Watson v Lucas

Rent Act 1977, section 2(1)(b) and Schedule 1, para 3–Whether a married man separated from his wife could be a member of the family of his deceased mistress, with whom he had lived for nearly 20 years–Difference of judicial opinion in the Court of Appeal–Review of the statutory succession cases from Gammans v Ekins (with facts very like those in the present case) to Carega Properties SA v Sharratt, where the House of Lords disappointed hopes of a definitive statement–Whether present case distinguishable from Dyson Holdings Ltd v Fox, the previous high point of the more relaxed modern outlook–County court judge in present case distinguished it from Dyson on the grounds inter alia that here the couple retained the use of their own names for most purposes and the man elected to remain married to the wife from whom he was estranged–View of majority of the Court of Appeal that the judge erred in giving greater weight to these factors than to the evidence of a permanent and stable, indeed life-long, association–Oliver LJ, treating Dyson with some reserve, disagreed and considered that the judge was not wrong in attributing weight to the factors which distinguished the present case–County court judge’s decision in favour of landlord reversed, but leave given to appeal to House of Lords In this case Dennis Lucas, the defendant in county court proceedings for possession brought by Mrs Edith Watson, appealed against the decision of Judge Granville Slack at Willesden County Court in her favour. The dwelling-house which was the subject of the proceedings was a flat in 21 St Julian’s Road, London NW6. The appellant claimed to be entitled to remain in occupation as a statutory tenant by succession to the deceased tenant of the flat, a Mrs Sullivan. Mark M George (instructed by Powell, Magrath & Co) appeared on behalf of the appellant; C R Semken (instructed by Sylvester, Amiel & Co) represented the respondent. Giving judgment, STEPHENSON LJ said: On February 29 1980 His Honour Judge Granville Slack declared in the Willesden County Court that Mr Lucas, the defendant to proceedings brought against him by the plaintiff, Mrs Watson, was not a member of the family of a Mrs Sullivan. Mrs Sullivan had been the protected tenant of a basement flat in 21 St Julian’s Road, London NW6, when she died on June 21 1977. Mr Lucas had been residing with her there at the time of her death and, as the judge held contrary to Mrs Watson’s evidence, for many years before her death. Mrs Watson bought the freehold of the whole house, no 21, in 1962, subject to Mrs Sullivan’s tenancy of the basement flat, and lived there on and off from 1962 to 1979. When she gave him notice to quit, as she did in 1977 and again in 1979, he was therefore entitled to remain in occupation of the flat as a statutory tenant by succession pursuant to section 2(1)(b) of the Rent Act 1977 and para 3 of Schedule 1 thereto, provided he was a member of her family. In his defence he raised that plea. The judge rejected it. He appeals against the judge’s decision. A good deal turns on this question and the outcome of this appeal because Mr Lucas’ continued occupation of the flat has prevented Mrs Watson from completing work on the house, which was required by the local authority; and that led Mrs Watson to couple with her claim for possession a claim for damages limited to £2,000 and Mr Lucas to counterclaim an injunction and damages limited to £1,000. Those proceedings the judge has adjourned pending the hearing of this appeal. The statutory provisions on which Mr Lucas relies are these: Section 2(1) and (5) provide: 2(1) Subject to this Part of this Act– (a)  after the termination of a protected tenancy of a dwelling-house the person who, immediately before that termination, was the protected tenant of the dwelling-house shall, if and so long as he occupies the dwelling-house as his residence, be the statutory tenant of it; and (b)  Part I of Schedule 1 to this Act shall have effect for determining what person (if any) is the statutory tenant of a dwelling-house at any time after the death of a person who, immediately before his death, was either a protected tenant of the dwelling-house or the statutory tenant of it by virtue of paragraph (a) above. (5)   A person who becomes a statutory tenant as mentioned in subsection 1(b) above is, in this Act, referred to as a statutory tenant by succession. The First Schedule contains these three paragraphs: 2.     If the original tenant was a man who died leaving a widow who was residing with him at his death then, after his death, the widow shall be the statutory tenant if and so long as she occupies the dwelling-house as her residence. 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 at the time of and for the period of 6 months immediately before his death then, after his death, that person or if there is more than one such person such one of them as may be decided by agreement, or in default of agreement by the county court, shall be the statutory tenant if and so long as he occupies the dwelling-house as his residence. 4.     A person who becomes the statutory tenant of a dwelling-house by virtue of paragraph 2 or 3 above is in this Part of this Schedule referred to as ‘the first successor.’ This protection was first conferred by section 12(1)(g) of the Increase of Rent and Mortgage Interest (Restrictions) Act 1920 by defining the expression tenant to include a member of his (or her) family residing with him at the time of his death. Section 13 of the amending Act of 1933 introduced the qualification of residence for not less than six months. The facts are these: Mr Lucas was born in 1933. He married at 20 and had one child, but after about a year his wife left him for Ireland and later had two more children. A few years later he got to know Mrs Sullivan. She was a widow with two children, who had been living in this flat with her husband since 1943 and with his children since his death in 1956. After about a year’s acquaintance he moved into the flat with her. That was in 1958 when he was 24 or 25 and she was 40, having been born in 1918. Her daughter had already married and left home. Her son, Michael, who gave evidence for Mr Lucas, was then a schoolboy and continued to live in the flat with his mother and Mr Lucas until he married in 1964. From 1958 until her death in 1977 Mr Lucas and Mrs Sullivan lived together in the flat as husband and wife, and he lives there still with another woman. The flat consists of two rooms, used as bedrooms when Michael was there, a kitchen and a hall, with its own front door and an outside water-closet. They slept in the same bed and had sexual intercourse, but no children. They both worked and he contributed out of his earnings to the housekeeping expenses. She paid the rent and bills, which were all in her name. They never married. He could have divorced his wife but he never did. She was a Roman Catholic and did not want a divorce. Indeed about 1970 she came over from Ireland and asked for a reconciliation, but he refused: he would not leave Mrs Sullivan, and his wife went back to Ireland. ‘For the most part,’ said the judge in the note which he based on notes of his judgment supplied by counsel appearing for the plaintiff and the defendant before him and in this court, ‘they retained their own names, with one exception, regarding the ex-servicemen’s club, of which Mr Lucas was a member before he met Mrs Sullivan. He could take her there as his wife, and she was regarded as his ‘Missis.’  When they went on excursions, it was as Mr and Mrs Lucas. There is no evidence that he ever used her name until after her death. He used his own name, and usually she continued to use her own name.’  This was the only respect in which they did not live as a married couple. There was no evidence from other relatives or friends as to how they were regarded. Michael said, ‘It was as if they were married.’  Both the Sullivan children called Mr Lucas ‘Dennis’ and Michael’s mother referred to him as ‘Dennis’ when speaking to her son. So their relationship was no different from that of step-father and step-son. Was Mr Lucas a member of Mrs Sullivan’s family?  On the authority of Brock v Wollams [1949] 2 KB 388, at p 395 per Cohen LJ, the question the county court judge should and did ask himself is: ‘Would an ordinary man, addressing his mind to the question whether Mr Lucas was a member of her family or not, have answered ‘yes’ or ‘no’?’  We are bound by the decision of this court in Dyson Holdings Ltd v Fox [1976] QB 503 to place the ordinary man not in 1920, when Parliament first used the phrase in this context, but in 1977, when Mrs Sullivan died. And we have also to place him in possession of the evidence which the judge had before he gives his answer. Gammans v Ekins [1950] 2 KB 328 would seem to have answered the question for us, for in that case this court decided that an unmarried man, who had lived with a tenant as man and wife for some 20 years before her death (see [1950] 2 All ER 140 at p 141 E), and taken her name, and had no children, was not a member of her family because (as I read the judgments) to hold that he was would be an abuse of language and presuppose an intention of Parliament to reward immorality with irremovability: the party to a union out of wedlock is out of the statutory protection. But that decision was held in Dyson v Fox to be no longer valid, either because it was wrongly decided (per Lord Denning MR) or (per James LJ and Bridge LJ) because it was outdated and no longer applied the popular meaning which would guide the ordinary man’s answer. These are only three of the reported cases in which the question who is a member of a deceased tenant’s family has been extensively canvassed since the Act of 1920 first made an answer necessary. The hope expressed in the most recent case in this court, Helby v Rafferty [1979] 1 WLR 13, that all these cases, including Dyson Holdings Ltd v Fox, might be reviewed by the House of Lords was disappointed by the House in Carega Properties SA v Sharratt [1979] 1 WLR 928. It would, in my judgment and in Hamlet’s words, be ‘weary, stale, flat and unprofitable’ to go through them all again. The judge went carefully through them and Mr George for the defendant has taken us through them. I am satisfied that the decisive question for us is whether this case is distinguishable from Dyson v Fox. It is submitted for the defendant that it is not and that the county court judge was only able to distinguish it on a ground which is erroneous in law or does not distinguish it in fact. In that case this court decided that an unmarried woman who had lived with an unmarried man as man and wife for 21 years before his death, but without children, and had taken his name, was a member of his family residing with him at the time of his death because, in the view of the majority, the union was permanent and stable enough to make her a member of his family in the eyes of the ordinary man and in the popular meaning of that phrase in 1975, or in 1961 when the man died. That decision and its ratio have been doubted and distinguished by this court in Helby v Rafferty. It was doubted because it did not follow the earlier decision of this court in Gammans v Ekins on a ground which seemed inconsistent with recognised principles of statutory construction. It was distinguished on the facts there found by the county court judge; not because the deceased tenant was (as in this case) a woman and the defendant, who claimed to be a member of her family residing with her at the time of, and for the period of six months immediately before, her death, was (as here) a man, but because during their five years’ cohabitation the parties never pretended to be married and the tenant wished to remain independent, so that their relationship lacked the permanence and stability necessary to create a family unit. All three members of this court in Helby v Rafferty regarded Dyson v Fox as binding upon them, and we must do the same. To go back to Gammans v Ekins would be to introduce impermanence and instability into our own decision in this case. Only the House of Lords can reinstate Gammans v Ekins and that they have so far declined to do. In Carega Properties SA v Sharratt [1979] 1 WLR 928 at p 930 Lord Diplock said: The facts of the instant case, if they are not unique, are certainly most unusual, and for that reason they do not, in my opinion, provide a suitable occasion for this House to undertake a general consideration of what persons may be included in the expression ‘a member of the original tenant’s family’ where at the time of the tenant’s death there did exist between him and the claimant to a statutory tenancy by succession a relationship of one or other of the various kinds to which I have referred above. In particular, the difficult question posed by Dyson Holdings Ltd v Fox [1976] QB 503 as to the extent, if any, to which changed social attitudes towards cohabitation between unmarried couples and the offspring of such liaisons may have enlarged the meaning of the expression ‘family’ in the Rent Act 1968 does not arise in the instant case and is best left for consideration in the light of the actual facts of a case in which it does arise. But can it be distinguished?  The learned county court judge thought it could. In his judgment he said: One feature not present in any of those cases was that Mr Lucas was a married man, and so long as he was a married man he could not become her husband in law, and I think that is a very important factor. He deliberately chose to remain a married man. Each remained independent of the other, each usually used his or her own name, Lucas or Sullivan. The Lucases and Sullivans were Catholics and deliberately chose not to get a divorce after 1969. This is not an easy case in which to apply the test, but I come to the conclusion that the defendant was not a member of Mrs Sullivan’s family, indeed could not become a member of her family so long as he remained a married man. If the ordinary man was aware of all the circumstances, he would say they were not members of the same family. Mr Lucas never got himself in a position to marry Mrs Sullivan. The facts have to be weighed. Mr Lucas remained a married man, he was never a member of Mrs Sullivan’s family. The conclusion I reach is that Mr Lucas was not a member of Mrs Sullivan’s family. As I read his judgment he denied Mr Lucas the protection of the statute (1) because usually he and Mrs Sullivan kept their own names and so remained independent of each other; (2) because he deliberately chose not to get a divorce from his legal wife and so get into a position to marry Mrs Sullivan. The county court judge seems to have regarded his married status as an absolute bar in law to his being in fact a member of her family. He was there accepting the argument of Mr Semken for Mrs Watson, plainly recorded in the judge’s note of the evidence and argument and repeated in this court, that a man cannot have two families, Mr Lucas had a family in Ireland and not having legally rid himself of wife and children in Ireland he could not be a member of another family here. That is, in my view, a plain error of law. The ordinary man often has two families, one by his first wife and another by his second, and often has, or at any rate knows a man who has, one by his wife and another by his mistress. I can see no legal impossibility in a man being a member of both families, though he will have difficulty in fact in residing with more than one and it will be impossible in fact for him to reside with more than one, as required by the statute, if he is to succeed to a protected tenancy; and the question whether he is a member of his mistress’s family cannot any longer be decided by any moral preference for the family to which his lawful wife belongs. There is no authority on the point, although it could have been taken in the unreported case of Trowchurch Ltd v Gertraud Pammer decided on March 28 1980 in the Bloomsbury and Marylebone County Court and may yet be taken on the appeal to this court which we were told is pending. But I agree with Mr George’s submission that there is no rule of law that a man who remains married throughout the time he has been living with an original tenant other than his wife may not become a member of the original tenant’s family. This error of law entitles us to reverse the judge’s decision if it was wrong in treating Mr Lucas’ married status, and the retention of their own names by himself and Mrs Sullivan with that one exception, as features of their relationship inconsistent with a family relationship between them in the eyes of an ordinary 1977 man. I agree with the judge that Mr Lucas’ remaining married is an important factor, as may be the retention of the couple’s own names. But not in this case. For both factors are of importance, in my judgment, only as indications of independence and instability. The retention of the woman’s name may indicate both, as I think it did in Helby v Rafferty, but if it does not indicate either, it has little relevance to the right answer to the ordinary man’s question. Without the guidance given in Dyson v Fox, I might have thought that both factors might influence his answer, but I understand the ratio of the majority decision as holding that a union between a man and a woman, which in all the circumstances, known and unknown to the ordinary man, looks permanent and stable to him, creates a family unit and both parties are members of it, whether or not it consists of more than those two. It was held long ago by the Divisional Court that a husband could be a member of his wife’s family (in fact there was a young child, but the decision does not seem to have turned on that fact): Salter v Lask [1925] 1 KB 584. And we are prevented by Dyson v Fox from regarding children as necessary to constitute a family, although I think we are still permitted to regard their absence as a possible indication of impermanence. The question ‘is there a family?’ or ‘has he a family?’  or ‘has he any family?’ is the wrong question, because there ‘a family’ means or may mean children. If this approach is right, and Mr Lucas’ decision to remain married is not an absolute impediment to membership of another woman’s family, but a factor to be weighed with all others including their decision to keep their own names in every part of their life together but one, I am of opinion that it was plainly outweighed by his refusal to give Mrs Sullivan up and return to his wife and by his continuing to live with Mrs Sullivan till she died some 10 years later. The judge must have given these two factors of names and marriage far too much weight or this refusal to be reconciled far too little, or he could not have answered the relevant question as he did. These two factors, in my judgment, are of far too little weight to contradict the evidence that this was a lasting, indeed a lifelong, association, more permanent and stable than many marriages including Mr Lucas’ own; and I doubt if the judge could have given them enough weight to justify his decision had he not fallen into the error of regarding Mr Lucas’ membership of Mrs Sullivan’s family as incompatible with his married status. Holding this man to be a member of this woman’s family will not promote the support of marriage or the reduction of illicit unions, which were among the avowed objects of divorce reform stated in the preamble to Sir Alan Herbert’s Matrimonial Causes Act of 1937. But though those objects might have influenced the judges who decided Gammans v Ekins, they had lost their relevance to the interpretation of this provision of the Rent Acts a quarter of a century later. Its irrelevance in other fields has been recognised by Parliament: as demonstrated by the Domestic Violence and Matrimonial Proceedings Act 1976 section 2(2), the Supplementary Benefits Act 1976, Schedule 1 Part I, para 3(1)(b)–and, I would add, more directly by the concept of a family in the Family Income Supplements Act 1970 section 1(1), which provides: 1(1) For the purposes of this Act a family shall consist of the following members of a household– (a)  one man or single woman engaged, and normally engaged, in remunerative full-time work; and (b)  if the person mentioned in paragraph (a) above is a man and the household includes a woman to whom he is married or who lives with him as his wife, that woman; and (c)  the child or children whose requirements are provided for, in whole or in part, by the person or either of the persons mentioned in the preceding paragraphs. The emphasis which in 1976 this court in Dyson v Fox put on the permanence and stability of an extra-marital relationship in constituting a family for the purpose of this provision led the court to regard the county court judge’s answer, that Miss Fox was not a member of Mr Wright’s family, as plainly wrong. I cannot find enough difference between her position and that of Mr Lucas to allow of Judge Granville Slack’s answer, that he was not a member of Mrs Sullivan’s family, being right. I do not find it helpful to consider a couple’s adoption of a common name to conceal the fact that the parties are not married to be a ‘masquerading’ (Asquith LJ’s word in Gammans v Ekins [1950] 2 KB 328 at p 331) or a ‘charade’ (Stamp LJ’s word in Helby v Rafferty [1979] 1 WLR 13 at p 19A). Mr Semken argued that it was essential to a family relationship that there should be a public assumption of the status of marriage by passing off or holding out; but though that may indicate an intention that the association shall last, its absence does not negative that intention and would be less likely to be thought to do so with every year that goes by. It was held in Brock v Wollams that, where the parties could not have been granted a legal adoption, a de facto adoption might constitute a family for the purpose of succession to a statutory tenancy. So a married couple who keep their own names may constitute a family de facto when there was not and could not be a marriage de jure. If the de facto cohabitation is strong enough to repel an attempt by the other party to the legal marriage to put an end to the cohabitation, as in this case, I do not see why it should be prevented from being, as well as being treated as, a family relationship in its ordinary popular meaning because the couple do not pose as married persons. In spite of their separate names they would, by the time Mrs Sullivan came to die, have come to be regarded as husband and wife, ‘as if they were married.’  That is how Mrs Sullivan’s son Michael told the judge that he regarded them when Mr Lucas moved in; and that I suspect was Mrs Watson’s view of them which led her to pretend in the witness-box that, though she had been living in the house since 1962, she knew nothing of his living in her basement flat until a few months before Mrs Sullivan’s death. The ordinary man has to consider whether a man or a woman is a member of a family in the light of the facts, and, whatever may have been held before Dyson v Fox, I do not think a judge, putting himself in the place of the ordinary man, can consider an association which has every outward appearance of marriage, except the false pretence of being married, as not constituting a family. If it looks like a marriage in the old and perhaps obsolete sense of a lifelong union, with nothing casual or temporary about it, it is a family until the House of Lords declares (as Mr Semken reserves his right to ask them to declare) that Dyson v Fox’s case was wrongly decided because the reasoning of the majority was wrong. The time has gone by when the courts can hold such a union not to be ‘familial’ simply because the parties to it do not pretend to be married in due form of law. I have no wish to extend Dyson v Fox (but we are bound to follow it) or to differ from an experienced county court judge on the meaning and application of what ought to be plain English words, but we are bound to move with the times. I would accordingly hold, not without reluctance, that the judge’s decision was wrong. I would allow the appeal and declare that Mr Lucas was a member of Mrs Sullivan’s Family. 101 Dissenting, OLIVER LJ said: The question raised on this appeal is a short and at first sight a simple one. It is whether, in the circumstances which have already been stated in the judgment of my Lord. Stephenson LJ, the defendant who had lived on terms of intimacy for some 20 years with the tenant of the basement flat at the plaintiff’s premises at 21 St Julian’s Road NW6, until the tenant’s death in 1977 was a member of the tenant’s family. If he was then, under the provisions of the third paragraph of the first Part of Schedule 1 to the Rent Act 1977, he is entitled to continue to occupy the flat as a statutory tenant in succession to her. If he was not, as the learned judge held, then the plaintiff is entitled to possession of the flat and the appeal must be dismissed. The particular provision under which the defendant claims to remain in the premises as the successor to the original tenant is one which has formed part of the Rent Act legislation in substantially, although not exactly, the same form since 1920 and it has been the subject of judicial construction in a number of cases. One of the curiosities of the statutory provisions is that while they cater expressly for the position of the widow of the tenant residing with him at his death, there is no corresponding reference to a widower, so that if he is given any protection at all it is solely as a member of his late wife’s ‘family.’  It was, however, held as long ago as 1925 that the expression ‘family’ was apt to include the surviving husband of the tenant (see Salter v Lask [1925] 1 KB 584). The matter was taken a stage further in 1949 in Brock v Wollams [1949] 2 KB 388 in which the expression ‘family’ was held to include an informally-adopted child and in 1953 in Hawes v Evenden [1953] 1 WLR 1169, a unanimous Court of Appeal held that an unmarried lady who had lived with the tenant for 12 years prior to his death and had had two children by him was entitled, in the circumstances that she and the children were living together with him as one family unit, to be considered a member of his family. It is clear, however, that in that case the conclusive features were the birth of the children and the establishment and continuance of what the ordinary man would regard as a single family unit. The court there applied the test propounded by Cohen LJ in Brock v Wollams, namely, would the ordinary man answer ‘yes’ or ‘no’ to the question whether the person claiming the protection of the Act was a member of the tenant’s ‘family.’  Hawes v Evenden answered a question which had been left unanswered in the earlier decision of Gammans v Ekins [1950] 2 KB 328. In that case, in which the position was uncomplicated by the birth of any children, the Court of Appeal firmly and unanimously rejected a claim by a man who, although unmarried had lived with the tenant for some years prior to her death as her husband to be a member of her family. Had the matter rested there, the instant case would have presented no difficulty and, speaking for myself, I should, even unassisted by authority, have required little argument to persuade me to the same conclusion as that reached in Gammans v Ekins. In 1976, however, a very similar case came before the Court of Appeal again in Dyson Holdings Ltd v Fox [1976] QB 503. That was a very hard case of an elderly lady of 74 who, although unmarried in fact, had lived with the tenant to all outward appearances as his wife for 40 years until his death in 1961. The 21 years immediately preceding his death were spent in the premises of which possession was sought and she had continued to reside there after his death up to the date of the hearing. She had taken the tenant’s name and, as Lord Denning expressed it, ‘in every respect they were man and wife save that they had not gone through a ceremony of marriage.’  The court held unanimously that she was a member of the tenant’s family and the majority decision (that of James and Bridge LJJ) was based upon two propositions, that is to say, first that the word ‘family’ fell to be construed, not according to its accepted meaning at the time when the legislation was enacted, but at any rate, as I read the majority judgments, according to its accepted meaning at the time of the hearing and secondly that there had, since Gammans v Ekins, taken place such a change in social attitudes as to justify the conclusion that the ordinary man would, given a sufficient degree of apparent stability and permanence in an extra-marital relationship, say that the parties to it were members of one another’s ‘family.’  In this way the court, while accepting that Gammans v Ekins was binding upon it, felt able to distinguish that case from the case before it. One of the curious features of the case is that, although the tenant in that case had died in 1961 (which would have appeared, prima facie, to be the appropriate time for deciding whether the defendant was a member of his family) the court seems to have disregarded the fact that between 1961 and the hearing the Court of Appeal had, in 1964, considered the case of Gammans v Ekins in Ross v Collins [1964] 1 WLR 425 and had unanimously followed and applied it. Ross v Collins was cited in Dyson but does not appear to have been referred to in any of the judgments. So what the court actually did in 1975 was to apply to the word ‘family’ in the context of the tenant’s death in 1961 a meaning contrary to that which had subsequently been applied to it as the correct meaning by the Court of Appeal in 1964. It is, I think, true to say that the ratio of Dyson’s case has not met with universal approbation. In particular, in Helby v Rafferty [1979] 1 WLR 13 the Court of Appeal expressed certain reservations about the approach to construction adopted by the court in Dyson. However, it recognised that Dyson was binding upon it (although it distinguished the case before it on the facts). Although I think that in the light of Ross v Collins it is arguable (I put it no higher) that there are in fact two conflicting decisions of the Court of Appeal as to the meaning to be applied to the word, this was not argued before us and I am content to adopt the starting position in the instant case that Dyson is binding upon this court. The only question is whether the learned county court judge was right in distinguishing it. What he had to do was to ask himself the question, ‘in the circumstances of the instant case would the ordinary man, cognisant of the relevant facts, say that the defendant in this case was a member of the tenant’s family?’; and in asking himself that question he had to bear in mind the way in which the matter was put by Bridge LJ in the Dyson case when he said (at p 513 of the report): The ordinary man in 1975 would, in my opinion, certainly say that the parties to such a union, provided it had the appropriate degree of apparent permanence and stability, were members of a single family whether they had children or not. The instant case is markedly similar in many respects to the Dyson case. The parties had shared the same dwelling for many years, each contributing to the household expenses and there was every outward indication that, had it not been for the tenant’s death, they would have gone on living together indefinitely. It differs from the Dyson case, however, in three respects. First it was the female partner who was the tenant and the male who survived. Secondly, neither party sought, save on occasions when they went together to, or on outings organised by, a club of which the defendant was a member, to convey the impression that their relationship was anything other than an extra-marital one. The tenant was a widow and continued to be known by her married name of ‘Mrs Sullivan.’  The defendant continued to be known as ‘Mr Lucas.’  Thirdly, whereas the parties in the Dyson case were a bachelor and a spinster respectively and thus able to marry if they wished, the defendant in the instant case was, at all material times, a married man with a wife and child living in Ireland. His evidence was that his wife had left him about a year after his marriage (which had taken place some five years before he went to live with Mrs Sullivan) taking the child of the marriage with her. She had two further children by another man in Ireland but in about 1969 she came back to him seeking a reconciliation which he declined. He continued to live with Mrs Sullivan and his wife returned to Ireland. She, apparently, was a Roman Catholic and did not want a divorce, and he took no steps to obtain one although he was aware that he could petition on the ground of five years’ separation. It was this last matter to which the learned judge attached particular importance. I do not think that it was argued that the first ground of distinction which I have mentioned carried any weight and the learned judge does not mention it. That clearly, I think, must be right because the test enunciated in the Dyson case clearly cannot depend for its application upon whether the survivor is male or female. He does appear to have taken some account of the fact that they normally and habitually used their own names. But the matter to which he attached very great importance was the fact that the defendant remained, at his own election, a married man. He said, ‘So long as he was a married man he could not become her husband in law, and I think that is a very important factor. He deliberately chose to remain a married man. Each remained independent of the other, each usually used his or her own name, Lucas or Sullivan. The Lucases and Sullivans were Catholics, and deliberately chose not to get a divorce after 1969.’ The learned judge then continued–and this is the passage in his judgment which is particularly criticised by the defendant–‘This is not an easy case in which to apply the test’ (that is the test of what the ordinary man would conclude) ‘but I come to the conclusion that the defendant was not a member of Mrs Sullivan’s family, indeed could not become a member of her family so long as he remained a married man.’  What is said about this is that this shows that the learned judge misdirected himself because, instead of looking, as Dyson directs him to do, at the question of whether as a matter of fact the relationship was one of such stability and permanence as to justify the inference that the parties were members of each other’s ‘family,’ he considered only the legal impediment to their actually marrying and treated that, as a matter of law, as an insurmountable impediment. It is, I think, dangerous to seek to construe a judgment–and, particularly, an extempore judgment–as if it were a statute and I do not, for my part, put upon it the interpretation for which the defendant contends. The reference to the inability of the defendant to become a member of the tenant’s family appears to have been something of an afterthought and I do not read it as more than a consideration which the learned judge treated as reinforcing the conclusion of fact at which he had arrived. One has to bear in mind that this passage came at the conclusion of two pages of notes of judgment in which the learned judge exhaustively reviewed the facts which had been proved, a review which would have been wholly unnecessary if the learned judge had really taken the view that a decision in the defendant’s favour was precluded by what would, on this view, have been the only relevant fact, namely that he remained married to someone else. That it formed a very important factor in his decision, however, is beyond doubt, and the real question on this appeal is, as it seems to me, whether he was right to attribute any weight to it and, if so, whether he attached undue weight to it. Now the essential question was one of fact: was the relationship one which displayed the requisite qualities of apparent permanence and stability not simply to constitute what the ordinary man might categorise as a permanent relationship but to constitute what he would categorise as a de facto family relationship?  In reaching the answer to that question I do not think that the learned judge was wrong to take into account both the manner in which the parties to the relationship treated themselves and their actual status. In the instant case the defendant and Mrs Sullivan never appear (except for the purely temporary purpose of satisfying the social susceptibilities of the other members of the defendant’s club) to have sought to convey to the world at large any impression that they were other than what they were in fact, that is to say, two people who were not husband and wife but had chosen to live together. As I read the learned judge’s judgment he did attach some significance to this and I do not, for my part, think that he was wrong to do so. While it may seem startling at first to attribute to persons who engage in a masquerade a higher status than that attributable to persons who do not seek to pose as other than what in fact they are, such a step is, to some extent, inherent in the very process of attributing the status of ‘family’ to those who do not in fact possess it. Normally one expects members of a family to be related by blood or marriage and the application of the term to what is loosely called ‘de facto’ relationship involves the artificial attribution to the persons concerned of a blood or de jure relationship which they do not in fact possess. In reaching a decision, therefore, whether the case is a proper one to make such an attribution, having regard to what the ordinary man would conclude, it cannot, I think, be irrelevant to consider how the parties treated themselves and whether they themselves manifested any desire to be so treated vis-a-vis their friends and neighbours by publicly assuming the appearance and incidents of a family unit. If they themselves openly treated themselves as engaged only in a concubinage which each was free to continue or not and from which each was free to withdraw at will and without obligation, then this does appear to me to be at least a factor to be taken into account in considering whether (to apply the test propounded by Bridge LJ in Dyson) the relationship had a sufficient degree of apparent permanence. That is not to say that pretence is a virtue but merely that one should not be too ready to accord to persons in the position of the defendant and Mrs Sullivan in the instant case the incidents of a status which they never sought to claim for themselves. Equally, they did not seek to put themselves in a position in which they could commit themselves de jure to a permanent relationship, involving mutual obligations although they knew that they could do so and the learned judge inferred from this that they had elected to preserve their independence. Of course, the same could be said of the parties in Dyson and thus it is argued that there is no valid distinction between the two cases. In the ultimate analysis, however, the question is, what would the ordinary man conclude in the circumstances of the particular case?  Here the defendant had, at all material times, a de jure wife who was alive and who, as a matter of law, constituted, with the child of his marriage, his family, so that the case raises directly for decision the question which was postulated by counsel in the Dyson case (at p 507) and not specifically answered by the Court of Appeal. That position he elected to continue and the ordinary man could, I think, be pardoned if he concluded that the defendant was intending, for whatever reason, social, religious, economic, or merely idiosyncratic, to preserve the familial status which the law and his Church had imposed upon him by his marriage rather than to substitute a different and de facto familial relationship to which he was not prepared to commit himself de jure or even as a matter of outward semblance. Mr George has drawn our attention to the provisions of the Supplementary Benefits Act 1976 and the Domestic Violence and Matrimonial Proceedings Act 1976 to support the submission that the legislature, in concurrence with current social mores, now recognises extra-marital relationships such as that with which the instant case is concerned as deserving of the law’s protection. But I am not, speaking for myself, much impressed by that argument, because the question is not whether the law will now countenance relationships which might previously have been disapproved, but whether the ordinary man would regard the relationship between a man and his concubine of many years as being a ‘family’ relationship. Indeed the argument is a double-edged one, for if we are to have reference to other statutes, it can with equal force be pointed out that when the legislature in 1975 came to extend the ambit of the Inheritance (Family Provision) Act 1938 to persons in the position of, for instance, Mrs Sullivan in the instant case, it felt it necessary to alter the title to the Inheritance (Provision for Family and Dependants) Act. I cannot help feeling that if the defendant had, for instance, gone to a solicitor in 1976 in order to have a will prepared and had been asked ‘Have you any family?’  his answer would have been, ‘Yes,102 I have a wife and child in Ireland but I have not seen them for years and I want to leave my property to a lady with whom I am living.’ Essentially every case of this kind turns upon its own facts and the impression that these facts make upon the mind of the judge assuming for the purpose the mantle of the ordinary man. Speaking for myself I would regard the decision in Dyson as one which rested on its own peculiar facts and as standing at the very limit of any ordinarily accepted or acceptable definition of a family relationship. I would certainly not regard it as establishing a principle which is susceptible of further extension or which ought to be extended. The learned judge in the instant case had the difficult task of applying to the facts before him a test which, though neither an easy one nor a certain one, is certainly one which required him to consider all the available factors. The decision was essentially one which depended on the inferences to be drawn from all of the facts before him, which included the fact that the defendant was and remained by choice married to a lady who was living at the material time and who was the mother of his only child. If and so far as the learned judge was of opinion that the continued existence of the defendant’s marriage was, in itself, and by itself, an absolute answer in law to his claim, I respectfully disagree. I think, however, that it was one factor, and a not unimportant factor, to be taken into account, although I accept that a man may sometimes have two families. One possible inference to be drawn from it was that, however agreeable and convenient his relationship with Mrs Sullivan may have been, he was not prepared to make himself responsible for her maintenance and support indefinitely or to put himself in a position in which he might have become susceptible to pressure to convert the relationship into a permanent one by marrying her. I have to confess to more than one change of mind during the course of the very full arguments presented to us, but I have concluded, not without hesitation, that I am not prepared to say that the learned judge, who saw and heard the witnesses, in making his assessment as a whole and in reaching the conclusion which he did reach on all the facts, was wrong to attribute weight to the factors which I have mentioned. They are, in my judgment, factors which do distinguish the instant case from Dyson, and I would dismiss the appeal. Agreeing with Stephenson LJ that the appeal should be allowed, SIR DAVID CAIRNS said: I do not find Cohen LJ’s test as helpful in this case as it was in Brock v Wollams. To the question ‘was Mr Lucas a member of Mrs Sullivan’s family?’  I have no idea what ‘an ordinary man’ would answer. Stephenson LJ would answer ‘yes’ and Oliver LJ would answer ‘no.’ I agree with both my brethren that we should proceed on the basis that Dyson v Fox was rightly decided, though I respectfully share the doubts that have been expressed about that decision and I recognise the force of the reasoning of Oliver LJ based on Ross v Collins. I have reached the conclusion that this case cannot sensibly be differentiated from Dyson v Fox. I attach little or no importance to the continued use by Mr Lucas and Mrs Sullivan of different surnames. It is the relations between the man and the woman that are relevant rather than the appearance that they present to the public. In this case the permanence of the relationship is not in doubt. That Mr Lucas had a wife in Ireland is to my mind, having regard to the brevity of cohabitation, the long separation and the refusal by Mr Lucas of reconciliation, neither an absolute bar to his being considered a member of Mrs Sullivan’s family nor a weighty reason for finding as a fact that he should not be so considered. Even if he could be regarded as a member of his wife’s family (which I doubt) I agree with Stephenson LJ that it is perfectly possible for a man to be a member of more than one family. In my opinion the learned judge was wrong in law in holding that the very fact of Mr Lucas’s continuing state of marriage prevented his being a member of Mrs Sullivan’s family. If this part of his judgment cannot strictly be considered to be one of the rationes decidendi I think it coloured his approach to the question of fact and led him to attach excessive weight to that matter. For these reasons I agree with Stephenson LJ that the appeal should be allowed. The appeal was allowed with costs. Legal aid taxation of the appellant’s costs was ordered and a declaration made that the appellant was a member of Mrs Sullivan’s family for the purposes of the Rent Act 1977. Leave to appeal to the House of Lords was granted.