Judgment - Fitzpatrick (A. P.) v. Sterling Housing Association Ltd.  continued

(back to preceding text)

    I must refer to one further authority: the decision of this House in Carega Properties S.A. v. Sharratt [1979] 1 W.L.R. 928. This was the only occasion on which your Lordships have previously considered the meaning of family in the Rent Act legislation. A widow aged 75 developed a platonic relationship with a young man aged 24. He lived in her flat for nearly 20 years until she died. They treated themselves as aunt and nephew. The trial judge held that through their relationship this elderly lady and young man achieved a familial nexus, meaning thereby a nexus such as one would only find within a family. The Court of Appeal reversed the judge's decision, and held that on the facts the relationship was not within the permissible limits of the meaning of the phrase 'a member of the . . . tenant's family'. This House upheld the decision of the Court of Appeal. Lord Diplock delivered the leading speech. He agreed with observations of Russell L.J., in Ross v. Collins [1964] 1 W.L.R. 425, 432, to the effect that an adult man and woman who establish a platonic relationship cannot establish a family nexus by acting as a devoted brother and sister or father and daughter would act. This is so, even if they address each other as such and even if they refer to each other as such and regard their association as tantamount to such.

    On this appeal your Lordships have not been invited to depart from your Lordships' decision in the Carega case. This does not preclude your Lordships from deciding this appeal to the effect I have already indicated. Had it done so, I would have wished to consider afresh the decision in that case. The reason why the decision in Carega is not an impediment is this. In Carega Lord Diplock, at page 930, stated that the facts of the case did not provide a suitable occasion for the House to undertake a general consideration of what persons may be included in the expression 'a member of the . . . tenant's family'. Further, he disavowed any intention to review the decision in the Dyson case, stating that this was best left for consideration in the light of the actual facts of a case in which it arises. The present case, like Dyson, but unlike Carega, is a sexual partnership case. I do not understand the House in Carega to have been expressing any views regarding this type of case.

    I must also mention the 'ordinary person' test enunciated by Cohen L.J. in Brock v. Wollams [1942] 2 K.B. 388, 395. He suggested that the trial judge should ask himself this question: would an ordinary person, addressing his mind to the question whether the defendant was a member of the family, have answered 'yes' or 'no'? This oft-quoted test has tended to bedevil this area of the law. It may be useful as a reminder that family is not a term of art. But the test gives uncertain guidance when, as here, the members of the Court of Appeal and also your Lordships are divided on how the question should be answered. Contrary to what seems implicit in this form of question, the expression family does not have a single, readily recognisable meaning. As I have emphasised, the meaning of family depends upon the context in which it is being used. The suggested question does not assist in identifying the essential ingredients of the concept of family in the present context.

    In the course of a well-reasoned and attractively presented argument, Mr. Chapman submitted that homosexual relationships have always existed and that at the inception of the Rent Act regime in the 1920s a homosexual partner would not have been regarded as a member of the tenant's family. In those days homosexual acts between men constituted a criminal offence. This remained so until they were de-criminalised by the Sexual Offences Act 1967.

    This submission raises the question whether the word family as used in the Rent Acts may change its meaning as ways of life and social attitudes change. Can the expression family legitimately be interpreted in 1999 as having a different and wider meaning than when it was first enacted in 1920? The principles applicable were stated cogently by Lord Wilberforce in Royal College of Nursing of the United Kingdom v. Department of Health and Social Security [1981] A.C. 800, 822. A statute must necessarily be interpreted having regard to the state of affairs existing when it was enacted. It is a fair presumption that Parliament's intention was directed at that state of affairs. When circumstances change, a court has to consider whether they fall within the parliamentary intention. They may do so if there can be detected a clear purpose in the legislation which can only be fulfilled if an extension is made. How liberally these principles may be applied must depend upon the nature of the enactment, and the strictness or otherwise of the words in which it was expressed.

    In the present case Parliament used an ordinary word of flexible meaning and left it undefined. The underlying legislative purpose was to provide a secure home for those who share their lives together with the original tenant in the manner which characterises a family unit. This purpose would be at risk of being stultified if the courts could not have regard to changes in the way people live together and changes in the perception of relationships. This approach is supported by the fact that successive Rent Acts have used the same undefined expression despite the far-reaching changes in ways of life and social attitudes meanwhile. It would be unattractive, to the extent of being unacceptable, to interpret the word family in the Rent Act 1977 without regard to these changes.

    The change in attitudes towards unmarried couples cohabiting as husband and wife exemplifies this point. In Gammans v. Ekins [1950] 2 K.B. 328 the court's decision was affected by its perception of the immorality of such a relationship. An immoral relationship did not come within the ambit of family in the Rent Acts. Asquith L.J., at page 331, said it would be anomalous that a person could acquire protection by living in sin even if the liaison was protracted in time and conclusive in character. Jenkins L.J., at page 332, described the relationship as no more than a liaison between two elderly people who chose to pose as husband and wife when they in fact were not. Evershed M.R., at page 334, was more hesitant, but his conclusion was that it might be no bad thing to show that one of the privileges derivable from marriage was not equally enjoyed by those living together as man and wife but in fact unmarried.

    In one respect of crucial importance there has been a change in social attitudes over the last half-century. I am not referring to the change in attitude toward sexual relationships between a man and woman outside marriage or toward homosexual relationships. There has been a widespread change in attitude toward such relationships, although differing and deeply felt views are held on these matters. These differing views are to be recognised and respected. The crucial change to which I am referring is related but different. It is that the morality of a lawful relationship is not now regarded as relevant when the court is deciding whether an individual qualifies for protection under the Rent Acts. Parliament itself made this clear in 1988, when amending the Rent Acts in the Housing Act 1988. Paragraph 2(3) of Schedule 1 envisages that more than one person may be living with the tenant as a surviving spouse under the extended definition. In so enacting the law Parliament was not expressing a view, either way, on the morality of such relationships. But by this provision Parliament made plain that, for purposes of Rent Act protection, what matters is the factual position. The same must be true of homosexual relationships.

    It is for this reason that I do not accept the argument that the inclusion of a tenant's homosexual partner within the ranks of persons eligible to qualify as members of his family is a step which should be left to Parliament. It really goes without saying that in cases such as this the courts must always proceed with particular caution and sensitivity. That is not to say the courts can never proceed at all. That is not what the Court of Appeal did in 1975 when deciding the Dyson case. Nor should this course commend itself to your Lordships in the present case.

    In this regard, at the risk of repetition, it is necessary to stress the limited nature of the decision in this case. The courts have already decided that the undefined expression 'family' is to be given a wide meaning in the context of the Rent Acts. The courts have already decided that family includes relationships other than those based on consanguinity or affinity. To include same sex partners is to do no more than apply to them the same rationale as that underlying the inclusion of different sex partners. The decision goes no further than this. The decision leaves untouched questions such as whether persons of the same sex should be able to marry, and whether a stable homosexual relationship is within the scope of the right to respect for family life in article 8 of the European Convention on Human Rights.

    I would allow this appeal. It is not disputed that if a same sex partner can qualify as a member of the tenant's family, the appellant does in fact qualify. He and the original tenant, until the latter's death, lived together for many years in a stable homosexual relationship. The judge found they enjoyed a very close, loving and monogamous homosexual relationship. In my view the appellant falls within paragraph 3.


My Lords,

    The late John Thomson was the statutory or protected tenant of a flat owned by the respondents from 1972 until his death in 1994. Since 1976 the appellant had lived in the flat with him in what is described in the agreed facts as a long-standing, close, loving and faithful monogamous homosexual relationship. The appellant claims to be entitled to succeed to the tenancy on the death of his partner as a statutory tenant by succession under the Rent Act 1977. For this purpose he has to bring himself within the provisions of Part 1 of Schedule 1 to that Act which sets out the provisions for determining who is the statutory tenant by succession after the death of a protected or statutory tenant. The appellant founds on two particular provisions.

    The first of these is contained in paragraph 2(2) of the Schedule which was added by section 39 and Schedule 4 of the Housing Act 1988. Prior to that addition the paragraph referred only to a spouse of the original tenant. It has not been suggested that the appellant could qualify as a spouse. Sub-paragraph (2) extended the scope of the paragraph by stating:

     "(2) For the purposes of this paragraph, a person who was living with the original tenant as his or her wife or husband shall be treated as the spouse of the original tenant."

It is on this extension of the scope of the provision relating to spouses that the appellant bases the first branch of his argument.

    I am not persuaded that the appellant qualifies under this provision. The language here plainly indicates a biological distinction between the sex of the original tenant and that of the successor. The careful use of the words "his" and "her" and the phrase "wife or husband" point to a heterosexual relationship. I see no difference between the language used and the phrase "his wife or her husband" and I cannot read the language as if it had said "his or her partner." Taking the words in their ordinary sense, the sub-paragraph does not in my view include a homosexual relationship. That was the view taken by the Court of Appeal in Harrogate Borough Council v. Simpson (1984) 17 H.L.R. 205 and I consider that that decision was correct.

    The alternative view which was adopted by Ward L.J. in the present case involves a construction of the word "as" which fastens too narrowly on a consideration of the manner of the cohabitation and ignores considerations of form, appearance and capacity. Even if the word "as" does not require a complete equation, nevertheless the approximation must be a closer one than can exist in the case of a homosexual couple. The essential characteristic of the relationship of husband and wife is the bond of marriage. The paragraph in my view is simply seeking to cover situations where the couple are husband and wife in every respect except that they are not married. Thus merely living in the same household will not be sufficient; the manner of their living together and the reason for their so doing may also have to be explored (Crake v. Supplementary Benefits Commission [1982] 1 All E.R. 498). I would only add that the view which I take of this head of the argument appears to accord with El-Al Israeli Airlines Ltd. v. Danilowitz [1994] Case 712/94, where the respondent before the Supreme Court of Israel did not attempt to challenge the view taken by the State Tribunal that a homosexual partner did not qualify as a "spouse (husband or wife)" nor as "a cohabitant publicly known as his/her wife/husband." That case, like that of Egan v. Canada (1995) 124 D.L.R. (4th) 609, where the relevant statute defined "spouse" as including a person of the opposite sex, was concerned primarily with a question of discrimination; but that question is not directly raised in the present appeal.

    The appellant then turns to paragraph 3 of the Schedule. The relevant part of that paragraph in its amended version reads:

     "(3) Where paragraph 2 above does not apply, but a person who was a member of the original tenant's family was residing with him in the dwelling-house at the time of and for a period of two years immediately before his death then, after his death, that person . . . shall be entitled to an assured tenancy of the dwelling-house by succession."

The appellant meets the requirement of residence at the time of the death and for the two years immediately prior thereto. So the question comes to be, was the appellant a "member of the original tenant's family?"

    The word "family" connotes essentially some grouping, usually of persons, who are connected with each other by some particular kind of bond. But the precise content of the group depends upon the context in which the term is used. In some contexts it may require to be restricted to members who are linked by ties of consanguinity or affinity or both consanguinity and affinity. A family so linked may be said to exist even although they do not live together or even meet each other. On a narrower view the context may in some cases even require the group to consist only of children. On the other hand the tie may in particular circumstances consist of a close and intimate degree of companionship between people who are living together in one dwelling even if no relationship of blood or marriage exists between them.

    The word "family" in the context of the rent legislation can be traced to section 12(1)(g) of the Increase of Rent and Mortgage Interest (Restrictions) Act 1920. The meaning of the word "family" in the context of the Rent Acts was explored long ago by Wright J. in Price v. Gould (1930) 46 T.L.R. 411 where it was recognised that the word was a "popular, loose and flexible expression, and not a technical term." In Stewart v. Mackay, 1947 S.C. 287 in relation to the phrase "the needs of the tenant and his family" it was considered that domestic servants and even lodgers might fall within the description if they had a sufficient degree of permanence and the general relationship. In Brock v. Wollams [1949] 2 K.B. 388 the narrow meaning of relations by blood or marriage was rejected, so also was the idea that "family" could be equated with "household." A bond which goes no further than the fact that the group are living under the same roof is not enough. The preferred meaning was that formulated by the "Cohen question" which was repeated by its author as a fair test in Standingford v. Probert [1950] 1 K.B. 377, 383. The test involves the artificiality of an imaginary recourse to a hypothetical representative of the general public; in substance it requires the application of the ordinary popular sense of the word. Once the test is established, the problem seems to me to be one of the application of the word rather than its construction. But, as was pointed out in Sefton Holdings Ltd. v. Cairns (1987) 20 H.L.R. 124, the question is whether the person was a member of the family, not whether he was living as a member of the family.

    Some of the most close family relationships may be created by choice, between persons who may otherwise have been strangers to each other. Marriage is the obvious example. Adoption of children is another. The element of a free mutual choice of a close intimate relationship and the voluntary determination to spend one's life with another is one form of a family bond. The kind of relationship with which the present case is concerned is one where the parties of their own choice live together in a situation of actual or potential sexual intimacy. Beyond that kind of case and the case of a relationship akin to that of parent and child the element of choice does not seem to operate to achieve a family bond. One cannot choose to become a brother or a sister, an aunt or an uncle. Some may choose to be a family member. Others have it thrust upon them.