal was based on s. 28 of the Federal Court Act , which for convenience's sake I shall reproduce here:

28. (1) Notwithstanding section 18 or the provisions of any other Act, the Court of Appeal has jurisdiction to hear and determine an application to review and set aside a decision or order, other than a decision or order of an administrative nature not required by law to be made on a judicial or quasi-judicial basis, made by or in the course of proceedings before a federal board, commission or other tribunal, on the ground that the board, commission or tribunal

...

(b) erred in law in making its decision or order, whether or not the error appears on the face of the record;

On this point I agree with Marceau J.A. of the Federal Court of Appeal, and adopt his reasons in this regard (at pp. 31-32):

The standard for reviewing the Tribunal's interpretation, said counsel, should be that established by the Supreme Court in Service Employees' International Union, Local No. 333 v. Nipawin District Staff Nurses Association et al., [1975] 1 S.C.R. 382; and National Bank of Canada v. Retail Clerks' International Union et al., [1984] 1 S.C.R. 269. I disagree. In both those cases referred to, and in the others where likewise the Supreme Court has limited the power of intervention of the reviewing courts to cases of patent unreasonableness, the tribunals were acting under the special protection of privative clauses. There is no such clause immunizing the decisions of a human rights tribunal. It may be difficult at times, in analyzing a decision, to extract the question of law from the facts of the case so as to verify the treatment given to it by the tribunal without interfering with factual findings not subject to review. But the facts in this case are clearly established and there is no danger of mixing them up with the purely legal question of interpretation involved. If the Tribunal was not correct in its answer to the question, however understandable may have been its error, the Court has the duty to intervene.

The question before the Court in this case is one of statutory interpretation: it is therefore a question of law. The appellant argued that, nevertheless, the Federal Court of Appeal should have exercised judicial restraint and upheld the Tribunal's decision. Absent a privative clause, the courts have shown curial deference vis-à-vis certain specialized tribunals when interpreting their own Act. The question is therefore whether a tribunal set up under the CHRA is such a body. On this point, this Court, my colleague L'Heureux-Dubé J. dissenting, just a few months ago, in Zurich Insurance Co. v. Ontario (Human Rights Commission), [1992] 2 S.C.R. 321, at p. 338, found that such a board does not have the kind of expertise that should enjoy curial deference on matters other than findings of fact:

In spite of the ability to overturn decisions of the Board on findings of fact, this Court has indicated that some curial deference will apply even to cases without privative clauses to reflect the principle of the specialization of duties (see Bell Canada v. Canada (Canadian Radio-television and Telecommunications Commission), [1989] 1 S.C.R. 1722, at p. 1746, Etobicoke, supra, at p. 211). While curial deference will apply to findings of fact, which the Board of Inquiry may have been in a better position to determine, such deference will not apply to findings of law in which the Board has no particular expertise.

It seems to me that this should have put the matter to rest. But if any additional reasons need be given for our having come to that conclusion, I would adopt in that regard the reasons of my colleague Justice La Forest in this case.

Having decided this question, I must now turn to the second one. It is important to remember that when this case was heard last June, the only question submitted to this Court was whether, by specifically denying homosexual couples access to certain benefits conferred on heterosexual couples, a union and the government had infringed the CHRA . There was no question of determining whether the government and the unions should or should not extend these types of benefits to homosexual couples, nor of deciding whether Parliament when enacting the CHRA should have prohibited discrimination on the basis of sexual orientation. Also of great importance to the dynamics of the analysis in this case is the fact that none of the provisions of the CHRA were challenged under the Charter . The question before this Court was thus strictly one of statutory interpretation.

Since then, as the result of two important decisions of Canadian courts, the situation in this country has evolved with respect to the questions at issue in this appeal. On July 9, 1992 this Court handed down its decision in Schachter v. Canada, [1992] 2 S.C.R. 679, confirming that, in a limited number of special circumstances, the courts may add to the text of legislative provisions so that they conform to the requirements of the Constitution. On August 6, 1992, the Ontario Court of Appeal, relying on the principles set forth in Schachter, added sexual orientation to the list of prohibited grounds of discrimination contained in s. 3 of the CHRA , as it was of the view that without this addition the provision was contrary to s. 15 of the Charter . The case in question was Haig v. Canada (1992), 9 O.R. (3d) 495. On November 9, 1992, the Minister of Justice, Kim Campbell, announced her intention not to appeal that decision.

As a result of these developments, the Court invited the parties to this appeal to submit new arguments. Relying on the reasons of the Ontario Court of Appeal in Haig, the appellant could then have challenged the constitutionality of s. 3 of the CHRA on the basis of the absence of sexual orientation from the list of prohibited grounds of discrimination. This would have enabled this Court to address the fundamental questions argued in the Ontario Court of Appeal in Haig. It would then have been possible to give a much more complete and lasting solution to the present problem.

The appellant chose not to take this approach, however, and insisted that this Court dispose of its action solely on the basis of the meaning of "family status". In these circumstances, as the Court did not have the benefit of any argument that would have enabled it to give an informed ruling on the questions decided by the Ontario Court of Appeal in Haig, and since on the present record it cannot do so, I can do no more than to dispose of this appeal on the basis of the law as it stood at the time of the events in question. Accordingly, the issue to be determined, on the facts of this case, is whether there was discrimination on the basis of Mr. Mossop's "family status" under the CHRA as it stood at the time the events occurred.

When Mr. Mossop was denied bereavement leave in June 1985, the CHRA did not prohibit discrimination on the basis of sexual orientation. In my opinion, this fact is a highly relevant part of the context in which the phrase "family status" in the Act must be interpreted. It is interesting to note in this regard that there was a recommendation by the Canadian Human Rights Commission that sexual orientation be made a prohibited ground of discrimination. Nevertheless, at the time of the 1983 amendments to the CHRA , no action was taken to implement this recommendation.

It is thus clear that when Parliament added the phrase "family status" to the English version of the CHRA in 1983, it refused at the same time to prohibit discrimination on the basis of sexual orientation in that Act. In my opinion, this fact is determinative. I find it hard to see how Parliament can be deemed to have intended to cover the situation now before the Court in the CHRA when we know that it specifically excluded sexual orientation from the list of prohibited grounds of discrimination contained in the Act. In the case at bar, Mr. Mossop's sexual orientation is so closely connected with the grounds which led to the refusal of the benefit that this denial could not be condemned as discrimination on the basis of "family status" without indirectly introducing into the CHRA the prohibition which Parliament specifically decided not to include in the Act, namely the prohibition of discrimination on the basis of sexual orientation.

While, with respect, I am not in agreement with all of Marceau J.A.'s judgment, I believe that he correctly identified the relationship which exists between sexual orientation and the discrimination at issue in this case (at p. 37):

... should it be admitted that a homosexual couple constitutes a family in the same manner as a husband and wife, it then becomes apparent that the disadvantage that may result to it by a refusal to treat it as a heterosexual couple is inextricably related to the sexual orientation of its members. It is sexual orientation which has led the complainant to enter with Popert into a "familial relationship" (to use the expression of the expert sociologist) and sexual orientation, therefore, which has precluded the recognition of his family status with regard to his lover and that man's father. So in final analysis, sexual orientation is really the ground of discrimination involved.

While it may be argued that the discrimination here applies to homosexual couples through their familial relationship or in their "family status" and does not apply to the sexual orientation of Mr. Mossop as an individual as such, I am not persuaded by this distinction. I cannot conclude that by omitting sexual orientation from the list of prohibited grounds of discrimination contained in the CHRA , Parliament intended to exclude from the scope of that Act only discrimination on the basis of the sexual orientation of individuals. If such an interpretation were to be given to the CHRA , the result would be somewhat surprising: while homosexuals who are not couples would receive no protection under the Act, those who are would be protected.

Whatever may be my personal views in that regard, I find that Parliament's clear intent throughout the CHRA , before and at the time of the amendment of 1983, was to not extend to anyone protection from discrimination based on sexual orientation.

Absent a Charter challenge of its constitutionality, when Parliamentary intent is clear, courts and administrative tribunals are not empowered to do anything else but to apply the law. If there is some ambiguity as to its meaning or scope, then the courts should, using the usual rules of interpretation, seek out the purpose of the legislation and if more than one reasonable interpretation consistent with that purpose is available, that which is more in conformity with the Charter should prevail.

But, I repeat, absent a Charter challenge, the Charter cannot be used as an interpretative tool to defeat the purpose of the legislation or to give the legislation an effect Parliament clearly intended it not to have.

Of course, if the effect of the legislation is in violation of the Charter , and a challenge of the constitutionality of the law is made before the courts, then the courts are commanded under s. 52 of the Constitution Act, 1982 to declare the section inoperative or to amend it when permissible along the lines set out in Schachter as did the Ontario Court of Appeal in Haig.

Before concluding, I should add that this does not mean that the hypothesis of overlapping grounds of discrimination should be ruled out in other contexts. Indeed, in this case, if Parliament had decided to include sexual orientation in the list of prohibited grounds of discrimination, my interpretation of the phrase "family status" might have been entirely different and I might perhaps then have concluded that Mr. Mossop's situation included both his sexual orientation and his "family status". For the reasons I have given, however, and in particular as there is no challenge under the Charter , I am unable to come to such a conclusion in the case at bar.

Nor should this decision be interpreted as meaning that homosexual couples cannot constitute a "family" for the purposes of legislation other than the CHRA . In this regard, each statute must be interpreted in its own context.

VI - Conclusion

For these reasons, I would dismiss the appeal.

//La Forest J.//

The reasons of La Forest and Iacobucci JJ. were delivered by

La Forest J. -- I have read the reasons of the Chief Justice and Justice L'Heureux‑Dubé. I share the general approach of the Chief Justice and would dispose of the case in the manner he proposes. I think it advisable, however, to deal more directly with some of the issues raised by my colleague, L'Heureux‑Dubé J. I shall, therefore, briefly set forth the main considerations that have led me to the conclusion I have reached.

Two issues were raised before us:

1.What is the standard of review of decisions of a human rights tribunal in the interpretation of its enabling legislation?

2.Did the Federal Court err in holding that the term "family status" did not include a homosexual relationship between two individuals?

1. The Standard of Review

I agree with the Chief Justice that the general question raised is one of statutory interpretation, and as such is a question of law over which the Federal Court of Appeal has jurisdiction. My colleague L'Heureux-Dubé J., however, maintains that we should defer to the views of a human rights tribunal not only on questions of fact, but also on questions of law. For this she relies on decisions on judicial review of administrative tribunals in specialized fields, and particularly labour law. With respect I cannot agree. First it must be recognized that these specialized tribunals frequently have strong privative clauses in their constituent legislation showing a legislative intention to limit judicial review. No such privative clause appears in the Canadian Human Rights Act , R.S.C., 1985, c. H-6 (formerly S.C. 1976-77, c. 33 as amended) (the "Act"). The normal approach to interpreting a tribunal's enabling statute should be that the courts retain their general supervisory jurisdiction. In the case of human rights tribunals, that supervision is by the Federal Court of Appeal under s. 28 of the Federal Court Act , R.S.C., 1985, c. F-7 . The enabling statute of an administrative tribunal empowers that body to do the things that are set forth in the statute. That is why a privative clause, which specifically addresses the position of the administrative tribunal vis-à-vis the courts, is of importance in determining a legislative intent to limit the judicial supervision of the tribunal. In the absence of other provisions indicating a disposition to limit judicial review, the normal supervisory role of the courts remains. The administrative tribunal, of course, is authorized to make determinations on these questions, but they are not to be insulated from the general supervisory role of the courts.

The courts have also been willing to show deference to administrative tribunals for reasons of relative expertise. This is in addition to the normal deference of reviewing courts in respect of questions of fact. But the position of a human rights tribunal is not analogous to a labour board (and similar highly specialized bodies) to which, even absent a privative clause, the courts will give a considerable measure of deference on questions of law falling within the area of expertise of these bodies because of the role and functions accorded to them by their constituent Act in the operation of the legislation. The Human Rights Commission undoubtedly serves many useful functions that help to educate, inform, and advise the government, the public and the courts on matters of human rights (s. 27). It also provides a procedure for initiating, investigating, and seeking voluntary settlement of human rights complaints. But it must be noted that in none of these roles is the work of the Commission binding on all parties. That power resides only with the tribunal in its adjudicative role under Part III of the Act. And the tribunal is not, simply by these other functions of the Commission, entitled to be free of normal review in its adjudicative function. The tribunals themselves are ad hoc bodies established to settle a particular dispute. In this respect their positions are similar to that of a labour arbitrator. But a human rights tribunal does not appear to me to call for the same level of deference as a labour arbitrator. A labour arbitrator operates, under legislation, in a narrowly restricted field, and is selected by the parties to arbitrate a difference between them under a collective agreement the parties have voluntarily entered. As well, the arbitrator's jurisdiction under the statute extends to the determination of whether a matter is arbitrable. This is entirely different from the situation of a human rights tribunal, whose decision is imposed on the parties and has direct influence on society at large in relation to basic social values. The superior expertise of a human rights tribunal relates to fact-finding and adjudication in a human rights context. It does not extend to general questions of law such as the one at issue in this case. These are ultimately matters within the province of the judiciary, and involve concepts of statutory interpretation and general legal reasoning which the courts must be supposed competent to perform. The courts cannot abdicate this duty to the tribunal. They must, therefore, review the tribunal's decisions on questions of this kind on the basis of correctness, not on a standard of reasonability.

2. Family Status

I turn, then, to the meaning to be attributed to the words "family status" under the ordinary rules of statutory interpretation. In determining the intent of Parliament, one must, of course, give to the words used in a statute their usual and ordinary sense having regard to their context and to the purpose of the statute. Here I shall focus particularly on the word "family" because the word "status" must inevitably attach to it. No one denies (and my colleague L'Heureux-Dubé J. concedes this) that the dominant conception of family is the traditional family. That, to use the term L'Heureux-Dubé J. uses, is the "unexamined consensus". That does not, of course, exhaust the meaning of the term, and we all know that in ordinary parlance it also comprises several derivative meanings that have a real connection with the dominant concept. I recognize, however, that particularly in recent years the word is loosely used to cover other relationships. The appellant here argues that "family status" should cover a relationship dependent on a same-sex living arrangement. While some may refer to such a relationship as a "family", I do not think it has yet reached that status in the ordinary use of language. Still less was it the case when the statute was enacted. In human terms, it is certainly arguable that bereavement leave should be granted to homosexual couples in a long-term relationship in the same way as it applies to heterosexual couples, but that is an issue for Parliament to address. It is not argued here that anything in the context supports the contention that this was the legislative purpose. The appellant's argument ultimately rests on the proposition that human rights statutes should be interpreted "purposefully" so as to favour all disadvantaged groups. I agree that the statute should be interpreted generously with a view to effect its purpose. But this brings us back to the question whether the addition of the words "family status" had as one of its legislative purposes the protection of persons living in the position of the appellant. As noted neither the language relied on nor the other grounds of discrimination listed support this. Nor is there any evidence in the surrounding context that this was the mischief Parliament intended to address, which could afford some credence to the argument that Parliament was using the words "family status" other than in their ordinary sense. As the Chief Justice observes, when one looks at extraneous evidence, there is nothing to show that Parliament intended to cover the situation of a same-sex couple. Indeed, so far as it goes -- and I do not attach any significance to it except in the negative way I have just mentioned -- this evidence would tend to support the opposite conclusion.

In sum, neither ordinary meaning, context, or purpose indicates a legislative intention to include same-sex couples within "family status". I underline that the present case is not an action under the Canadian Charter of Rights and Freedoms where the Court may review the actions of Parliament or the government, and I would refrain from saying anything about the issues such an action might raise. Nor do I think it appropriate, in the absence of argument, to consider the application of the Act to arrangements other than the one before us.

//L'Heureux-Dubé J.//

The following are the reasons delivered by

L'Heureux-Dubé J. (dissenting) -- I have had the opportunity of reading the reasons of Chief Justice Lamer and Justice La Forest, and with respect, I cannot agree with them nor with their disposition of this appeal. As the Chief Justice notes, this appeal concerns the interpretation of the term "family status", one of the enumerated grounds of discrimination in s. 3 of the Canadian Human Rights Act , R.S.C., 1985, c. H‑6 (formerly S.C. 1976‑77, c. 33 as amended) (the "Act"). The Court is asked to determine whether a Canadian Human Rights Tribunal (the "Tribunal") committed a reviewable error in interpreting the term "family status" so as to include a relationship between a couple of the same sex, in the context of the denial of a bereavement leave benefit provided for in a collective agreement. Although the Chief Justice has set out the facts and the judgments below, in view of my differing reasons, I would like to allude briefly to both.

Facts

The complainant Mr. Mossop and his male companion Mr. Popert first met in 1974 and lived together from 1976 on in a jointly owned and maintained home. The two men shared day‑to‑day activities, maintained a sexual relationship, and were known to their friends and families as a homosexual couple.

On June 3, 1985, Mr. Mossop attended the funeral of Mr. Popert's father. At the time of the complaint, Mr. Mossop was employed by the Department of the Secretary of State. His terms of employment were governed by a collective agreement between the Treasury Board and the Canadian Union of Professional and Technical Employees ("CUPTE"). Article 19.02 of this agreement provided for bereavement leave of up to four days upon the death of a member of an employee's immediate family. The day after the funeral of his companion's father, Mr. Mossop applied for bereavement leave. His application was turned down on the basis that the definition of immediate family in the collective agreement did not include the father of a same‑sex companion. With the approval and support of CUPTE, Mr. Mossop filed a grievance against his employer. When the grievance was rejected on the basis that the denial of benefits complied with the terms of the collective agreement, Mr. Mossop lodged complaints with the Canadian Human Rights Commission against his employer (to which was later added the Treasury Board) and his union. The Tribunal, appointed pursuant to s. 49 (formerly s. 39) of the Act, concluded that the Treasury Board and CUPTE had committed a discriminatory practice under s. 10(b) of the Act, which prohibits employers or unions from entering into agreements which restrict employment opportunities on a discriminatory basis: (1989), 10 C.H.R.R. D/6064.

By originating notice, the Attorney General of Canada made an application to the Federal Court of Appeal pursuant to s. 28 of the Federal Court Act, R.S.C., 1985, c. F‑7 , to review and set aside the decision of the Canadian Human Rights Tribunal. The Federal Court of Appeal allowed the application and the decision of the Tribunal was set aside: [1991] 1 F.C. 18, 71 D.L.R. (4th) 661, 32 C.C.E.L. 276, 114 N.R. 241, 90 C.L.L.C. ¶ 17,021, 12 C.H.R.R. D/355.

Relevant Legislation

Canadian Human Rights Act , R.S.C., 1985, c. H-6

2. The purpose of this Act is to extend the laws in Canada to give effect, within the purview of matters coming within the legislative authority of Parliament, to the principle that every individual should have an equal opportunity with other individuals to make for himself or herself the life that he or she is able and wishes to have, consistent with his or her duties and obligations as a member of society, without being hindered in or prevented from doing so by discriminatory practices based on race, national or ethnic origin, colour, religion, age, sex, marital status, family status, disability or conviction for an offence for which a pardon has been granted.

10. It is a discriminatory practice for an employer, employee organization or organization of employers

. . .

(b) to enter into an agreement affecting recruitment, referral, hiring, promotion, training, apprenticeship, transfer or any other matter relating to employment or prospective employment,

that deprives or tends to deprive an individual or class of individuals of any employment opportunities on a prohibited ground of discrimination. [Emphasis added.]

The Collective Agreement

2.01For the purpose of this Agreement,

. . .

(s)a "common‑law spouse" relationship is said to exist when, for a continuous period of at least one year, an employee has lived with a person of the opposite sex, publicly represented that person to be his/her spouse, and lives and intends to continue to live with that person as if that person were his/her spouse.

. . .

19.02 For the purpose of this clause, immediate family is defined as father, mother, brother, sister, spouse (including common‑law spouse resident with the employee), child (including child of common‑law spouse), or ward of the employee, father‑in‑law, mother‑in‑law, and in addition a relative who permanently resides in the employee's household or with whom the employee permanently resides.

(a)Where a member of his immediate family dies, an employee shall be granted bereavement leave for a period of up to four (4) consecutive days and not extending beyond the day following the funeral. During such period he shall be granted special leave with pay for those days which would have been regularly scheduled working days. In addition, he may be granted up to three (3) days' special leave with pay for the purpose of travel to and from the place of the funeral. [Emphasis added.]

Judgments

Canadian Human Rights Tribunal (1989), 10 C.H.R.R. D/6064

The Tribunal identified the fundamental question as being whether the denial of a bereavement leave was based on the prohibited ground of "family status". It noted that the interpretation of the term "family status" must be governed by the Act itself, and by the principles of interpretation of human rights codes established in the jurisprudence. The Tribunal found that the meaning of the term "family status" in s. 3 of the Act was ambiguous, and that certain interpretive aids were of little assistance. The Parliamentary record was inconclusive, there was no general understanding because of diverging societal views on homosexuality, and the "plain meaning rule" could not apply to a term which was as ambiguous as "family status". In the Tribunal's opinion, the term "family status" had to be assigned a meaning which was both reasonable and consonant with the context of the term, the intention of Parliament, and the object and scheme of the Act. Testing the terms against an understanding of how people were living and how language reflects reality, the Tribunal found that the Act was not meant to promote certain types of status over others, and that same‑sex couples were not prima facie excluded from the scope of "family status".

Noting that a collective agreement is the type of agreement subject to s. 10(b) of the Act, the Tribunal examined the provisions related to bereavement leave, and found that the benefit was available to an employee upon the death of members of the "immediate family". As the definition of "immediate family" in arts. 19.02 and 2.01(s) of the collective agreement included certain legal and factual relationships while excluding others, the Tribunal concluded that the agreement treated some families differently than others, that the collective agreement deprived Mr. Mossop of the employment opportunity of bereavement leave on a prohibited ground of discrimination, "family status", and that Treasury Board and CUPTE, in entering into the agreement, had committed a discriminatory practice under s. 10(b) of the Act.

The Tribunal ordered that a day of bereavement leave be designated, that the holiday leave credit which had been used to account for Mr. Mossop's absence be restored, that Treasury Board and CUPTE each pay Mr. Mossop $250 to compensate for injury to his feelings and self‑respect, that Treasury Board and CUPTE cease to apply arts. 19.02 and 2.01(s) of the collective agreement in so far as they do not allow bereavement leave in situations where a person would meet the definition of "common law spouse" except for the sex of that person, and that the collective agreement be amended so that the definition of common‑law spouse include persons of the same sex who would meet the definition in its other respects.

Federal Court of Appeal, [1991] 1 F.C. 18

Marceau J.A. stated that any error of law by a Canadian Human Rights Tribunal is reviewable on an application pursuant to s. 28 of the Federal Court Act . In his opinion, since there was no privative clause immunizing the decisions of the Human Rights Tribunal, the Court had a duty to intervene in the face of any error on a question of legal interpretation. He held that the Tribunal had committed such an error of law in interpreting the term "family status" in the Act as including a same‑sex relationship.

While agreeing that human rights legislation has often been said to be quasi‑constitutional, Marceau J.A. expressed the view that a "living‑tree" approach should not be taken towards interpreting such legislation. He felt that, in so doing, a tribunal would step outside the bounds of its constitutional responsibilities and usurp the function of Parliament. He rejected the Tribunal's functional or sociological approach, finding that though there may be room for extension of the term, the core meaning of "family" is well understood. Marceau J.A. considered it noteworthy that the word "family" did not appear in isolation, but was coupled with the word "status". He opined that status is primarily a legal concept, and that only through a legal approach could one arrive at a proper understanding of the phrase "family status". He commented that, in his view, the purpose of the 1983 amendment of the Act was to have the English text express what the French text was already saying. Thus, though the French text of the Act refers to "situation de famille" rather than "statut familial", he concluded that the French term must be taken to express the narrower meaning rendered by the English text.

In any event, Marceau J.A. was of the opinion that it was sexual orientation and not "family status" which was the real ground of discrimination involved. At the time, sexual orientation was not a prohibited ground under s. 3 of the Act, and he remarked that the fact that sexual orientation is a prohibited ground of discrimination under the Canadian Charter of Rights and Freedoms was not relevant to the issue of the proper characterization of the issue here. In his words, at p. 38:

. . . I do not see the Charter as capable of being used as a kind of ipso facto legislative amendment machine requiring its doctrine to be incorporated in the human rights legislation by stretching the meaning of terms beyond their boundaries.

Stone J.A. (Heald J.A. concurring) agreed with the result reached by Marceau J.A. He commented on three additional aspects of the appeal. First, in his view, the determination of Parliament's objective in adding "family status" as a prohibited ground of discrimination was essential to interpreting the term correctly. Since, prior to the adoption of an amendment on July 1, 1983, the original English text of the Act included only "marital status" whereas the original French text included only "situation de famille", Stone J.A. concluded that the amendment was introduced to resolve a discrepancy between the two texts, and that accordingly, the term "family status" did not expand the scope of the Act to provide protection against discrimination based on sexual orientation. Secondly, he noted that the question of whether the term "family status" includes or excludes common law relationships was not at issue. Finally, with respect to the Charter issue, Stone J.A. expressed the view that, while human rights legislation should be interpreted in a broad and purposive manner, he did not accept that the Charter mandates that the courts ascribe to a statutory term a meaning which it was not intended to possess. Having already decided that the term "family status" did not import sexual orientation as a prohibited ground of discrimination, he concluded that the Charter could not alter the construction of the term.

Issues

I would formulate the issues in this case differently than the majority in order to more accurately reflect my view of the case:

1.What is the standard of review of decisions of administrative tribunals on questions of law arising out of a tribunal's interpretation of its own enabling legislation?

2.Did the Tribunal's interpretation of the term "family status" in s. 3 of the Act meet the proper standard of review?

3.Did the Tribunal's conclusion that art. 19.02 of the collective agreement entered into between the Treasury Board and CUPTE infringes s. 3 of the Act meet the proper standard of review?

Standard of Review

Courts and academic writers have invested, as Brian Langille puts it in "Judicial Review, Judicial Revisionism and Judicial Responsibility" (1986), 17 R.G.D. 169, at p. 183, "imperials of ink" in addressing the problem of the proper institutional relationship between courts and administrative bodies. Specifically, under what conditions should courts intervene and substitute their views for the views of administrative bodies which have been given legislative mandates to deal with particular subject matters? It is no simple task to develop an approach to this question given the complex interrelationship of important and sometimes competing values and considerations.

On the one hand, there is the desire to respect legislative intention to create administrative bodies which can deal with issues in an economically efficient and expeditious fashion. There is a certain value in "finality", and the economic value in such systems may be reduced where appeals through the courts are readily available. Further, administrative bodies may develop valuable expertise in dealing with particular issues.

On the other hand, there is great variety and diversity among administrative bodies, and not all administrative bodies are specialized or have equal expertise. Since some administrative bodies have a greater potential than others to impact on personal rights and freedoms, it may not be desirable to have one unilateral standard of deference. Further, there is a need to ensure that economic efficiency does not come at the expense of the vulnerable. Finality is a hollow victory where obtained in a manner not according with principles of fundamental justice. In order to maintain public confidence in administrative systems, there is also the need to guard against potential administrative abuse of authority.

Given this diversity and the factors listed above, it is no simple task to devise a model for negotiating the relationship between courts and administrative bodies. However, it has been mediated primarily through the device of judicial review. The central principle of judicial review can be articulated with relative simplicity. As McLachlin J. for the majority noted in Lester (W.W.) (1978) Ltd. v. United Association of Journeymen and Apprentices of the Plumbing and Pipefitting Industry, Local 740, [1990] 3 S.C.R. 644, at p. 669, "Courts should exercise caution and deference in reviewing the decisions of specialized administrative tribunals".

The principle of deference is not automatic. It is well settled that, despite a generally deferential stance, courts will intervene in the face of a jurisdictional error, or a patently unreasonable error of fact or law (Canadian Union of Public Employees, Local 963 v. New Brunswick Liquor Corp., [1979] 2 S.C.R. 227 ("C.U.P.E."); National Bank of Canada v. Retail Clerks' International Union, [1984] 1 S.C.R. 269; Syndicat des employés de production du Québec et de l'Acadie v. Canada Labour Relations Board, [1984] 2 S.C.R. 412; Syndicat des professeurs du collège de Lévis‑Lauzon v. CEGEP de Lévis‑Lauzon, [1985] 1 S.C.R. 596; TWU v. British Columbia Telephone Co., [1988] 2 S.C.R. 564 (affirming Lambert J.A. dissenting at (1985), 65 B.C.L.R. 145 (C.A.)); U.E.S., Local 298 v. Bibeault, [1988] 2 S.C.R. 1048; Canada (Attorney General) v. Public Service Alliance of Canada, [1991] 1 S.C.R. 614 ("P.S.A.C."); CAIMAW v. Paccar of Canada Ltd., [1989] 2 S.C.R. 983; National Corn Growers Assn. v. Canada (Import Tribunal), [1990] 2 S.C.R. 1324; Lester (W.W.) (1978) Ltd. v. United Association of Journeymen and Apprentices of the Plumbing and Pipefitting Industry, Local 740, supra). Though the principle can be stated with relative ease, its application may often involve difficulties. Some reflections on the rationale for deference may provide some insight into the task of determining when deference is appropriate.

The deferential stance is rooted in part in a respect for governmental decisions to create administrative bodies with delegated powers. The constituting statute provides some indication of the scope of jurisdiction which has been conferred on the administrative body. Though it is difficult to make pronouncements about legislative intent, it can be presumed that the legislature intended that more deference would be shown to bodies with broad powers than to bodies with highly circumscribed powers. Where a broader jurisdiction is given, it may be appropriate to conclude that more rather than less deference should be given. Similarly, where a board is given a clearly specified and defined jurisdiction, even if over a narrow subject matter, great deference may be given to decisions that fall within that area. The purpose of the statute may also provide some evidence as to the appropriate measure of deference.

As a related matter, deference has often been linked to the existence of a privative clause. These clauses, which expressly limit access to court review, may provide further express evidence of legislative intent that courts should assume a more deferential role. It should be noted that while the principle of deference has often been applied in the presence of a privative clause, such clauses are not determinative. This Court has on several occasions reiterated that the absence of a privative clause does not necessarily result in review of all questions of fact and law. Estey J. (dissenting in part) in Douglas Aircraft Co. of Canada v. McConnell, [1980] 1 S.C.R. 245, at pp. 274‑75, stated that, even in the absence of a privative clause, a board's decisions on issues of fact, procedure and evidence should be beyond judicial reach, and that even matters of law should not be subject to review unless those errors were such as to assume "jurisdictional proportions". In my view, the privative clause is an important consideration, but does not definitively answer the question of whether or not deference would be appropriate in a given situation.

The rationale for deference also rests in part on an acceptance of the specialized nature of certain boards. Wilson J., in her concurring opinion in National Corn Growers, supra, commented that the increasingly deferential stance of the Court was in part linked to the increasing appreciation of the expertise of certain bodies. As she notes at p. 1336:

Part of this process has involved a growing recognition on the part of courts that they may simply not be as well equipped as administrative tribunals or agencies to deal with issues which Parliament has chosen to regulate through bodies exercising delegated power, e.g., labour relations, telecommunications, financial markets and international economic relations. Careful management of these sectors often requires the use of experts who have accumulated years of experience and a specialized understanding of the activities they supervise.

Courts have also come to accept that they may not be as well qualified as a given agency to provide interpretations of that agency's constitutive statute that make sense given the broad policy context within which that agency must work.

The Court has given recognition to the importance of expertise. In Bell Canada v. Canada (Canadian Radio‑television and Telecommunications Commission), [1989] 1 S.C.R. 1722, at p. 1746, Gonthier J. observed that, even where a court is granted statutory appellate jurisdiction over a board, "curial deference should be given to the opinion of the lower tribunal on issues which fall squarely within its area of expertise". In Zurich Insurance Co. v. Ontario (Human Rights Commission), [1992] 2 S.C.R. 321, in which I was dissenting, but not on this point, Sopinka J., for the majority, further added at p. 338 that "curial deference will apply even to cases without privative clauses to reflect the principle of the specialization of duties". (It should be noted here that the legislation at issue in Zurich, supra, not only included a clause providing for an appeal on any question of law or fact, but also expressly allowed for an appellate court to substitute its own findings for that of the Board.) (See also Gendron v. Supply and Services Union of the Public Service Alliance of Canada, Local 50057, [1990] 1 S.C.R. 1298, at p. 1321.) A related consideration is the connection of the board to the context. That is, even a body made up of "non‑specialists" may develop a certain "field‑sensitivity" where that body is in a position of proximity to the community and its needs. Where the question is one that requires a familiarity with and understanding of the context, there is a stronger argument that a higher degree of deference may be appropriate.

The rationale for deference is also influenced by the nature of the question or interest being considered. Some questions are appropriately left to boards, others should be determined by courts. Courts have recognized that statutory interpretation is not such a strict science, and that there are situations where it may be less appropriate to speak of "the correct answer", and more appropriate to speak about ranges of acceptable answers. Where the answer depends upon a policy choice, the question is simply who is best placed to make those choices. Where the administrative body has the jurisdiction to make policy choices, there are good reasons for the court to show a more deferential stance. However, there are questions where it would be clearly inappropriate to defer. Constitutional questions, for example, are not appropriate ones for showing deference. This is not to say that administrative boards are not competent to hear these concerns. This Court has clearly stated that administrative bodies can play a role in such determinations, and can be of great assistance in compiling the record. The standard of review on such questions, however, will be one of correctness. (See Douglas/Kwantlen Faculty Assn. v. Douglas College, [1990] 3 S.C.R. 570; Cuddy Chicks Ltd. v. Ontario (Labour Relations Board), [1991] 2 S.C.R. 5; Tétreault‑Gadoury v. Canada (Employment and Immigration Commission), [1991] 2 S.C.R. 22.)

The rationale for deference may differ depending on whether the question is one of fact or of law. In general, deference is given on questions of fact because of the "signal advantage" enjoyed by the primary finder of fact. Less deference is warranted on questions of law, in part because the finder of fact may not have developed any particular familiarity with issues of law. While there is merit in the distinction between fact and law, the distinction is not always so clear. Specialized boards are often called upon to make difficult findings of both fact and law. In some circumstances, the two are inextricably linked. Further, the "correct" interpretation of a term may be dictated by the mandate of the board and by the coherent body of jurisprudence it has developed. In some cases, even where courts might not agree with a given interpretation, the integrity of certain administrative processes may demand that deference be shown to that interpretation of law.

As this discussion makes clear, there are a variety of reasons intermingled in the generally deferential stance. This intermingling makes it difficult to devise a strict set of rules or criteria applicable to all situations. Not all boards will be entitled to the same measure of deference, and even boards with a great deal of expertise may from time to time err in a manner that mandates intervention. In brief, deference will not always be appropriate, and courts should not blindly abandon their inherent supervisory role, which is one of their most important inherent powers. (Crevier v. Attorney General of Quebec, [1981] 2 S.C.R. 220). As I noted in TWU v. British Columbia Telephone Co., supra, at p. 586 (dissenting in the result):

The two principles of non‑intervention within jurisdiction, on the one hand, and judicial review of excess of jurisdiction, on the other, are principles of equal importance in the area of administrative law, and one cannot be allowed to override the other.

It is one thing for a court to defer to a specialized body acting within its jurisdiction, and quite another to refuse to properly exercise its supervisory power where appropriate. The difficulty is in determining whether deference is warranted in a specific case. The best approach to this quandary is one which recognizes the need for flexibility. In Douglas Aircraft, supra, at p. 278, Estey J. observed:

[Judicial review] is and must remain a remedy with adaptable flexibility. The administrative juridical mechanism itself is still growing and evolving. With it will grow and evolve the concomitant interrelationship between the administrative tribunal and the superior court...

Such flexibility is found where review takes place in line with the "rational compass of the statute" (see P. Weiler, "The `Slippery Slope' of Judicial Intervention: The Supreme Court and Canadian Labour Relations 1950‑1970" (1971), 9 Osgoode Hall L.J. 1, at p. 33). This Court has favoured this approach which J. H. Grey has commented upon in "Sections 96 to 100: A Defense" (1985), 1 Admin. L.J. 3, at p. 11:

There is no single immutable standard of justice to be applied to all judicial decisions or fairness to be applied to administrative ones . . . . What Crevier does entrench is some degree of review. The courts will not interfere at the same moment on all issues or against all tribunals. However, they now clearly possess a constitutional right to step in when the bounds of tolerance are exceeded by any decision‑maker. Clearly, the precise location of the bounds of tolerance is left to the court and that is quite consistent with the general trends in modern administrative law. [Emphasis in original.]

The task for courts is, in light of the value of deference, to determine the appropriate standard of review in a given context. The pragmatic and functional approach articulated by Beetz J. for the Court in Bibeault, supra, provides, in my opinion, the proper framework and the question to ask is the one he set out at p. 1087: "Did the legislator intend the question to be within the jurisdiction conferred on the tribunal"?

In that case, Beetz J. emphasized that the concept of jurisdiction is the foundation of judicial review, and that it cannot be emptied of its meaning or content. He clarified its meaning in the context of judicial review at p. 1090:

Jurisdiction stricto sensu is defined as the power to decide. The importance of a grant of jurisdiction relates not to the tribunal's capacity or duty to decide a question but to the determining effect of its decision. As S. A. de Smith points out, the tribunal's decision on a question within its jurisdiction is binding on the parties to the dispute.... The true problem of judicial review is to discover whether the legislator intended the tribunal's decision on these matters to be binding on the parties to the dispute, subject to the right of appeal if any.

If the question is one which was intended to be within a board's jurisdiction, the role of the courts is a superintending one, and intervention will be warranted only where the decision is patently unreasonable. This view, according to Beetz J. at p. 1090:

. . . puts renewed emphasis on the superintending and reforming function of the superior courts. When an administrative tribunal exceeds its jurisdiction, the illegality of its act is as serious as if it had acted in bad faith or ignored the rules of natural justice. The role of the superior courts in maintaining the rule of law is so important that it is given constitutional protection: Crevier v. Attorney General of Quebec, [1981] 2 S.C.R. 220. Yet, the importance of judicial review implies that it should not be exercised unnecessarily, lest this extraordinary remedy lose its meaning. [Emphasis added.]

The concept of jurisdiction has been criticized by some as overly vague. As many writers have commented, concepts such as "jurisdiction" and "patently unreasonable" seem highly subjective, and do not appear to provide an objective set of criteria. (See P. L. Bryden, "Case Comment: United Association of Journeymen and Apprentices of the Pipefitting Industry v. W.W. Lester (1978) Ltd." (1992), 71 Can. Bar Rev. 580; A. J. Roman, "The Pendulum Swings Back" (1991), 48 Admin. L.R. 274; J. M. Evans, "Jurisdictional Review in the Supreme Court: Realism, Romance and Recidivism" (1991), 48 Admin. L.R. 255; D. J. Mullan, "Of Chaff Midst the Corn: American Farm Bureau Federation v. Canada (Canadian Import Tribunal) and Patent Unreasonableness Review" (1991), 45 Admin. L.R. 264; D. Mullan, "The Re‑Emergence of Jurisdictional Error" (1985), 14 Admin. L.R. 326.)

In Bibeault, supra, Beetz. J was not unaware of these difficulties. He clearly recognized that, though the principle of jurisdiction is simple, its application may nonetheless present difficulties. This is particularly so given the uncertainty inherent in language, and the related difficulty of relying on statutory texts to determine the bounds of a tribunal's jurisdiction. Application may become even more complex where the point in issue is whether or not the tribunal has the "jurisdiction" to determine the bounds of its own "jurisdiction". At p. 1087, Beetz J. remarked that there was unfortunately no simple test for making these determinations, "given the fluidity of the concept of jurisdiction and the many ways in which jurisdiction is conferred on administrative tribunals."

Because of the difficulties of application, it is important for courts to remain conscious of the rationale for deference. As former Chief Justice Laskin wrote in Canada Labour Relations Board v. Halifax Longshoremen's Association, [1983] 1 S.C.R. 245, at p. 256:

It is rarely a simple matter to draw a line between a lawful and unlawful exercise of power by a statutory tribunal, however ample its authority, when there are conflicting considerations addressed to the exercise of power. This Court has, over quite a number of years, thought it more consonant with the legislative objectives involved in a case such as this to be more rather than less deferential to the discharge of difficult tasks by statutory tribunals like the Board.

Courts also must not lose sight of Dickson J.'s warning in C.U.P.E., supra, at p. 233:

The question of what is and is not jurisdictional is often very difficult to determine. The courts, in my view, should not be alert to brand as jurisdictional, and therefore subject to broader curial review, that which may be doubtfully so.

In short, the approach articulated by this Court requires a focus on jurisdiction, a focus which accounts for the general values of deference, and the ease with which questions can be improperly branded as jurisdictional. Though it was first used in the context of a board protected by a privative clause, it is a principled approach of general application which does not focus on formal categories, but rather seeks to determine the rationale behind deference in a specific context. Using the pragmatic and functional approach, Beetz J. made it explicit in Bibeault, at p. 1088, that:

. . . the Court examines not only the wording of the enactment conferring jurisdiction on the administrative tribunal, but the purpose of the statute creating the tribunal, the reason for its existence, the area of expertise of its members and the nature of the problem before the tribunal.

Implicitly here, one must ask: "Who should answer this question, the board or a court?" Without providing an exhaustive list, Beetz J. suggests that there are at least three general areas that can be considered: the statute or legislative text; the administrative body; and the nature of the problem involved. In the same vein, Dickson J. in C.U.P.E., supra, at p. 236, suggested a similar group of characteristics that might lead a court to show more rather than less deference to an administrative tribunal's decisions. He considered such matters as: legislative intent to confide certain decisions to a board; broad supervisory and administrative powers given to a board; the need to balance several social goals; the sensitivity and expertise required by board members; and the existence of a privative clause. I note that the methodology suggested by both Beetz and Dickson JJ. is consistent with that proposed by H. W. MacLauchlan in "Judicial Review of Administrative Interpretations of Law: How Much Formalism Can We Reasonably Bear?" (1986), 36 U.T.L.J. 343. In this article MacLauchlan, though writing prior to Bibeault, supra, provides a helpful method of analysis which is consistent with the pragmatic and functional model articulated in that case.

First, then, is an examination of the statute. In so doing, there are many questions that can be considered. What is the purpose of the board? To what social needs is it responding? What is the scope of powers that have been granted to it? Are these powers defined in broad or narrow terms? Does it have policy-making powers? Is there language suggesting that more or less deference should be given? Is there a privative clause?

Second, it is valuable to focus clearly on the board whose decision is being impugned. Is it a specialized board? Does it have a developed body of jurisprudence that guides it and functions as precedent? How are its members selected, and how do they participate in decision‑making? Is there a context in which the members work that provides them with field‑sensitivity or other advantages?

Third, the nature of the problem under scrutiny is an important consideration. Does the matter squarely or by implication fall within the powers of the board? Does it require specialized knowledge to answer? Would it best be decided in a context-specific setting, or is it a question of general application? Does the problem seem to have only one "correct" answer, or are there a variety of possible and reasonable interpretations? Does the integrity of the administrative scheme require that the problem be answered by the board? Are there reasons to think that a court would be better suited to deal with the matter? Does it involve a question of constitutional interpretation that would be inappropriately left to be determined by a board?

If, after considering these factors (the list is not necessarily exhaustive), one concludes that courts should answer the question, then the question is one which does not lie within the jurisdiction of the board and the test of correctness should apply. If one concludes that the question should be answered by the board, then the question is one within the board's jurisdiction and courts should only intervene if the decision is patently unreasonable. By remaining focused on who is best placed to make the decision being challenged (even where that decision is a determination about the extent of jurisdiction), the pragmatic and functional approach offers the best method of determining whether deference or intervention is appropriate and merited in the specific situation.

This pragmatic approach is not a new standard of review. Courts have a history of determining the appropriate standard of review in accordance with the "rational compass of the statute". In Ontario Human Rights Commission v. Borough of Etobicoke, [1982] 1 S.C.R. 202, the Court found review to be appropriate as the Ontario Human Rights Code contained a very broadly worded statutory right of appeal. This provision was similar to one contained in the Individual's Rights Protection Act considered in Dickason v. University of Alberta, [1992] 2 S.C.R. 1103. In that case, Cory J. for the majority found that review was appropriate, and affirmed at p. 1125 that "the standard of review of decisions made by administrative bodies will always be governed by their empowering legislation." In Zurich, supra, the majority found it significant that the legislation contained a wide‑ranging appeal clause which allowed appeals to the courts on questions of law or fact, and authorized a court to substitute its opinion for that of the board. The Court observed (at p. 337) that "the legislature was not of the opinion that the conclusions of the Board of Inquiry should be given great deference as a result of accumulated expertise or specialized understanding." As is clear from these decisions, the standard of review will vary from body to body, based on the legislation establishing the board or tribunal, the expertise of that body, and the nature of the question.

Regarding s. 28 of the Federal Court Act , I would note that a distinction is to be made between jurisdiction and discretion. Review is a discretionary remedy. Though courts have a discretionary power to review, this does not mean that it will always be appropriate to exercise that discretion. There is nothing in s. 28 that dictates review on every error of law. Given the rationale for deference, and the importance of the court's supervisory power, an error should be a serious one to merit a court's intervention. It may be useful to recall Dickson C.J.'s comments in Fraser v. Public Service Staff Relations Board, [1985] 2 S.C.R. 455, at p. 464, that a reviewing court, whether pursuant to s. 28 of the Federal Court Act or to common law principles, must exercise its powers of review with caution and sensitivity:

A restrained approach to disturbing the decisions of specialized administrative tribunals, particularly in the context of labour relations, is essential if the courts are to respect the intentions and policies of Parliament and the provincial legislatures in establishing such tribunals . . .

It is with this background in mind that, in order to determine the appropriate standard of deference in the present instance, one has to inquire into the most appropriate adjudicator of the term "family status" in s. 3 of the Act. This involves a consideration of the Act, the Commission itself, and the nature of the problem involved. I first turn to the Act. The Act establishes a Human Rights Commission which is given broad powers in a highly specialized and sensitive area. The powers, duties and functions of the Commission are set out in s. 27:

27. (1) In addition to its duties under Part III with respect to complaints regarding discriminatory practices, the Commission is generally responsible for the administration of this Part and Parts I and III and

(a) shall develop and conduct information programs to foster public understanding of this Act and of the role and activities of the Commission thereunder and to foster public recognition of the principle described in section 2;

(b) shall undertake or sponsor research programs relating to its duties and functions under this Act and respecting the principle described in section 2;

(c) shall maintain close liaison with similar bodies or authorities in the provinces in order to foster common policies and practices and to avoid conflicts respecting the handling of complaints in cases of overlapping jurisdiction;

(d) shall perform duties and functions to be performed by it pursuant to any agreement entered into under subsection 28(2);

(e) may consider such recommendations, suggestions and requests concerning human rights and freedoms as it receives from any source and, where deemed by the Commission to be appropriate, include in a report referred to in section 61 reference to and comment on any such recommendation, suggestion or request;

(f) shall carry out or cause to be carried out such studies concerning human rights and freedoms as may be referred to it by the Minister of Justice and include in a report referred to in section 61 a report setting out the results of each such study together with such recommendations in relation thereto as it considers appropriate;

(g) may review any regulations, rules, orders, by‑laws and other instruments made pursuant to an Act of Parliament and, where deemed by the Commission to be appropriate, include in a report referred to in section 61 reference to and comment on any provision thereof that in its opinion is inconsistent with the principle described in section 2; and

(h) shall, so far as is practical and consistent with the application of Part III, endeavour by persuasion, publicity or any other means that it considers appropriate to discourage and reduce discriminatory practices referred to in sections 5 to 14.

The Act, as may be seen, endows the Commission with a wide range of powers and duties. The Commission must: hear and determine complaints regarding discriminatory practices; develop and conduct information programs to foster public understanding of the Act; undertake research programs which relate to its duties and functions; maintain liaisons with other bodies to foster common policies and practices; carry out studies concerning human rights; review regulations and other instruments made pursuant to Acts of Parliament; and endeavour through persuasion, publicity and other means to discourage or reduce discriminatory practices. The breadth of these powers points in the direction of a more deferential approach.

The Act does not contain an express clause precluding judicial review, but does make provision for the internal review and rehearing of certain decisions. Sections 55 and 56 provide for an internal appeal process whereby, in specified situations, a matter may be reheard by a Review Tribunal. Beyond this, however, there is no express provision which would authorize a reviewing court to substitute its findings or opinions for those of the Tribunal, contrary to what was the case in the legislation examined in Zurich, Dickason, and Etobicoke, supra (in the context of similar provisions in Young Offenders legislation, see also R. v. M. (S.H.), [1989] 2 S.C.R. 446). While a privative clause would have provided further support for a deferential approach, there is nothing in the Act which suggests that deference is not warranted.

As for the Commission itself, Parliament unquestionably intended to create a highly specialized administrative body, one with sufficient expertise to review Acts of Parliament and, as specifically provided for in the Act, to offer advice and make recommendations to the Minister of Justice. In the exercise of its powers and functions, the Commission would inevitably accumulate expertise and specialized understanding of human rights issues, as well as a body of governing jurisprudence. The work of the Commission and its tribunals involves the consideration and balancing of a variety of social needs and goals, and requires sensitivity, understanding, and expertise.

As to the nature of the problem at issue, tribunals appointed under the Act have a mandate to determine whether or not a given practice or policy has discriminated against a complainant on the basis of one of the enumerated grounds set out in s. 3 of the Act. The enumerated grounds of discrimination are not defined within the Act and, as is clear later in these reasons, it was the intention of the legislature that, in the event of any ambiguity, the Commission and its tribunals would have the task of determining the meaning and scope of these concepts. This is at the core of the Board's jurisdiction. Besides, the task of determining whether or not a complainant comes within the scope of one of these grounds is a question of fact to be determined by the tribunal. As Thurlow C.J. wrote in Canadian Pacific Air Lines, Ltd. v. Williams, [1982] 1 F.C. 214, at p. 215, a Tribunal "has jurisdiction to determine whether what is alleged by the complainant is capable of being discrimination and, if so, whether discrimination has been established." Here, the Tribunal was required to determine whether Mr. Mossop had suffered discrimination on the basis of his "family status". This problem is one which the Board is eminently suited to decide and a question best decided in a context‑specific setting.

When all these factors are considered, I would draw the following conclusions. First, the Tribunal has the jurisdiction to determine questions of fact. It hears witnesses and decides on credibility. As is the case for courts of appeal reviewing findings of fact of superior courts, courts should defer to these findings unless they are patently unreasonable. Such determinations lie at the heart of the tribunal's jurisdiction and expertise, and the enabling legislation does not contain any provision suggesting that the principle of great deference to findings of fact of a trial court should be altered. Second, the Tribunal has jurisdiction to interpret its Act and, consequently, the meaning of the term "family status" in s. 3 of the Act; courts should defer to the Tribunal's interpretation since the legislature specifically intended that the Commission and its tribunals carry out the task of interpreting the grounds of discrimination in the Act. While courts do have a role to play in this task, that role is a limited one, confined to overturning an interpretation which, using the test articulated by Dickson J. in C.U.P.E., supra, at p. 237, is "so patently unreasonable that its construction cannot be rationally supported by the relevant legislation and demands intervention by the court upon review".

Turning to the specific issue before us, the Tribunal concluded that Mr. Mossop's relationship is one which could come within the scope of "family status". The Tribunal came to this conclusion based on its interpretation of its own Act, its assessment of viva voce expert evidence, the evidence in the record, and its legislative mandate. The Tribunal has the jurisdiction and expertise necessary to interpret the grounds of discrimination set out in its own legislation, and accordingly, unless its interpretation of the scope of "family status" is patently unreasonable, deference should be shown and the Court should not interfere. The question of whether or not discrimination is in fact made out is also a matter lying at the heart of the Tribunal's jurisdiction. Again, unless the Tribunal has made a patently unreasonable finding in arriving at that conclusion, it is entitled to deference.

Applying what is, in my view, the proper standard of review and dealing first with the Tribunal's interpretation of "family status" in s. 3 of the Act, is this interpretation patently unreasonable?

"Family Status" in s. 3 of the Act

1. Interpretation of Human Rights Legislation

It is well established in the jurisprudence of this Court that human rights legislation has a unique quasi‑constitutional nature, and that it is to be given a large, purposive and liberal interpretation. In this regard, see Insurance Corp. of British Columbia v. Heerspink, [1982] 2 S.C.R. 145; Ontario Human Rights Commission v. Simpsons‑Sears Ltd., [1985] 2 S.C.R. 536; Bhinder v. Canadian National Railway Co., [1985] 2 S.C.R. 561; Canadian National Railway Co. v. Canada (Canadian Human Rights Commission), [1987] 1 S.C.R. 1114 ("Action Travail des Femmes"); Robichaud v. Canada (Treasury Board), [1987] 2 S.C.R. 84; Zurich, supra (for a general review, see Alan L. W. D'Silva, "Giving Effect to Human Rights Legislation ‑‑ A Purposive Approach" (1991), 3 Windsor Rev. L. & S. Issues 45). This long line of cases mandates that courts interpret human rights legislation in a manner consistent with its overarching goals, recognizing as did my colleague Sopinka J. for the majority in Zurich, supra, at p. 339, that such legislation is often "the final refuge of the disadvantaged and the disenfranchised". In interpreting a statute, Charter values must not be ignored. As McIntyre J. observed in RWDSU v. Dolphin Delivery Ltd., [1986] 2 S.C.R. 573, at p. 602, referring to Re Blainey and Ontario Hockey Association (1986), 26 D.L.R. (4th) 728:

Blainey then affords an illustration of the manner in which Charter rights of private individuals may be enforced and protected by the courts, that is, by measuring legislation ‑‑ government action ‑‑ against the Charter .

In Hills v. Canada (Attorney General), [1988] 1 S.C.R. 513, at p. 558, for the majority, I stressed that "the values embodied in the Charter must be given preference over an interpretation which would run contrary to them". (See also Slaight Communications Inc. v. Davidson, [1989] 1 S.C.R. 1038, at p. 1078.)

The respondent Attorney General of Canada argued that, although the Act should be interpreted as remedial legislation, the ordinary rules of interpretation should apply. Therefore, the Court of Appeal correctly examined the "plain meaning" and textual context of the term, as well as the intention of Parliament. As can be discerned from my opinion in Thomson Newspapers Ltd. v. Canada (Director of Investigation and Research, Restrictive Trade Practices Commission), [1990] 1 S.C.R. 425, at p. 570, I would agree that the rules of interpretation which have guided the courts to this day have not been set aside, and that they continue to play a role in the interpretation of legislation, including constitutional and quasi‑constitutional documents. However, I also note Laskin C.J.'s words in Miller v. The Queen, [1977] 2 S.C.R. 680, at p. 690, that the Court had a duty "not to whittle down the protections of the Canadian Bill of Rights by a narrow construction of what is a quasi‑constitutional document". This observation applies a fortiori to human rights legislation, and has continued to be an important caution in the era of the Charter . In Action Travail des Femmes, supra, Dickson C.J. reviewed the jurisprudence on the interpretation of human rights legislation and, at p. 1134, stated the principle as follows:

Human rights legislation is intended to give rise, amongst other things, to individual rights of vital importance, rights capable of enforcement, in the final analysis, in a court of law. I recognize that in the construction of such legislation the words of the Act must be given their plain meaning, but it is equally important that the rights enunciated be given their full recognition and effect. We should not search for ways and means to minimize those rights and to enfeeble their proper impact. Although it may seem commonplace, it may be wise to remind ourselves of the statutory guidance given by the federal Interpretation Act which asserts that statutes are deemed to be remedial and are thus to be given such fair, large and liberal interpretation as will best ensure that their objects are attained.

The Court has repeatedly warned of the dangers of strict or legalistic approaches which would restrict or defeat the purpose of quasi‑constitutional documents. For example, in Tremblay v. Daigle, [1989] 2 S.C.R. 530, at p. 553, this Court made it clear that the meaning of highly controversial terms "cannot be settled by linguistic fiat". (On avoiding narrow and technical interpretations of Charter rights, see also Law Society of Upper Canada v. Skapinker, [1984] 1 S.C.R. 357, at pp. 365‑67; R. v. Duarte, [1990] 1 S.C.R. 30; on avoiding inflexible categorizations, see Re B.C. Motor Vehicle Act, [1985] 2 S.C.R. 486.)

The remarks of Wilson J. in Thomson Newspapers Ltd. v. Canada (Director of Investigation and Research, Restrictive Trade Practices Commission), supra, at p. 470, are apposite here:

The principle of statutory construction, expressio unius, is ill‑suited to meet the needs of Charter interpretation. It is inconsistent with the purposive approach to Charter interpretation which has been endorsed by the Court and which focuses on the broad purposes for which the rights were designed and not on mechanical rules which have traditionally been employed in interpreting detailed provisions of ordinary statutes in order to discern legislative intent.

Although made in the context of the Charter , and while it is clear that there are differences between constitutional and quasi‑constitutional documents, these comments apply here since both types of documents require an interpretive approach that is broad and purposive, and identifies the values which the legislation was designed to protect. McIntyre J. clearly was of this view when, speaking for a unanimous Court in Ontario Human Rights Commission v. Simpsons‑Sears, supra, at pp. 546‑47, he wrote:

It is not, in my view, a sound approach to say that according to established rules of construction no broader meaning can be given to the Code than the narrowest interpretation of the words employed. The accepted rules of construction are flexible enough to enable the Court to recognize in the construction of a human rights code the special nature and purpose of the enactment (see Lamer J. in Insurance Corporation of British Columbia v. Heerspink, [1982] 2 S.C.R. 145, at pp. 157‑58), and give to it an interpretation which will advance its broad purposes. Legislation of this type is of a special nature, not quite constitutional but certainly more than the ordinary ‑‑ and it is for the courts to seek out its purpose and give it effect. The Code aims at the removal of discrimination. This is to state the obvious. Its main approach, however, is not to punish the discriminator, but rather to provide relief for the victims of discrimination.

In short, though traditional interpretational tools ought not be ignored, they must be applied in the context of a broad and purposive approach.

2. Purpose of the Act

The purpose of the Act, set out in s. 2 recited earlier, is to ensure that people have an equal opportunity to make for themselves the life that they are able and wish to have without being hindered by discriminatory practices. The social cost of discrimination is insupportably high, and these insidious practices are damaging not only to the individuals who suffer the discrimination, but also to the very fabric of our society. This Court decried the multiple harms caused by discrimination in the context of hate promotion in R. v. Keegstra, [1990] 3 S.C.R. 697. Dickson C.J. remarked, at pp. 746‑47, that the consequences of such discriminatory practices "bear heavily in a nation that prides itself on tolerance and the fostering of human dignity through, among other things, respect for the many racial, religious and cultural groups in our society". As McIntyre J. confirmed in Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143, at p. 172, "[d]iscrimination is unacceptable in a democratic society because it epitomizes the worst effects of the denial of equality". The Act, in prohibiting certain forms of discrimination, has the express purpose of promoting the value of equality which lies at the centre of a free and democratic society. Our society is one of rich diversity, and the Act fosters the principle that all members of the community deserve to be treated with dignity, concern, respect and consideration, and are entitled to a community free from discrimination.

The Tribunal did not fail to appreciate these factors, and understood the importance of interpreting the Act in light of its purpose. At page D/6089, it explained:

The Supreme Court of Canada has forcefully stated its view of the context of human rights codes generally, the object of such codes, and what flows therefrom. Codes, based on their broad purposes relating to "individual rights of vital importance" (Action travail des femmes, supra, at p. 182), are legislation of "a special nature" (O'Malley, supra, at p. 546). The Court has noted that the broad purposes of the Act are not obscure, but are "patently obvious" in s. 2 of the Act, and has emphasized, drawing upon s. 2, that the goal is that of equal opportunity for each individual to achieve "the life that he or she is able and wishes to have" (Action travail des femmes, at p. 182).

Against this framework, I would like to briefly address the arguments advanced before us that, in its interpretation, the Tribunal failed to adequately consider both textual context and legislative intent, and that this failure constituted an error of law.

3. Textual Interpretation

It was argued that a correct interpretive approach would warrant that a textual interpretation be determinative, and that the coupling of the terms "family" and "status" in the English text of s. 3 of the Act required the Tribunal to construe "family status" as including only those families who have recognizable status at law. This is the way that the Court of Appeal approached the matter. Leaving aside for the moment the broad and purposive approach which, in my view, should guide the interpretation of human rights legislation, even if one were to take a textual approach to the interpretation of s. 3 of the Act, the result of such an interpretive exercise would not lead to the conclusions of the Court of Appeal, but rather would, in my view, support the Tribunal's findings.

First, the word "status" is capable of bearing several meanings. The Concise Oxford Dictionary of Current English (8th ed. 1990) provides the following definition:

1. social position, rank, relation to others, relative importance ... 2. (Law). person's relation to others as fixed by law.... 3. position of affairs;

While the term "status" may be used to indicate status at law, it may indicate more factual matters of rank, social position, or relation to others. The use of the term "status" is not sufficient by itself to restrict the notion of "family status" to only those families that are recognized at law. Reference to the French version of the term, "situation de famille", is warranted here. Le Petit Robert (1990) provides the following definitions of the term "situation":

1. Le fait d'être en un lieu; manière dont une chose est disposée, située ou orientée.... 2. Ensemble des circonstances dans lesquelles une personne se trouve.... 3. Emploi, post rémunérateur régulier et stable....

"Situation" is not a legal term, is broader than the English term "status", and encompasses a host of meanings. When the meaning of the French term is considered, it is apparent that the scope of "family status" has potential to be very broad. In French, the term "situation de famille" would not be used to express a legal notion. "État matrimonial" would.

As noted in the Court of Appeal reasons, until 1983, the French text of the Act prohibited discrimination on the basis of "situation de famille" while the English text of the Act prohibited discrimination on the basis of "marital status." In 1983 the Act was amended, the French text to include both "situation de famille" and "état matrimonial", the English text to include both "marital status" and "family status". The amendment did not simply modify the existing terms, but in fact expanded the Act. Had the intention been to narrow the scope of protection, it would have been simple to use one term or the other. In my view the purpose of the amendment could only have been to envelop the two notions. Furthermore, if both terms "situation de famille" and "état matrimonial", "marital status" and "family status", in the French or in the English texts, were similar, there would have been no need to juxtapose them. One such expression would have been sufficient as the legislator is not presumed to use meaningless words (P.-A. Côté, The Interpretation of Legislation in Canada (2nd ed. 1992), at p. 232).

In any event, I have difficulty with the Court of Appeal's proposition that the meaning of "family status" and its French equivalent can be determined through reference to what it found to be the more restrictive English term. First, as I have noted above, the English term is not necessarily more restrictive. Second, even if it could be said that "family status" encompassed only families with legal status, it would be highly inappropriate to interpret the term by relying on the more narrow meaning of the French and English texts. It is an established principle of interpretation in Canada that French and English texts of legislation are deemed to be equally authoritative (R. v. Turpin, [1989] 1 S.C.R. 1296), and where there is a discrepancy between the two, it is the meaning which furthers the purpose of the legislation which must prevail (R. v. Collins, [1987] 1 S.C.R. 265; see also P.-A. Côté, supra, at pp. 272‑79). In this case, given that the purpose of the Act is to prevent discrimination and provide an equal opportunity to make the type of life one wishes, the broader of the two meanings should prevail.

A textual interpretation seems to me to support the conclusion of the Tribunal that "family status" should not be restricted to a narrow legal meaning. Nothing in the textual context indicates that the protection of the Act is to be extended only to certain types of legally validated families. On the contrary, the term "family status" suggests a broader protection that would prohibit discrimination against individuals on the basis of the internal structuring of their families. But, as I said above, a strict textual interpretation is not warranted here. That leaves then the argument concerning legislative intent.

The intervener Focus on the Family asserted that the Tribunal should have considered the proceedings in Parliament in an attempt to more accurately discern the intention of Parliament at the time of the legislative enactment. They submitted that the extension of the term "family status" to include same‑sex relationships would usurp the legislative function of Parliament, and give the term a meaning never intended by Parliament. This argument implies that the Tribunal exceeded its jurisdiction by interpreting the scope of "family status" as it did.

First, as regards the amendment itself and referring to the evidence before the Tribunal, this intervener argued that legislative debate supports their assertion that the members of the legislature intended the amendment simply to bring the French and English texts into conformity. As Lord Watson observed in Salomon v. Salomon & Co., [1897] A.C. 22, at p. 38, the ""Intention of the Legislature" is a common but very slippery phrase". Legislative intention can be difficult to ascertain, and it is dangerous to rely only on the legislative record in order to infer that intent. While such record may be of some assistance in certain types of cases (for example Re Anti‑Inflation Act, [1976] 2 S.C.R. 373), legislative intent is derived primarily from the legislation itself (Hills v. Canada (Attorney General), supra, at p. 549; Hunter v. Southam Inc., [1984] 2 S.C.R. 145; Re B.C. Motor Vehicle Act, supra).

With this caveat, had Parliament intended that the protection for families be restricted to legally recognized families, the amendment to the Act could have made this clear. However, this was not done. Instead, the amendment increased the scope of protection by adding a new ground of discrimination to each text: both "marital status" and "family status" became prohibited grounds. As the terms are juxtaposed, it is reasonable to conclude that "family status" must be something other than "marital status", just as "situation de famille" must be something other than "état matrimonial". Since "état matrimonial" is closer to a legal notion, "situation de famille" or "family status" can only be broader. It was, of course, open to Parliament to define the concept of "family status" within the Act. It did not choose to do so, even in the face of debate about the meaning of the term. Instead, Parliament determined that the task of dealing with any ambiguity in any concepts in the Act should be left to the administrative board charged with the task of implementing the Act. I refer, in this regard, to the comments of the Minister of Justice as reported in the Minutes of Proceedings and Evidence of the Standing Committee on Justice and Legal Affairs, Issue No. 114, December 20, 1982, at p. 17: "It will be up to the commission, the tribunals it appoints, and in the final cases, the courts, to ascertain in a given case the meaning to be given to these concepts." When asked why he was reluctant to define these terms within the Act itself, the Minister responded as follows:

The reason for my reluctance to have such definitions included, Mr. Chairman, is that it is not in accord with the scheme of the bill. These words are being interpreted by the Canadian Human Rights Commission. We trust them to interpret and issue regulations.

It is true, of course, that a court can always pronounce on the validity of this; but in most cases, the action of the commission is accepted. Generally speaking, we think that is a better way to proceed. [Emphasis added.]

(Minutes of Proceedings and Evidence of the Standing Committee on Justice and Legal Affairs, Issue No. 115, December 21, 1982, at p. 73).

Though the members of Parliament may perhaps not at that precise moment have envisaged that "family status" would be interpreted by the Tribunal so as to extend to same‑sex couples, the decision to leave the term undefined is evidence of clear legislative intent that the meaning of "family status", like the meaning of other undefined concepts in the Act, be left for the Commission and its tribunals to define. In my view, if the legislative record helps here in the search for legislative intent, it rather supports the Tribunal's wide and broad discretion in the interpretation of the provisions of its own Act.

An interpretation of a human rights document, or for that matter any legislation, that may not conform with Parliament's intention can be easily cured by Parliament itself. Because legislation can be amended more readily than a Constitution, legislatures which find the interpretations given by administrative tribunals inconsistent with legislative intent can always amend the legislation, or pass new legislation in order to modify that interpretation. This has been done on a number of occasions (Contributory Negligence Act, R.S.A. 1980, c. C‑23, which allows for partial recovery by contributorily negligent tort claimants; R. v. Krannenburg, [1980] 1 S.C.R. 1053, and the resulting Criminal Code , R.S.C., 1985, c. C‑46, s. 485(1) , concerning loss of jurisdiction over the offence; the Income Tax Act, S.C. 1970‑71‑72, c. 63, s. 18(1)(l)(ii) reversing the decision of the Exchequer Court in Royal Trust Co. v. Minister of National Revenue, [1957] C.T.C. 32, allowing club fees as a deductible expense; the U.K. case of London Graving Dock Co. v. Horton, [1951] A.C. 737 (H.L.) discussed in Waldick v. Malcolm, [1991] 2 S.C.R. 456, at pp. 476‑77.)

Even if Parliament had in mind a specific idea of the scope of "family status", in the absence of a definition in the Act which embodies this scope, concepts of equality and liberty which appear in human rights documents are not bounded by the precise understanding of those who drafted them. Human rights codes are documents that embody fundamental principles, but which permit the understanding and application of these principles to change over time. These codes leave ample scope for interpretation by those charged with that task. The "living‑tree" doctrine, well understood and accepted as a principle of constitutional interpretation, is particularly well suited to human rights legislation. The enumerated grounds of discrimination must be examined in the context of contemporary values, and not in a vacuum. As with other such types of legislation, the meaning of the enumerated grounds in s. 3 of the Act is not "frozen in time" and the scope of each ground may evolve.

Textual context should not detract from the purposive approach mandated by human rights documents, and legislative intent is best inferred from the legislation itself. Any ambiguities as to the exact boundaries of the enumerated grounds of discrimination were specifically left to the Tribunal's interpretation. Consequently, the Tribunal cannot be reproached for having applied recognized principles of interpretation of human rights legislation, in light of the particular purpose of its Act, as it specifically states at pp. D/6088‑89:

. . . the interpretation of the term "family status" as used in the Act must be governed by the principles of interpretation of human rights codes in general and the Act in particular, enunciated by the Supreme Court of Canada in O'Malley, Bhinder, Action travail des femmes and Robichaud. This is not simply a mechanical exercise, because the principles of interpretation are themselves expressed in broad terms.

. . .

This approach recognizes the dynamic relationship between specific words and the context in which they are used. In this principle, the regulator of the dynamic forces is the requirement that there be harmony between the words and their context. Driedger's approach has been quoted at length to support the observation of McIntyre J. in O'Malley, supra, that "the accepted rules of construction are flexible enough to enable the Court to recognize in the construction of a human rights code the special nature and purpose of the enactment". [Emphasis in original.]

That being said, I turn now to the Tribunal's actual interpretation of "family status" in s. 3 of the Act, to determine whether, according to the proper standard of review, it was patently unreasonable.

Across the political spectrum, there is broad appreciation of the vital importance of strong, stable families. Sylvia Law, in "Homosexuality and the Social Meaning of Gender", [1988] Wis. L. Rev. 187, suggests that this appreciation is hardly surprising since families are a central institution in most societies. She comments, at pp. 220‑21:

Families bear primary responsibility for the nurture and acculturation of children, and for the care of the sick and the old. They provide communities that give personal meaning and value to our lives. Families mediate between the state and the isolated individual. They perform essential income redistribution functions, from adults in their prime earning years to the old, the young and the women who care for them. [Footnotes omitted.]

In her article "`A Family Like Any Other Family:' Alternative Methods of Defining Family in Law" (1990‑1991), 18 N.Y.U. Rev. L. & Soc. Change 1027, at p. 1029, Kris Franklin notes:

Families have long been viewed as among the most essential and universal units of society. This sense of the shared experience of family has led to an often unexamined consensus regarding what exactly constitutes a family. Thus, while "(w)e speak of families as though we all knew what family are," we see no need to define the concepts embedded within the term. [Footnotes omitted.]

This "unexamined consensus" leads many to feel that the term "family" in fact has a plain meaning. This belief is reflected in the decision of the Court of Appeal where Marceau J.A. asks the question, at. p. 34, "Is it not to be acknowledged that the basic concept signified by the word has always been a group of individuals with common genes, common blood, common ancestors?" However, the unexamined consensus begins to fall apart when one is required to define the concepts which are embedded in the term "family" in the context of "family status". How are the boundaries of that status to be drawn? Is there a plain meaning for "family status"? Could it not be said that "family status" is an attribute of those who live as if they were a family, in a family relationship, caring for each other?

Those who support a formalistic or more restrictive meaning for the term family point out that there is a dominant conception of family that has been traditionally enforced by laws and social custom. The form taken by this dominant conception is commonly referred to as the traditional family. It is widely understood that the traditional family is one composed of a married man and woman and their children (see Bruce Ryder, "Equality Rights and Sexual Orientation: Confronting Heterosexual Family Privilege" (1990), 9 Can. J. Fam. L. 39). For example, Census Canada, for the purpose of the census, adopted a definition of the family as:

. . . a husband and wife (with or without children who have never been married, regardless of age) or a parent with one of more children never married, living in the same dwelling. A family may consist, also, of a man or woman living with a guardianship child or ward under 21 years of age for whom no pay was received.

(Jean E. Veevers, The Family in Canada, 1971 Census of Canada, Vol. 5, Pt. 3, Catalogue 99‑725, bulletin 5.3‑3 (1977), at p. 3).

The traditional conception of family is not the only conception. The American Home Economics Association (AHEA) defines a family as "two or more persons who share resources, share responsibility for decisions, share values and goals, and have commitments to one another over time" (adopted in 1973, see I. Diamond, Families, Politics and Public Policy: A Feminist Dialogue on Women and the State (1983), 1, at p. 8). K. G. Terkelsen in "Toward a Theory of the Family Life Cycle" in E. Carter and M. McGoldrick, eds., The Family Life Cycle: A Framework for Family Therapy (1980), 21, at p. 23, defines a family as a:

small social system made up of individuals related to each other by reason of strong reciprocal affections and loyalties, and comprising a permanent household (or cluster of households) that persists over years and decades.

M. Stuart in "An Analysis of the Concept of Family" in A. Whall and J. Fawcett, eds., Family Theory Development in Nursing: State of the Science and Art (1991), 31, at p. 40, says that there are five critical attributes to the concept of family:

1.The family is a system or unit.

2.Its members may or may not be related and may or may not live together.

3.The unit may or may not contain children.

4.There is commitment and attachment among unit members that include future obligation.

5.The unit caregiving functions consist of protection, nourishment and socialization of its members.

In Families in Canada Today: Recent Changes and Their Policy Consequences (2nd ed. 1988), Dr. Margrit Eichler, who testified before the Tribunal, suggests that attempts to define the family invariably result in exclusion, and that it is more appropriate to talk about families, or familial interaction. She states that this interaction occurs in several dimensions: the procreative dimension; the socialization dimension; the sexual dimension; the residential dimension; the economic dimension; and the emotional dimension. In Nurses and Families: A Guide to Family Assessment and Intervention (2nd ed., forthcoming), Wright and Leahey comment as follows (at pp. 3‑3 and 3‑4):

Designating a group of people with a term such as "couple", "nuclear family", "single‑parent family", specifies attributes of membership but these distinctions of grouping are not more or less "families" by reason of labelling. It is the attributes of affection, strong emotional ties, a sense of belonging and durability of membership that determine family composition.

We have found the following definition of family to be most useful in our clinical work: the family is who they say they are.

The multiplicity of definitions and approaches to the family illustrates clearly that there is no consensus as to the boundaries of family, and that "family status" may not have a sole meaning, but rather may have varied meanings depending on the context or purpose for which the definition is desired. This same diversity in definition can be seen in a review of Canadian legislation affecting the "family". The law has evolved and continues to evolve to recognize an increasingly broad range of relationships. Different pieces of legislation contain more or less restrictive definitions depending on the benefit or burden of the law to be imposed. These definitions of family vary with legislative purpose, and depend on the context of the legislation. By way of example, one may be part of a family for the purpose of receiving income assistance under welfare legislation, but not for the purpose of income tax legislation.

Despite the demonstrable range of definitions for family, the respondent Attorney General of Canada took a more narrow approach, arguing that the legal status of same‑sex partners, either as family or as spouses, has not received extensive judicial consideration. Furthermore, he submits that where it has been considered, courts have almost consistently rejected the proposition that the concept of spouse can include same‑sex partners.

The Tribunal, of course, was not concerned here with the definition of "spouse" but rather with "family status" in the particular context of an employee benefit. Be that as it may, both legislatures and courts have long been involved in the process of defining the relationships which will or will not be recognized for various purposes. The resulting legislation and decisions have often had an impact on the ways that families structure themselves. However, the family is not merely a creation of law, and while law may affect the ways in which families behave or structure themselves, the changing nature of family relationships also has an impact on the law. A review of our own legislation reveals the extent to which laws have changed to reflect the realities of families. The treatment of common law spouses is but one example. Law and Family have long been engaged in an Escherian dialectic, each shaping the other while at the same time being shaped. Therefore, in attempting to define "family status", the Tribunal found that it would be inadequate to consider the law without examining the family itself. In this case, the Tribunal considered that the interpretation of "family status" in the Act must account not only for current legal and societal conceptions, but also for the lived experience of family.

While it is arguable that the "traditional family" has an ideological stronghold, it is clear that a large number of Canadians do not live within traditional families. One cannot ignore the fact that, between 1970 and 1987, the divorce rate in Canada rose from 18.6 to 43.1 percent (Marriage and Conjugal Life in Canada: Current Demographic Analysis, by Jean Dumas and Yves Péron (1992), at p. 55). The Statistics Canada 1990 Family and Friends survey indicated that half of divorcees aged 30‑39 and more than one‑third of those aged 40‑49 were living common‑law (Marriage and Conjugal Life in Canada, supra, at p. 50). Many children do not live in nuclear families. For example, in the United States in 1982, it was reported that 25 percent of children under the age of 18 did not live with both biological parents (Bureau of the Census, U.S. Dep't. of Commerce, Current Population Reports, Special Studies Series P‑20, No. 380, Marital Status and Living Arrangements: March 1982, at 5, Table F (1983), cited in Bartlett, "Rethinking Parenthood as an Exclusive Status: The Need for Legal Alternatives When the Premise of the Nuclear Family Has Failed" (1984), 70 Va. L. Rev. 879, at pp. 880‑81). Single‑parent families, especially mother‑led, are prevalent; an increasing number of parents never marry; divorce is common, as is remarriage; significant numbers of families are comprised of a husband and wife with no children at home; lesbians and homosexuals establish long‑term and committed relationships, and many are involved in raising and nurturing children (M. Mendola, The Mendola Report: A New Look at Gay Couples (1980); Phyllis Chesler, Mothers on Trial: The Battle for Children and Custody (1986)). As stated in the 1990 Statistics Canada Report Women in Canada: A Statistical Report (2nd ed. 1990), at p. 7:

The traditional family, made up of a married couple -‑ with, or without children present -‑ continued to decline in importance to the profit of other forms of living arrangements, both within and outside of the family.

The evidence before the Tribunal was that the traditional family form is not the only family form, but co‑exists with numerous others. Though there are Canadians whose experience of family does in fact accord with the traditional model, the way many people in Canada currently experience family does not necessarily fit with this model. For example, Harriet Michel, in "The Case for the Black Family" (1987), 4 Harv. BlackLetter J. 21, gives a substantively different conception of family. She suggests that, in the U.S. black community,

[b]ecause slavery often denied us the right to live with our biological kin, our definition of family became any group of individuals who collectively come together in order to provide economic and emotional support to the members of the groups.

Similarly, Carol B. Stack, in All Our Kin: Strategies for Survival in a Black Community (1974), examined the family networks of the black urban poor. She describes networks of kinship among these groups that are quite different from the traditional understandings of family. These kinship groups include relations based on blood, but extend to include people who would not fit within the traditional model of family. Stack concludes that in studying these families, it is important to look at the way they live, without trying to fit the relationships into the traditional categories. She notes, at p. 31,

An arbitrary imposition of widely accepted definitions of the family, the nuclear family, or the matrifocal family blocks the way to understanding how people in The Flats describe and order the world in which they live.

An important insight could be drawn from these comments. While the structure of the family may be a question of choice for some, for others the structure of family may be in part a natural response to social and political pressures. The definition of "family" is thus capable of adapting to changes in the family structure which are caused by such pressures.

In defining the scope of the protection for "family status", the Tribunal considered it essential not only to look at families in the traditional sense, but also to consider the values that lie at the base of our support for families. To look beyond the specific forms a family might take is to ask what value one sees in the family and what lies at the base of society's desire to recognize and support families. In order to define "family status", it is no error to examine the underlying values of families so that, as Lisa R. Zimmer says in "Family, Marriage, and the Same‑Sex Couple" (1990), 12 Cardozo L. Rev. 681, at p. 699, "actual families, rather than theoretical stereotypes, may enjoy their protected status" (emphasis in original).

There are a variety of values that may lie at the base of society's support of the family. The state focuses on the family as an organizing structure of society. It is argued that the state has an interest in the family as a vehicle to promote social stability. The comments made in Moge v. Moge, [1992] 3 S.C.R. 813, at p. 848, are germane here:

Many believe that marriage and the family provide for the emotional, economic, and social well‑being of its members. It may be the location of safety and comfort, and may be the place where its members have their most intimate human contact. Marriage and the family act as an emotional and economic support system as well as a forum for intimacy. In this regard, it serves vital personal interests, and may be linked to building a "comprehensive sense of personhood". Marriage and the family are a superb environment for raising and nurturing the young of our society by providing the initial environment for the development of social skills. These institutions also provide a means to pass on the values that we deem to be central to our sense of community.

The Tribunal found that these values are not exclusive to the traditional family and can be advanced in other types of families. For example, while many see marriage as an indicator of stability, it appears from the current rate of marriage breakdown that heterosexual union is not an absolute guarantee of stability. As Hugessen J. noted in Schaap v. Canadian Armed Forces, [1989] 3 F.C. 172, at p. 184:

It is a commonplace that the existence of the marriage bond is no guarantee of the permanency and stability of a relationship, just as its absence is no sure indicator of a mere passing fancy.

Stability is a desirable value, but may be achieved in a variety of family forms (Lisa R. Zimmer, supra, at note 103; "Homosexuals' Right to Marry: A Constitutional Test and a Legislative Solution" (1979), 128 U. Pa. L. Rev. 193). It could certainly not have been lost on the Tribunal that long‑lasting and stable relationships have been maintained outside the bounds of legal marriage, as well as within same‑sex relationships.

These comments also apply to intimate and emotional relationships. These relationships are often conceptualized as heterosexual, and it is often thought that people will form their most intimate and emotionally rewarding relationships with a heterosexual partner. However, the Tribunal was of the view that many individuals have found their most fulfilling relationships outside of marriage, and even outside of sexual relationships completely. In its view, it is to be bound by myth to assume that only heterosexual couples are capable of forming loving caring stable relationships. Claudia Lewis, in "From This Day Forward: A Feminine Moral Discourse on Homosexual Marriage" (1988), 97 Yale L. J. 1783, at p. 1791, argues that it is in the nature of humankind to attempt to form family bonds, that this desire is deeply socially ingrained, and that it is not dependent on heterosexual orientation. Similarly Lisa R. Zimmer, supra, at p. 706 says:

. . . choice of a life partner is just as important for homosexuals as it is for heterosexuals. It is a decision which "fundamentally affect[s] a person" because it is a decision of self‑defining import.... [T]he underlying values of family are not altered by sexual orientation...

If there is value in encouraging individuals to form stable and emotionally intimate relationships, such relationships can be forged and maintained in a wide variety of family forms. The emotional and economic safety nets forged by same‑sex couples and their families were found not to be without value to society at large.

It was argued by the intervener Focus on the Family that one of the values of the family is its importance to society in fostering procreation, and that procreation requires families to be heterosexual. The argument is that procreation is somehow necessary to the concept of family, and that same‑sex couples cannot be families as they are incapable of procreation. Though there is undeniable value in procreation, the Tribunal could not have accepted that the capacity to procreate limits the boundaries of family. If this were so, childless couples and single parents would not constitute families. Further, this logic suggests that adoptive families are not as desirable as natural families. The flaws in this position must have been self‑evident. Though procreation is an element in many families, placing the ability to procreate as the inalterable basis of family could result in an impoverished rather than enriched vision.

It was also argued by the intervener Focus on the Family that the traditional family provides the most favourable environment for raising children. Many people in society are still influenced by myth and stereotype concerning the ability of same‑sex couples to raise and nurture the children of one or the other. Children raised by lesbians and homosexual men, according to Susan Golombok, are not necessarily different from children raised by heterosexuals (S. Golombok, A. Spencer and M. Rutter, "Children in Lesbian and Single‑Parent Households: Psychosexual and Psychiatric Appraisal" (1983), 24 J. Child Psychol. Psychiat. 551; comments in A. Sage, "Can gay couples be good parents", The Independent on Sunday, March 10, 1991; Gregory M. Herek, "Myths About Sexual Orientation: A Lawyer's Guide to Social Science Research" (1991), 1 Law & Sexuality 133). What is important is that children be nurtured. The critical factor is not the family form, nor the presence of mixed sex role models, but the provision of a loving and nurturing environment. From this perspective, the ideal family is one which meets the needs of its members, and best attempts to realize the values that lie at the base of family. As Jane Larson says:

It is the social utility of families that we all recognize, not any one proper form that "the family" must assume; it is the responsibility and community that the family creates that is its most important social function and its social value.

("Discussion" (1992), 77 Cornell L. Rev. 1012, at p. 1014.)

Families, both traditional and otherwise, have not always succeeded in the promotion of these values. As Adrienne Rich notes in "Husband‑Right and Father‑Right" in On Lies, Secrets, and Silence (1979), at p. 219:

The `preservation of the family' is quoted as an abstract principle without considering the quality of life within the family, or the fact that families may be held together by force, legally sanctioned terrorism, and the threat of violence.

Not all families are or have been sites of happiness, fulfilment and joy for their members. This Court has had many occasions to consider incidents of emotional, physical and sexual violence within families. It goes without saying that the values that lie at the base of society's desire to protect families are not always or even necessarily met within the family.

The reality is, as Didi Herman writes in "Are We Family?: Lesbian Rights and Women's Liberation" (1990), 28 Osgoode Hall L.J. 789, at p. 802, that families are "sites of contradiction". Some people find family life oppressive, others seek supportive family relations but cannot find them. While the family may provide emotionally satisfying experiences, it may also be the site of brutal, violent and terrifying experiences. However, despite the very real potential for oppression within the family, most people continue to believe that the family also has the potential to be the site of our most important human connections, and that it is these intimate connections that offer the greatest possibilities for individual fulfilment. It is the connections themselves, not simply the form they take, that are important. Jewelle Gomez, in "Repeat After Me: We Are Different. We Are the Same." (1986), 14 N.Y.U. Rev. L. & Soc. Change 935, at p. 939, says:

. . . our family as we construct it is our survival mechanism, and few of us would be willing to relinquish it. It is impossible for us to imagine moving forward without our families.

Given the range of human preferences and possibilities, it is not unreasonable to conclude that families may take many forms. It is important to recognize that there are differences which separate as well as commonalities which bind. The differences should not be ignored, but neither should they be used to de‑legitimize those families that are thought to be different, and as Audre Lorde puts it in "Age, Race, Class, and Sex: Women Redefining Difference" in Sister Outsider (1984) 114, at p. 122:

. . . we must recognize differences among [people] who are our equals, neither inferior nor superior, and devise ways to use each others' difference to enrich our visions and our joint struggles.

Though not all people structure their family relations in the same manner, it is clear that there are common problems. Today, families of all descriptions are subject to pressures that are often shattering. Changes in the economy mean that most families feel that they need the wages of two adult workers. At the same time, families must often care for the young, elderly, and ill, with little help from the larger community. Although these problems are shared by all forms of families, they are further exacerbated in families whose legitimacy is called into question. Given these pressures and responsibilities, it would seem that it is in society's interest to improve conditions to enable families to function as best they can, free from discrimination.

In light of all this, it is interesting to note that, in some ways, the debate about family presents society with a false choice. It is possible to be pro‑family without rejecting less traditional family forms. It is not anti‑family to support protection for non‑traditional families. The traditional family is not the only family form, and non‑traditional family forms may equally advance true family values.

The above discussion is not intended to provide an authoritative definition of what constitutes the family, but is rather to illustrate that a purposive approach to the term "family status" can result in an interpretation that can vary depending on the specific context. In the context of the claim made by Mr. Mossop, the Tribunal found that a unilateral and inflexible definition of "family status" could not accord with the purpose of the Act. As the Tribunal explained, "[t]he possibilities inherent in the term "family" are many and complex." (p. D/6092) Cognizant that the purpose of the Act could be subverted by an inappropriate definition, the Tribunal fully appreciated the values at the base of "family", and had a clear understanding of the need to interpret the term in a manner that would accord with the purpose of the Act. The Tribunal expressed the view, at p. D/6094, that:

In applying the principles of interpretation, the task of the Tribunal is to select a meaning which the term "family status" is reasonably capable of bearing, and that best accords with the intention of Parliament, the object of the Act and the scheme of the Act. The Tribunal rejects the view that it must select an exhaustive or all‑inclusive meaning. The Act will be invoked in many different sets of circumstances, and the term "family status" inherently has some scope, as it cannot be said to refer to an immutable characteristic, apart perhaps from consanguinity, which may itself be uncertain. The question for the Tribunal, then, is not what is the reasonable meaning, but what is a reasonable meaning, which best accords with the Act. [Emphasis in original.]

The Tribunal did not conclude that there is one definition which will serve for all purposes, but rather determined that the task was to find a reasonable meaning which advanced the rights contained in the Act. Following this approach, it made this assertion, at p. D/6094:

The Tribunal, giving the term "family status" a reasonable meaning which is neither the narrowest meaning of the term nor a minimizing of rights under the Act, holds that, prima facie, homosexuals in a relationship are not excluded from relying on that prohibited ground of discrimination.

On the evidence before it and in the context of the Act, the Tribunal concluded that the potential scope of the term "family status" is broad enough that it does not prima facie exclude same‑sex couples. In making this finding, the Tribunal used the proper interpretational approach and considered the purpose of the Act and the values at the base of the protection of families. This is a matter that lay at the heart of the Tribunal's specialized jurisdiction and expertise, and it cannot be said that this conclusion is at all unreasonable, a fortiori patently unreasonable. As the Tribunal notes, at p. D/6094:

It must be remembered that to exclude any person from invoking a prohibited ground of discrimination bars any further consideration of the matter under the Act, with potentially serious consequences for individuals. In the view of this Tribunal, such an approach to definition does not give effect to, or advance, the special purpose of the Act as stated in s. 2.

Whatever definition I myself would have adopted, I would not substitute my opinion for the Tribunal's conclusion based on its specialized field of expertise and the evidence before it. Its conclusion is in accord with the broad and purposive interpretation warranted in human rights legislation and is one that, given its purpose, the Act can bear.

Against this broad conceptual definition of the term "family status" in s. 3 of the Act, the Tribunal proceeded to determine, as a question of fact, whether the relationship of Mr. Mossop, Mr. Popert and their relatives was one of a "family status" nature in the specific and factual context of Mr. Mossop's claim for bereavement leave.