|[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]|
England and Wales Court of Appeal (Civil Division) Decisions
You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> The Secretary of State for Work And Pensions v Akhtar  EWCA Civ 1353 (10 September 2021)
Cite as:  HRLR 19,  1 WLR 421,  EWCA Civ 1353,  WLR(D) 483
[New search] [Context] [View without highlighting] [Printable PDF version] [View ICLR summary:  WLR(D) 483] [Buy ICLR report:  1 WLR 421] [Help]
Neutral Citation Number:  EWCA Civ 1353
Case No: C3/2019/2519
IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM ADMINISTRATIVE APPEALS CHAMBER
UPPER TRIBUNAL JUDGE WIKELEY
Royal Courts of Justice
Strand, London, WC2A 2LL
LORD JUSTICE UNDERHILL
Vice-President of the Court of Appeal (Civil Division)
LADY JUSTICE MACUR
LORD JUSTICE MOYLAN
- - - - - - - - - - - - - - - - - - - - -
The Secretary of State for Work and Pensions
- and -
- - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - - - -
Ms Z Leventhal, Mr J Anderson and Mr A Habteslasie (instructed by The Government Legal Department) for the Appellant
Ms C Rooney (instructed by Bhatt Murphy Solicitors) for the Respondent
Hearing dates: 15th and 16th December 2020
- - - - - - - - - - - - - - - - - - - - -
Covid-19 Protocol: This judgment was handed down remotely by circulation to the parties’ representatives by email, release to BAILII and publication on the Courts and Tribunals Judiciary website. The date and time for hand-down is deemed to be 10:30am 10th September 2021.
Lord Justice Moylan:
1. This appeal concerns entitlement to Bereavement Payment (“BP”) and Widowed Parent’s Allowance (“WPA”) under sections 36 and 39A of the Social Security Contributions and Benefits Act 1992 (“the SSCBA 1992”). The Secretary of State for Work and Pensions (“the SSWP”) appeals from the decision of Upper Tribunal Judge Wikeley (“the UTJ”) that the claimant, Nasim Akhtar (who I will call “NA” to be consistent with the judgment below) is entitled to both benefits because of the effect of the Social Security and Family Allowances (Polygamous Marriages) Regulations 1975 (“the 1975 Regulations”).
2. In summary, in 2008 NA married the deceased, Mr A, at a time when he was already married. The ceremony of marriage in Pakistan created a valid polygamous marriage under the law of Pakistan but was void under English law because Mr A was domiciled in England at that date. Mr A subsequently divorced his first wife, Ms B, in England in 2009 so that his marriage to NA became monogamous and remained so until his death in 2016.
3. The UTJ decided that the 1975 Regulations could be read down, pursuant to section 3 of the Human Rights Act 1998 (“the HRA 1998”), so as to apply to NA because, otherwise, the legislative framework would be discriminatory in breach of her rights under the European Convention on Human Rights. The UTJ took this course because the primary focus of the argument before him was a human rights challenge advanced on behalf of NA to the previous interpretation of the 1975 Regulations, namely that they only apply to a marriage which is valid under the law of England and Wales. As a result, the UTJ’s judgment focused on whether the legislation was discriminatory and whether it should be read down so as to comply with the ECHR.
4. The SSWP appealed contending that the UTJ had been wrong to conclude that the legislative framework was discriminatory and had been wrong to read down the 1975 Regulations so as to apply to NA.
5. When the appeal was first listed, the court raised questions about the meaning and effect of the 1975 Regulations, in particular as to whether the manner in which they appeared to have been applied was consistent with their proper interpretation. The parties, understandably, were not in a position to deal with this, so the appeal was adjourned to enable a more detailed analysis of the 1975 Regulations to be undertaken.
6. This led to us being provided with a statement from Helen Walker, Deputy Director at the Department for Work and Pensions, which contained a detailed exposition of the background to and history of the 1975 Regulations and referred to much of the background material which I deal with below. I am extremely grateful for the depth of the research reflected in the detail provided in and the documents provided with this statement. I am also very grateful to counsel for their further detailed submissions, including the additional written submissions (and documents) provided following the conclusion of the hearing by Ms Leventhal on 1 February 2021. I would also note that, prior to the hearing in this court, Ms Rooney was commendably acting pro bono.
7. Also by way of introduction, the Supreme Court decided in In re McLaughlin  1 WLR 4250 (“McLaughlin”) that the equivalent legislation in Northern Ireland in respect of WPA (section 39A of the Social Security Contributions and Benefits (Northern Ireland) Act 1992) was incompatible with article 14, read with article 8, of the ECHR “in so far as it precludes any entitlement to widowed parent’s allowance by a surviving unmarried partner of the deceased”, Lady Hale at . A declaration was made to that effect.
8. As explained in the Headnote, the Supreme Court decided that there was no proportionate justification for denying “a mother and her children the benefit of the father’s national insurance contributions purely on the basis that the parents had not been married to one another”. It was, therefore, accepted in the present case that the rejection of NA’s claim to WPA was a breach of her human rights and the only issue was the appropriate remedy.
9. The Supreme Court’s decision was applied by Holman J in Jackson v Secretary of State for Work and Pensions  1 WLR 1441. He decided, applying McLaughlin, that the legislation applicable to bereavement support payment (which has replaced BP and WPA, as referred to below) was also discriminatory and made a declaration of incompatibility.
10. The UTJ expressed concern, at , at “the apparently glacial pace of the Secretary of State’s consideration” of the decision in McLaughlin. He also referred to the House of Commons Work and Pensions Select Committee in April 2019 describing the “profound injustice” of the bereavement benefits system. In the light of the history and, in particular, the passage of time since the decision in McLaughlin, we asked Ms Leventhal what steps the Government had taken to address the issue. Ms Leventhal drew our attention to a response, dated 28 July 2020, to a written question in which the Parliamentary Under-Secretary for Work and Pensions said:
“It is our intention to take forward a Remedial Order to remove the incompatibilities from the legislation governing Widowed Parent’s Allowance and Bereavement Support Payment by extending these benefits to cohabitees with children. We intend to lay the Order before the House in due course.”
11. Since then, in a written response by the Parliamentary Under Secretary of State for Work and Pensions in the House of Lords, Baroness Stedman-Scott, dated 18 April 2021, it was said that the Government hopes “to lay a proposal for a draft Order before Parliament before Summer Recess”. Very recently, on 15 July 2021, a draft order, the Bereavement Benefits (Remedial) Order 2021, has been laid before Parliament.
12. The issues which arise in this appeal can be summarised as follows:
(i) Do the 1975 Regulations, with sections 36 and 39A of the SSCBA 1992, only apply if the relevant marriage is valid under English law and specifically, not void under section 11 of the Matrimonial Causes Act 1973 (“the MCA 1973”) and, accordingly, do they not apply to NA.
If they do not apply to NA:
13. I am largely using the same initials to describe the parties and other relevant individuals as those used in the first instance judgment to maintain consistency: NA v Secretary of State for Work and Pensions  1 WLR 6321.
14. Mr A was born in Pakistan in 1958. He married his first wife, Ms B, in Pakistan in 1976. This was a valid marriage in Pakistan and, in broad terms, was recognised as such by the law of England and Wales even though it was potentially polygamous because, at the date of the marriage, Mr A was domiciled in Pakistan. Subsequently both Mr A and his first wife moved to live in the UK. He became a British citizen in 1993.
15. In 2001 Mr A pronounced a talaq in the UK, with the intention of divorcing Ms B, and they separated. The talaq was not effective under English law to determine their marriage: section 44 of the Family Law Act 1986.
16. In 2008 Mr A married NA in Pakistan. The ceremony effected a valid polygamous marriage in Pakistan but it was void under section 11 of the MCA 1973 because Mr A was domiciled in England and Wales at that date and could not validly contract a marriage which was actually polygamous and, in English law terms, bigamous.
17. In 2009 Mr A and Ms B were validly divorced in England. Ms B died in 2011.
18. In 2010, NA moved to the UK, being granted entry clearance as Mr A’s spouse. She and Mr A had a daughter who was born in 2012.
19. Mr A died in 2016.
Social Security Legislation
20. The judgment below sets out the history of the relevant legislation under the heading, “The shifting framework of provision for bereavement benefits”, at -.
21. The legislation applicable to NA’s claims is that in force at the date of Mr A’s death. BP and WPA have since been replaced by bereavement support payment under the Pensions Act 2014, but these changes are not relevant to this case.
22. The SSCBA 1992 provided as follows:
“Benefits for widows and widowers
36. Bereavement payment
(a) either that person was under pensionable age at the time when the spouse or civil partner died or the spouse or civil partner was then not entitled to a Category A retirement pension under section 44 below or a state pension under Part 1 of the Pensions Act 2014; and
(2) A bereavement payment shall not be payable to a person if that person and a person whom that person was not married to, or in a civil partnership with, were living together as a married couple at the time of the spouse’s or civil partner’s death.
(3) In this section “the appointed day” means the day appointed for the coming into force of sections 54 to 56 of the Welfare Reform and Pensions Act 1999.”
“39A. Widowed parent’s allowance
(a) a person whose spouse or civil partner dies on or after the appointed day is under pensionable age at the time of the spouse’s or civil partner’s death, or
(b) a man whose wife died before the appointed day —
(i) has not remarried before that day, and
(ii) is under pensionable age on that day.
(2) The surviving spouse or civil partner shall be entitled to a widowed parent’s allowance at the rate determined in accordance with section 39C below if the deceased spouse or civil partner satisfied the contribution conditions for a widowed parent’s allowance specified in Schedule 3, Part I, paragraph 5 and —
(a) the surviving spouse or civil partner is entitled to child benefit in respect of a child or qualifying young person falling within subsection (3) below;
(3) A child or qualifying young person falls within this subsection if the child or qualifying young person is either—
(a) a son or daughter of the surviving spouse or civil partner and the deceased spouse or civil partner; or
(b) a child or qualifying young person in respect of whom the deceased spouse or civil partner was immediately before his or her death entitled to child benefit; or
(c) if the surviving spouse or civil partner and the deceased spouse or civil partner were residing together immediately before his or her death, a child or qualifying young person in respect of whom the surviving spouse or civil partner was then entitled to child benefit.
(4) The surviving spouse shall not be entitled to the allowance for any period after she or he remarries or forms a civil partnership, but, subject to that, the surviving spouse shall continue to be entitled to it for any period throughout which she or he —
(a) satisfies the requirements of subsection (2)(a) or (b) above; and
(b) is under pensionable age.
(5) A widowed parent’s allowance shall not be payable –
(a) for any period falling before the day on which the surviving spouse’s or civil partner’s entitlement is to be regarded as commencing by virtue of section 5(1)(k) of the Administration Act; or
(b) for any period during which the surviving spouse or civil partner and a person whom she or he is not married to, or in a civil partnership with, are living together as if they were a married couple or civil partners.”
23. It can be seen that under both section 36 and section 39A a person is only entitled to the benefit if they are a “spouse or civil partner” of the deceased. They are also both contributory benefits so they depend on the national insurance contributions made by the deceased.
24. Polygamous marriages are specifically addressed in social security legislation. I deal with the history of the relevant legislation below. At present, I just set out the relevant provisions.
25. Section 162 of the Social Security Act 1975 (“the SSA 1975”) (since superseded by section 121(1) of the SSCBA 1992) provided as follows:
“Treatment of certain marriages
(a) for a voidable marriage which has been annulled, whether before or after the date when the regulations come into force, to be treated for the purposes of such provisions of, or of any regulations under, this Act, subject to such exceptions or conditions as may be prescribed, as if it had been a valid marriage which was terminated by divorce at the date of annulment;
(b) as to the circumstances in which, for the purposes of this Act —
(i) a marriage celebrated under a law which permits polygamy, or
(ii) any marriage during the subsistence of which a party to it is at any time married to more than one person,
is to be treated as having, or not having, the consequences of a marriage celebrated under a law which does not permit polygamy;
and regulations made for the purposes of subsection (b) above may make different provision in relation to different purposes and circumstances.”
It can be seen that (b)(i) is dealing with a marriage that is potentially polygamous and (b)(ii) is dealing with a marriage that is actually polygamous.
26. The 1975 Regulations were made under this provision. Regulation 1 defines the terms “polygamous marriage” and “monogamous marriage” as follows:
““polygamous marriage” means a marriage celebrated under a law which, as it applies to the particular ceremony and to the parties thereto, permits polygamy;
“monogamous marriage” means a marriage celebrated under a law which does not permit polygamy, and “in fact monogamous” is to be construed in accordance with regulation 2(2) below.”
Regulation 2 provides as follows:
“General rule as to the consequences of a polygamous marriage for the purpose of the Social Security Act and the Family Allowances Act
2. (1) Subject to the following provisions of these regulations, a polygamous marriage shall, for the purpose of the Social Security Act and the Family Allowances Act and any enactment construed as one with those Acts, be treated as having the same consequences as a monogamous marriage for any day, but only for any day, throughout which the polygamous marriage is in fact monogamous.
(2) In this and the next following regulation —
(b) the day on which a polygamous marriage is contracted, or on which it terminates for any reason, shall be treated as a day throughout which that marriage was in fact monogamous if at all times on that day after it was contracted, or as the case may be, before it terminated, it was in fact monogamous.”
While not directly relevant to the present case, I also set out part of Regulation 3:
“Special rules for retirement pension for women
3. (1) Subject to the provisions of paragraphs (2) and (3) of this regulation, where on or after the date on which she attained pensionable age a woman was a married woman by virtue of a polygamous marriage and either —
(a) throughout a day, falling on or after the date on which both she and her spouse have attained pensionable age and in respect of which neither of them has an entitlement to a Category A or Category B retirement pension which is deferred, that marriage was in fact monogamous, or
(b) throughout the day on which her spouse died that marriage was in fact monogamous, that marriage, whether or not it has at all times been or continues to be in fact monogamous, shall, for the purposes of determining her right to and the rate of a retirement pension of any category under the Social Security Act be treated as having the same consequences as a monogamous marriage from and including the date on which she attained pensionable age or, if the marriage was contracted after that date, from and including the date of the marriage …”
27. Pausing there, it is relevant to note, as submitted by Ms Leventhal, that sections 36 and 39A of the SSCBA 1992, section 162 of the SSA 1975 and the 1975 Regulations all use the words “marriage” and “spouse” and that both sections 36 and 39A distinguish between a person who is a “spouse” and a person who is “living together (with another person) as a married couple” (section 36(2)) or who are “living together as if they were a married couple or civil partners” (section 39A(5)(b)).
28. It is also relevant to note that the effect of the 1975 Regulations, when they apply, is that a polygamous marriage is “treated as having the same consequences as a monogamous marriage” (my emphasis) for the purposes of the relevant legislation for each day on which it is “in fact monogamous”. As was pointed out during the hearing by Macur LJ, this is a deeming provision which does not, and does not purport, to address wider issues as to the nature and effect of a polygamous marriage. In particular, they do not purport to make a void marriage valid.
29. As to the continuing effect of the 1975 Regulations, despite the repeal of the SSA 1975, in her written submissions after the hearing, Ms Leventhal pointed to section 17(2)(b) of the Interpretation Act 1978 which provides:
“(2) Where an Act repeals and re-enacts, with or without modification, a previous enactment then, unless the contrary intention appears, —
(b) in so far as any subordinate legislation made or other thing done under the enactment so repealed, or having effect as if so made or done, could have been made or done under the provision re-enacted, it shall have effect as if made or done under that provision.”
30. Ms Walker’s statement dealt with the position in respect of other benefits. It appears that, for those in respect of which a party’s marital status is relevant, their status as a spouse will depend on whether the polygamous marriage is recognised as valid under English law. If it is void, they will not be able to claim as a spouse.
31. NA claimed both BP and WPA. It was initially considered that she was entitled to these benefits but a revised decision was issued in October 2016. This refused her claims on the basis that, because she and Mr A were not validly married for the purposes of English law, she was not a “spouse” under the relevant legislation and was not entitled to these benefits. It was accepted that Mr A had made national insurance contributions which satisfied the relevant contribution conditions as set out in the legislation.
32. NA appealed to the First-tier Tribunal. NA was at that stage represented by her nephew who, as described by the UTJ, at , has fought his “aunt’s corner” with “commitment and tenacity”. Her appeal was dismissed, at , on the basis that her marriage to Mr A was not valid under English law because it was polygamous and because Mr A was “domiciled in the UK at the time of the second marriage”. NA was, therefore, not a “spouse” within the meaning of the relevant legislation and was not entitled to either BP or WPA. The FTT, at , also “dismissed the argument advanced by (NA’s) nephew that she had been unjustifiably discriminated against on the grounds of her marital status”.
33. On her appeal to the Upper Tribunal, NA was represented pro bono, through the Free Representation Unit, by Ms Rooney. Her appeal was allowed. The UTJ summarised his decision as follows:
“ Not everyone will wish to read to the end of what is necessarily a lengthy decision. I therefore summarise the gist of my decision here. I accept Ms Rooney’s submission that the state’s refusal to provide the appellant with a bereavement payment is contrary to article 14 of the Convention read in conjunction with article 1 of the First Protocol. The bereavement payment is within the ambit of article 14, the appellant is in an analogous situation to a “lawful” widow and the difference in treatment is not objectively justified or proportionate. The same is true as regard the refusal of widowed parent’s allowance, but in any event the appellant is the victim of unlawful discrimination on the same basis as the applicant in In re McLaughlin  1 WLR 4250. I further conclude, for the purposes of the appellant’s entitlement to both bereavement payment and widowed parent’s allowance, that the relevant secondary legislation (the Social Security and Family Allowances (Polygamous Marriages) Regulations 1975) can be read down under section 3 of the Human Rights Act 1998 so as to be Convention-compliant. I therefore allow the appellant’s appeal to the UT, set aside the decision of the FTT and remake the decision under appeal in the terms set out above.”
NA was, therefore, entitled to both BP and WPA.
34. The judgment contains a clear and detailed account of the legal framework and history. I do not propose to set out the extensive analysis contained in it but I do summarise, albeit at some length, the UTJ’s analysis of the effect of the social security legislation, including the 1975 Regulations, and his analysis of the discrimination claim. The former runs from - and the latter from - of his judgment.
35. The UTJ, at , records the essence of the SSWP’s submission as follows:
“The essence of the Secretary of State's submission is summed up in para 10250 of the Decision Makers’ Guide, published by the Department for Work and Pensions. Chapter 10 of this guidance deals with evidence of age, marriage and death, and para 10250 states that: ‘A void marriage cannot be treated as valid under any circumstances. For benefit purposes it must be regarded as never having existed.’ The guidance gives R(G) 3/59 14 November 1958, a decision of the National Insurance Commissioner (a forerunner of the UT), as authority for that proposition.”
As explained by the UTJ, at , the claimant in R(G) 3/59 14 November 1958, National Insurance Commissioner, was not entitled to “a child’s special allowance on the death of her partner” because her marriage had been bigamous. As a result, “there was no marriage”. Specific provision had been made for voidable marriages under the National Insurance (Child’s Special Allowances) Regulations 1957. Regulation 3 provided that a voidable marriage which had been annulled was to be treated as a valid marriage which had been determined by divorce at the date of the annulment. No similar provision had been made for void marriages. These Regulations were replaced by the Social Security (Child's Special Allowance) Regulations 1975.
36. The UTJ noted, at , that “(o)ther than a passing reference to R(G) 3/59, the jurisprudence of the former National Insurance Commissioners and Social Security Commissioners and now the UT (Administrative Appeals Chamber) did not feature prominently in the submissions of counsel”. He surmised that this was “because there was no real dispute as to the existing line of authority in the case law” as to the meaning, for example, of “spouse” in section 36 of the SSCBA 1992.
37. The UTJ then turned to consider the 1975 Regulations, again noting that he had not been “taken by counsel to any of the jurisprudence in this jurisdiction (or its predecessors) on the validity of polygamous marriages entered into abroad”. He again surmised that this was “because the line of authority is undisputed” and referred to his previous decision of Secretary of State for Work and Pensions v N  UKUT 68 (AAC). In that case, the UTJ had considered the effect of the 1975 Regulations; he quoted, at , his conclusions:
“. Thus regulation 2(1) provides that ‘a polygamous marriage shall … be treated as having the same consequences as a monogamous marriage for any day, but only for any day, throughout which the polygamous marriage is in fact monogamous’. However, this does not have the effect of converting a void marriage into a valid one simply by virtue of the parties being in practice monogamously married immediately prior to one party’s death. Instead, it means that a valid polygamous marriage can be treated as ‘a monogamous marriage for any day … throughout which the polygamous marriage is in fact monogamous’. As Mr Commissioner Howell put it in unreported decision CG/2611/2003 [29 October 2003] at para 6:
‘A person seeking to claim widow’s benefit under the Social Security Contributions and Benefits Act 1992 has to be either the surviving member of a monogamous marriage recognised as valid under United Kingdom law or the surviving member of a valid marriage under a law which permits polygamy but in fact the only spouse of the deceased at the date of his death: section 121 (1) (b), and regulation 2 of the Social Security and Family Allowances (Polygamous Marriages) Regulations 1975.’
. The key expression in this passage for present purposes is ‘a valid marriage’. If Mr S had been domiciled in Bangladesh in 1983 he would have had capacity to enter into a valid polygamous marriage. If the sequence of events had then continued as before, Mrs N would be able to claim bereavement benefit on his death as she would be, in the words of Mr Commissioner Howell, ‘the surviving member of a valid marriage under a law which permits polygamy but in fact the only spouse of the deceased at the date of his death’. If, however, Mr S had been domiciled in the United Kingdom in 1983, then he would not have had capacity to enter into a polygamous marriage abroad in the first place and, by the law of England and Wales the second marriage in Bangladesh was void from the outset and could not be rescued by regulation 2. In effect it never existed as a valid marriage for the purposes of social security law (see R(G) 3/59).”
The widow’s claim to bereavement benefit in that case failed because her marriage to the deceased in 1983 had been void under section 11(d) of the MCA 1973.
38. Having concluded that NA did not qualify for the claimed benefits because her marriage to Mr A was void under English law, the UTJ considered whether this orthodox application of the legislation breached her rights under the ECHR. The position in respect of WPA was straightforward because it was accepted, at -, that NA’s “position was on all fours with the claimant in” McLaughlin.
39. The position was more complicated in respect of BP. It was accepted that BP fell within the ambit of A1PI for the purposes of the claim under article 14. On the question of whether NA was in an analogous position to a widow of a legal marriage, it was argued on her behalf, at , that, “as the widow of a religious marriage”, she was. The SSWP argued, at , that this was “a false analogy”; to characterise NA’s marriage “as a religious marriage was to miss the point - it was a marriage that was void at the outset because of section 11(d) of the MCA 1973”.
40. The UTJ, at , rejected the “binary approach” advanced by the SSWP. In his view, there was “a spectrum of potentially analogous situations” from, at one end, the surviving spouse of a lawful marriage to, at the other end, the surviving partner of a short-term cohabiting couple: “In between, there is a positive kaleidoscope of other types of quasi-matrimonial relationships”.
41. The UTJ’s ultimate conclusion, at , was that NA “as the sole surviving widow of an overseas religious marriage, is in an analogous position to that of a ‘lawful’ widow under a marriage recognised by the law of England and Wales”. This was based, at , on the fact that the “marriage was valid under the law of Pakistan and valid more generally under Islamic law” and the fact that “the parties regarded themselves as bound by the ‘corpus of rights and obligations which differentiate it markedly from the situation of a man and woman who cohabit’ (to adopt the language of Lindsay v United Kingdom 9 EHRR CD555, as cited in Shackell  ECHR 784)”.
42. It was accepted, at , that, if NA was in an analogous position to the widow of a valid marriage, the difference in treatment was based on “other status” within the scope of article 14.
43. The UTJ dealt with the issue of justification at some length. He accepted, at , “the well-established overarching principle that courts and tribunals will respect the Secretary of State’s judgement on matters of social policy in the welfare benefits context unless that judgement can be demonstrated to be ‘manifestly without reasonable foundation’ (see e g Humphreys v Revenue and Customs Comrs  1 WLR 1545, paras 19–20 per Baroness Hale JSC)”.
44. He then dealt with each of the matters relied on by the SSWP, namely (i) “to promote and prioritise legal marriage”; (ii) for public policy reasons, the law is antithetical to polygamous marriages; (iii) the need for a bright line rule; (iv) the “Beveridge contributory principle”; (v) administrative workability; and (vi) the matters identified by the ECtHR in Yiğit v Turkey (2011) 53 EHRR 25 when upholding a law which limited entitlement to a survivor’s pension and health insurance to a party to a valid marriage (and excluded a party to a religious ceremony of marriage which was not valid under Turkish law).
45. The UTJ, at , rejected (i) for a number of reasons including that “this justification is primarily framed in public policy terms of the state favouring marriage over cohabitation, rather than the state favouring marriage as understood under the Marriage Act 1949 and the MCA 1973 as against marriage which is valid in a jurisdiction abroad but not recognised in the UK”. In addition, NA and Mr A had entered into a “public contract”.
46. As for (ii), at , the “supposed public policy principle is not absolute as our law recognises polygamous marriages in some circumstances and for some purposes”. This was also “not a case of multiple wives”; Mr A had never lived with both wives, had divorced his first wife and NA was his sole surviving widow.
47. As to (iii), at , this was not a bright line case. The UTJ considered that section 11(d) of the MCA 1973 “draws a distinctly dim line” because it turns on domicile, an issue “which can be notoriously difficult to assess and is a mixed question of fact and law”. As to (iv), the UTJ referred to significant developments which had taken place in society since 1948. There had been “significant expansions in the right to rely on a deceased partner’s national insurance record … driven by human rights considerations”; by “comparison the extension of coverage to someone in the appellant’s position is marginal”.
48. The UTJ was also not persuaded by (v) and (vi). As to (v), at , based on over 25 years’ experience of sitting in this jurisdiction, he doubted whether “there would be anything more than a marginal increase in such difficult cases”. Indeed, there might be fewer appeals.
49. As to (vi), at , the UTJ noted that “the principle of secularism which underpinned the modern Turkish state was a vital, if not overriding, consideration in the court’s reasoning”. While the “principle of secularism in relation to matrimonial law in Turkey admitted of no exceptions”, in “contrast, the UK’s disapproval of polygamy on public policy grounds lacks that absolutist nature”. The 1975 Regulations recognised some polygamous marriages for social security purposes and “recognition has also been indirectly accorded by case law”, for example in Din v National Assistance Board  2 QB 213. The UTJ also did not consider that “the rule enshrined in section 11(d) of the MCA 1973 can be regarded as clear, accessible and straightforward in the way that the Turkish Civil Code was found to be”.
50. The UTJ concluded, at , that the distinction between NA and a “lawful” widow was not justified:
“86 In any justification case there must be a reasonable relationship of proportionality between the aim and the means pursued. The distinction that the law makes between a ‘lawful’ widow and someone in the appellant’s shoes is justified to the extent that it prevents more than one spouse claiming national insurance benefits on the basis of the contributions paid by one and the same husband (see Bibi  1 FLR 375 and R(P) 2/06 25 November 2005). However, for the reasons set out above I do not consider that the distinction between the appellant and a ‘lawful’ widow can be justified or is proportionate in circumstances where the appellant is the only surviving spouse of Mr A and in circumstances where the law of England and Wales already recognises the validity of some polygamous marriages based on a criterion (domicile) which lacks a clear bright line and may only be established (or indeed disproved) after the event.”
51. On the issue of remedy, the UTJ first rejected Ms Rooney’s submission that section 36 of the SSCBA 1992 could be read as applying to NA. The UTJ rejected this argument, at , on the basis that: “In the absence of any other definition of ‘spouse’ in the SSCBA 1992, one must fall back on the understanding supplied by matrimonial legislation”. Accordingly, spouse meant a person whose marriage was recognised as valid under English law (including, of course, English private international law rules).
52. The UTJ did, however, accept, at , that the 1975 Regulations could be read in “two ways”, the latter of which would bring NA within them. The “conventional or orthodox reading” of regulation 1(2), as applying only to “a polygamous marriage which is recognised as being valid by the law of England and Wales … involves reading into the definition the words in italics: “‘polygamous marriage’ means a valid marriage recognised according to the law of England and Wales and celebrated under a law which, as it applies to the particular ceremony and to the parties thereto, permits polygamy …”.
53. The alternative reading was as follows, at -. NA’s marriage to Mr A was within the definition of polygamous marriage in regulation 1(2) because it was “celebrated under a law which, as it applies to the particular ceremony and to the parties thereto, permits polygamy”. The law of Pakistan “as it applied to both (a) the particular ceremony and (b) the parties, plainly permits polygamy”.
54. The UTJ considered, at , that the requirement in regulation 2(2)(a), defining when a polygamous marriage will be “in fact monogamous”, “is not limited in point of time to the precise date on which the marriage was contracted”. Applied to the facts of this case, “From the relevant date of Ms B’s divorce in 2009 ‘neither party to it has any spouse additional to the other’ and so was ‘in fact monogamous’”. The UTJ next considered regulation 2(2)(b). This “might be read as requiring monogamy throughout the marriage” but there was “an alternative reading”, namely that the expression “’in fact monogamous’ … can be read in contradistinction to being ‘in law monogamous’”. Although Mr A’s marriage to NA in 2008 could not be “in law monogamous”, it was “in fact monogamous” because Mr A had “only ever lived with one spouse”.
55. The UTJ rejected, at , Ms Leventhal’s submission that this approach would be “inconsistent with the definition of ‘spouse’ as understood in the terms of section 11(d) of the MCA 1973”. In his view:
“the meaning of ‘spouse’ is not central to the interpretation and application of the 1975 Regulations. Rather, the focus of the exercise is the expression ‘polygamous marriage’ and how that should be read in a Convention-compliant manner. I conclude, for the purposes of the appellant’s entitlement to a bereavement payment, that the 1975 Regulations can be read down under section 3 of the Human Rights Act 1998 so as to be Convention-compliant.”
This meant that NA was entitled both to BP, at , and to WPA, at , as her marriage was within the scope of the 1975 Regulations so that it was to be treated as having the same consequences as a monogamous marriage, for each day that it was in fact monogamous, for the purposes of sections 36 and 39A of the SSCBA 1992.
56. The UTJ’s decision, that the legislation breached NA’s rights under the ECHR and that the legislation could be read down, is challenged by the SSWP. In simple terms, it is contended that parties to a marriage which is void under English law are not in an analogous situation to a party to a marriage which is valid under English law and that the difference in treatment between them is justified. Further, it is argued that the 1975 Regulations only apply to polygamous marriages which are valid under English law and do not include a marriage which is void under English law. This is said to be consistent with the language, context and purpose of the legislation and, if required, a necessary implication.
57. It is relevant to look at the position of polygamous marriages both under matrimonial law and under social security legislation. I propose to set this out, largely, in chronological order.
58. Another potential issue was whether the matrimonial law on polygamous marriages was the same in England and Wales and the rest of the UK. This could be relevant because the 1975 Regulations (and other social security legislation) would or, at least, might apply across the UK. In a Note provided after the hearing by counsel for the SSWP, it was explained that the position in Scotland appears to be materially similar to that in England and Wales and that, accordingly, there appear to be no relevant differences in matrimonial law, at least for the purposes of interpreting the effect of the 1975 Regulations.
59. A marriage can be actually or potentially polygamous. The general rule is that a marriage is potentially polygamous if, by the law of the place of celebration, either party is entitled to have another spouse: Dicey, Morris & Collins on The Conflict of Laws 15th Ed. (2012) (“Dicey”), at 17R-135, p. 964. As set out in a joint consultative document published in 1982 by The Law Commission (Working Paper No. 83) and The Scottish Law Commission (Consultative Memorandum No. 56) on Polygamous Marriages (“the 1982 WP”), at [2.3]
“The nature of a marriage as polygamous or monogamous is determined by the law of the country in which it is celebrated. If a marriage is celebrated abroad in a polygamous form in a country whose law permits the particular parties in question to contract a polygamous marriage then it will be regarded in this country as polygamous.”
An exception to this general rule was established by Hussain v Hussain  Fam 26, which interpreted the effect of subsections 11(b) and 11(d) of the MCA 1973 as meaning that the marriage in that case was not polygamous because its provisions meant that the husband, who was domiciled in England, did not have the capacity to contract a polygamous marriage. I deal with this decision further below.
Validity of Marriage
60. Under English rules of private international law: (a) the general rule is that the formal validity (i.e. the formalities) of a marriage is governed by the law of the country where the marriage was celebrated, Dicey at 17R-001, p. 917; and (b) the general rule is that capacity to marry (or essential validity) is governed by the law of each party’s antenuptial domicile, Dicey at 17R-057, p. 939. Bigamy is “a matter of capacity”, Dicey, at 17-082, p. 948.
61. If a marriage is valid in respect of both form and capacity it will be recognised as valid under English law and, as a result, the parties will be recognised as having the status of husband or wife.
62. A void marriage is “is one that will be regarded by every court in any case in which the existence of the marriage is in issue as never having taken place”, Rayden & Jackson on Relationship Breakdown, Finances and Children 18th Ed (“Rayden”), at [3.28]. A voidable marriage “is a valid marriage until such time as it is annulled”, Rayden, at [3.30]-[3.40].
63. The court has long had power to grant a decree of nullity in respect of a void marriage, including in respect of a bigamous marriage. This is relevant because the grant of a decree of nullity is the gateway to the court exercising its powers to make a financial order and, consequently, this provides an example of a party to a void marriage being given rights.
64. In Kassim (orse. Widmann) v Kassim (orse. Hassim)  P 224, at p. 233, Ormrod J (as he then was) said:
“the jurisdiction in nullity and other matters formerly enjoyed before 1857 by the ecclesiastical courts was transferred to this court by section 6 of the Matrimonial Causes Act, 1857, and is now exercised by it under section 21 of the Supreme Court of Judicature (Consolidation) Act, 1925”.
The marriage in that case, which had taken place in England, was bigamous. Ormrod J decided, at p. 234, that “the only order I can make in this case is a decree nisi of nullity” and not merely a declaration as sought by the husband. This was important because, at p. 232, a declaration would give the court no power to make any order for maintenance whereas a nullity decree would give the court “the necessary jurisdiction to deal with … ancillary relief”. The court has had the power to order maintenance in nullity (and other) proceedings since the Matrimonial Causes Act 1907: see K v K (otherwise R)  P 140. Since then, of course, the scope of the financial orders which the court has power to make, including on the grant of a decree of nullity, has increased very significantly: see section 23 the MCA 1973.
65. The court’s approach to polygamous marriages provided an exception to that taken, as referred to above, in respect of both the recognition of marriages and the grant of a decree or other matrimonial remedies. As Ms Leventhal noted in her submissions, in the 19th century and at least for a period in the 20th century, the effect of Hyde v Hyde and Woodmansee (1866) LR 1 P & D 130 was considered to be that a polygamous marriage, whether actually or only potentially polygamous, would not be recognised under English law even if it was otherwise both formally and essentially valid. This can be seen from Rayden on Divorce 1926 Ed., at , p. 34:
“… if a marriage is potentially polygamous, it will not be recognised, even if it is the first marriage, and the husband has not, in fact, taken any other wife; and though it may be recognised as a valid marriage by the lex loci celebrationis”;
and at , p.50:
“The essential validity of a marriage is governed by the lex domicilii of the parties (provided that the lex domicilii does not sanction polygamy)”.
66. Hyde v Hyde is often cited for the definition of marriage given by Lord Penzance. He defined marriage, at p. 133, as “the voluntary union for life of one man and one women, to the exclusion of all others”. As noted in Rayden and Jackson on Divorce and Family Matters 17th Ed (1997), at [4.4], it was “the monogamous concept of marriage that mattered”. This was explained, in particular, by Sir James Hannen P in Brinkley v Attorney-General (1860) 15 PD 76. He first said, at p. 79, that:
“The principle which has been laid down by those cases (Hyde v Hyde and In re Bethell (1887) 38 Ch. D. 220) is that a marriage which is not that of one man and one woman, to the exclusion of all others, though it may pass by the name of a marriage, is not the status which the English law contemplates when dealing with the subject of marriage.”
67. The actual scope of the decision in Hyde v Hyde was limited to the grant of matrimonial remedies. Lord Penzance’s decision, at p. 135, was that a party to a polygamous marriage was not entitled to any remedy in English law. This was based on his conclusion that the matrimonial law of England was not capable of being rationally applied to polygamy because it was based on there being only one wife and one husband. The law was “wholly inapplicable to polygamy”. Although, as set out in the Headnote, the marriage in that case was valid by the law of the place of celebration (Utah) and both parties “were single and competent to contract marriage”, Lord Penzance decided, at p. 138, that parties to a polygamous marriage were “not entitled to the remedies, the adjudication, or the relief of the matrimonial law of England”. This was the extent of his decision. He expressly stated, at p. 138, that he did “not profess to decide upon the rights of succession or legitimacy which it might be proper to accord to the issue of the polygamous unions, nor upon the rights or obligations in relation to third persons which people living under the sanction of such unions may have created for themselves”.
68. The fact that Hyde v Hyde did not set down a broader rule and that a polygamous marriage could be recognised as valid by English law is addressed in Dicey, at 17-190, p. 979:
“In Hyde v Hyde, Sir J.P. Wilde (Lord Penzance) made it clear that all he intended to decide was that “as between each other” the parties to a polygamous marriage were “not entitled to the remedies, the adjudication, or the relief of the matrimonial law of England.” … Yet for some time there was a tendency to assume that all polygamous marriages were wholly unrecognised by English law. However, since the decision of the Committee of Privileges in the Sinha Peerage Claim, it is now clear that they are recognised for many purposes.”
In the Sinha Peerage Claim  1 All ER 348n, the Committee for Privileges of the House of Lords recognised a potentially polygamous marriage in India for the purposes of succession to the title of Baron. Lord Maugham LC said, at p. 349:
“it cannot, I think, be doubted now (notwithstanding some earlier dicta by eminent judges) that a Hindu marriage between persons domiciled in India is recognised in our court, that the issue are regarded as legitimate, and that such issue can succeed to property in this country with a possible exception (of real estate).”
69. In Baindail (orse Lawson) v Baindail  P 122 the husband’s marriage in India was recognised for the purposes of a nullity petition brought by his second wife in respect of her (later) marriage in England. English law recognised the husband’s status as a married man. This was explained by Lord Greene MR, at p. 127:
“The proposition I think would not be disputed that in general the status of a person depends on his personal law, which is the law of his domicile. By the law of the respondent's domicile at the time of his Hindu marriage he unquestionably acquired the status of a married man according to Hindu law; he was married for all the purposes of Hindu law, and he had imposed upon him the rights and obligations which that status confers under that law.”
Lord Greene concluded, at p. 130, that: “On principle it seems to me that the courts are for this purpose bound to recognize the Indian marriage as a valid marriage and an effective bar to any subsequent marriage in this country”.
Legal Framework 1946-1975
70. What, then, was the status of polygamous marriages under social security legislation?
71. The statement from Helen Walker sets out that, prior to 1956, the various National Insurance Acts made no specific provision for polygamous marriages. There were, however, a number of decisions by National Insurance Commissioners to the effect that a party to an actually or to a potentially polygamous marriage was not, for example, a “widow” for the purposes of the National Insurance Act 1946 (“the NIA 1946”).
72. The first and, what clearly became, the leading authority was the decision, on appeal from a Commissioner, of a Tribunal of Commissioners in R(G) 18/52. The marriage had taken place in Sierra Leone and both parties were domiciled there. It was accepted in that case, at , that the “test which must be applied” was:
“Whether the words ‘marriage’, ‘husband’, ‘wife’ and ‘widow’ when used in an Act of Parliament or Statutory Instrument are intended to include polygamous marriages and the parties thereto, must be decided in the light of the language of the Act or instrument in question taken as a whole, and of its manifest scope and objects.”
It was argued on behalf of the claimant, at , that, applying this test, the term “widow” ought to be interpreted as including a widow who had in fact “been her husband’s sole wife, even though he had the right under the law of the place in which his marriage was solemnised and that of his domicile to have more than one wife at a time”.
73. This argument was rejected. The Tribunal acknowledged, at -, “that in certain Acts of Parliament the word ‘Marriage’ has been held to include polygamous marriage” (section 39 of the Marriage Act 1836: R v Rahman  2 All ER 165) and that in some cases polygamous marriages had been recognised (Baindail v Baindail and the Sinha Peerage Case). However, it concluded, at , that: “In none of (these) decisions … did the court purport to lay down any rule by which it can be determined whether a party to a polygamous marriage is to be held to be a ‘wife’ or ‘widow’ within the meaning of a particular statute”. And, at :
“Having regard to the terms and objects of the National Insurance Act 1946, it seems to us reasonably plain that “widow” in section 17 means a woman who was married to her husband by a marriage in the sense in which that term is used in the law of Great Britain, that is to say - ‘the voluntary union for life of one man and one woman to the exclusion of all others’ per Lord Penzance” in Hyde v Hyde.”
74. It could be argued that the broad proposition as to the effect of “the law of Great Britain” was overstated but the Tribunal explained its interpretation of the effect of section 17 as follows, at :
“The absence of any provision in Section 17 for the possibility of more than one widow surviving the husband points clearly to this conclusion, which is reinforced by the fact that express provision is made in Section 17(4) for the case of a widow who has had more than one husband. Again, in Section 22 of the Act relating to death grant, where the legislature contemplated the possibility of more than one person having a title, provision is made in Subsection (6) to meet the case. On the other hand, although by Section 59(1) of the Act the Minister is authorised to make regulations modifying the provisions of the Act in their application in relation to married women, the proviso to that subsection forbids the Minister, save as expressly authorised by the following provisions in the section, to modify any provision of the Act which has special application to a married woman or widow as such.”
75. In response to the Insurance Officer’s submissions, at , that it would be “the height of injustice” to hold that the widow of a potentially polygamous, but actually monogamous, marriage should “be deprived of the benefit to which their husband’s contributions would have entitled them” and that the “legislature … cannot have intended to perpetrate this injustice”, the Tribunal stated that: “We appreciate these considerations”. However, they went on to note, at , that section 58 of the NIA 1946 gave the Minister power to “modify its provisions so far as they apply to mariners” (the husband in that case was a seaman) and that it “may be that the legislature thought that the power conferred upon the Minister by this section would enable him to ensure that justice was done to this class”.
76. In 1954 the National Insurance Advisory Committee was asked by the relevant Minister “to review the present provisions (other than the contribution conditions and rates of benefit) governing widow’s benefits … under the” NIA 1946. In their 1956 Report, the Committee specifically considered marriages contracted abroad and referred to R(G) 18/52. The Report noted, at , that:
“(some) marriage ceremonies celebrated abroad which permit a man to have more than one wife are not recognised for all purposes as valid marriages by the laws of this country even though they are binding in the countries where they are contracted and prevent the contraction of other marriages by the parties in Great Britain. A woman so married (even though the only wife of the deceased) cannot be treated as his wife or widow under the National Insurance or Industrial Injuries Act.” (my emphasis)
As a result, “contributions paid by such men provide nothing for their wives or widows” save for a limited exception in respect of unemployment or sickness benefit which was based on her being “a woman having the care of his children” and not on the marriage.
77. In the context of the present case, the reference to marriages not being recognised “for all purposes as valid marriages by the laws of this country” is clearly significant. As is the reference to the marriage preventing “the contraction of other marriages by the parties in Great Britain”. This would seem to be a reference to the effect of Baindail v Baindail which, as set out above, concerned a polygamous marriage in India which was recognised as valid for the purposes of determining the husband’s status and, accordingly, whether he had the capacity to contract a second marriage in England.
78. The Committee considered, at , “that any departure from the existing definition of marriage for National Insurance purposes is not to be undertaken lightly”. They were, however, “agreed that such a departure could and should be made where the marriage is and has throughout been monogamous in fact”. It is relevant also to note that the approach being recommended was seen as being a departure from the existing definition of marriage. The specific recommendation was as follows:
“Where a marriage, contracted abroad, is not at present recognised for National Insurance purposes because of its potentially polygamous nature, but the husband is or afterwards becomes insured under the National Insurance Acts, widow’s benefit under these Acts should be payable to the widow (if she is otherwise qualified), provided that the statutory authorities are satisfied that there has been such a marriage and that it has been monogamous throughout.”
It can be seen that the recommendation was confined to potentially polygamous marriages and the context, to repeat, was that the Committee referred expressly to “valid marriages”, at .
79. This led to the enactment of section 3 of the Family Allowances and National Insurance Act 1956 (“the NIA 1956”) which provided:
“As from the appointed day, a marriage performed outside the United Kingdom under a law which permits polygamy shall be treated for any purpose of the Family Allowances Acts, 1945 and 1952, the National Insurance (Industrial Injuries) Acts, 1946 to 1954, the National Insurance Acts, 1946 to 1955, and this Act as being and having at all times been a valid marriage if and so long as the authority by whom any question or claim arising in connection with that purpose falls to be determined is satisfied that the marriage has in fact at all times been monogamous.”
This was then replaced by materially identical provisions in section 113(1) of the National Insurance Act 1965 (“the NIA 1965”), section 86(5) of the National Insurance (Industrial Injuries) Act 1965 and section 17(9) of the Family Allowances Act 1965. It is relevant to note the reference to a marriage performed “under a law”, an expression repeated in the 1975 Regulations. This can only be a reference to the law of the place where the marriage was performed or celebrated.
80. The next decision to which we were referred was R(G) 1/70. That case concerned the meaning of widow in section 26 of the NIA 1965 and whether section 113(1) applied in the circumstances of the case. The deceased husband had been married twice in India and the issue was whether his second wife was entitled to a widow’s allowance. It is clear that both marriages were valid under Indian law and that both the husband and each of his wives were domiciled in India at the date of the respective marriages. The deceased’s second wife claimed widow’s allowance.
81. The Commissioner said:
“7. For many purposes English Law also regards as valid a polygamous marriage which is valid by the law of the territory where it was celebrated, but not for all purposes. The question to be determined in this appeal is not whether in general the claimant’s marriage to M.S.R. was valid under English law, but the narrow and technical question whether her marriage is valid for the purposes of the 1965 Act.”
The answer to this question depended on the “interpretation of the language of the Act”. The Commissioner decided, at , that R(G) 18/52, although decided under the NIA 1946, “remains an effective authority for the interpretation of the 1965 Act”. As a result, he concluded that the words “widow” and “marriage”, when used in the NIA 1965, “denote a matrimonial relationship of a monogamous character and do not include polygamous relationships”.
82. The Commissioner noted that this “strict interpretation is, however, modified by section 113(1) of the 1965 Act”, requiring consideration of whether the marriage had “at all times been monogamous”. There was a factual dispute as to whether the husband had effectively divorced his first wife in India prior to his marriage to the claimant. The claim failed because the alleged “customary divorce”, relied on as having dissolved the first marriage, was not recognised under Indian law.
83. The Law Commission dealt with the law of nullity of marriage in its 1968 Working Paper No. 20, Nullity of Marriage and in its 1970 Report on the Nullity of Marriage (Law Com. No. 33). This recommended the codification of the law of nullity in a comprehensive statute which became the Nullity of Marriage Act 1971 (“the NoMA 1971”). Polygamous marriages were outside the scope of this review, which is reflected in the fact that the Act only dealt with actually and not potentially polygamous marriages. Section 1 of the NoMA 1971 set out the “only” grounds on which a marriage after the commencement of the Act (31 July 1971) would be void which included, at (b), “that at the time of the marriage either party was already married”.
84. The NoMA 1971 expressly provided that the rules of private international law would continue to apply:
“4. Marriages governed by foreign law or celebrated abroad under English law.
(1) Where, apart from this Act, any matter affecting the validity of a marriage would fall to be determined (in accordance with the rules of private international law) by reference to the law of a country outside England and Wales, nothing in section 1, 2 or 3(1) of this Act shall —
(a) preclude the determination of that matter as aforesaid; or
(b) require the application to the marriage of the grounds or bar there mentioned except so far as applicable in accordance with those rules”.
This was replaced by section 14 of the MCA 1973 which is to the same effect.
85. The Law Commission addressed the issue of polygamous marriages in its 1971 Report on Polygamous Marriages (Law Com. No. 42) (“the 1971 Report”) which had been preceded by its 1968 Working Paper No 21, Polygamous Marriages (“the 1968 WP”). The scope of these was “limited to the question of recognition of polygamous marriages for the purposes of family law and social security legislation”: the 1971 Report at . The principal focus was whether, what was then called, “matrimonial relief” should be available to parties to a polygamous marriage (i.e. revoking the effect of Hyde v Hyde) although, as mentioned, they also addressed social security because, as explained in the 1971 Report at , “this field has been the subject first of decisions and later of legislation affecting polygamous marriages”.
86. The 1971 Report contained the following definition of a polygamous marriage, at :
“2. For the purposes of this Report a polygamous marriage can be defined as a marriage under a system of law which permits one of the parties to the marriage to take another spouse at a later date even though the marriage still subsists. The term “polygamous marriage'' includes:
(a) a potentially polygamous marriage, in which neither party has, at the relevant time, any other spouse, but in which one party is capable of taking another spouse; and
(b) an actually polygamous marriage, in which one party has, at the relevant time, another spouse or other spouses in addition to the other party.
Both these types of marriage are in law polygamous marriages. The terms "potentially polygamous" and "actually polygamous" will be used to distinguish them where necessary.”
The 1971 Report included within the expression polygamous marriage both a marriage in which a husband was permitted to have more than one wife and a marriage in which a wife was permitted to have more than one husband (at  n.3).
87. The 1968 WP noted, at [8(6)], that the “crucial question” in determining whether a marriage was potentially polygamous was “whether the law under which the marriage is celebrated permits polygamy; if it does not, the marriage is monogamous”. The 1968 WP also noted, at [8(6)], that:
“It was at one time supposed that the monogamous or polygamous character of a marriage had to be determined once and for all at the date of its inception. But now it is clear that a potentially polygamous marriage may become monogamous by reason of subsequent events, and that, therefore, English matrimonial relief may subsequently become available to the parties.”
A number of examples were given including, as determined in Ali v Ali  P 564, “if the husband changes his domicile from a country whose law permits polygamy to a country whose law does not”.
88. As was emphasised in Ms Leventhal’s submissions, the Law Commission considered, but decided against, proposing any change to the legal test for determining whether a polygamous marriage was valid under English law. As set out in the 1971 Report, at :
“In our Working Paper we considered whether it might be acceptable to test the validity of a polygamous marriage solely by reference to the law of the place of celebration, and without any reference to the law of either party’s domicile. But we concluded that there was no justification, nor indeed reason, for changing the present law … Our consultations have confirmed us in this view.”
Accordingly, at , the position remained that, “if a person domiciled in England goes through a polygamous form of marriage abroad, that marriage will, under English law, be void, even if it was only potentially polygamous”. This was because, as set out in the 8th Edition of Dicey and Morris on the Conflict of Laws, at Rule 35, “a man or woman whose personal law does not permit polygamy has no capacity to contract a valid polygamous marriage”.
89. The 1971 Report also noted, at :
“The two basic principles of the present law concerning polygamy are:
(a) Neither party to a polygamous marriage is entitled to any matrimonial relief in England whether the marriage is potentially or actually polygamous.
(b) However, a polygamous marriage which is valid by the law of the place of celebration and by each party's personal law is generally recognised as valid in England, except for purposes of matrimonial relief.”
The 1968 WP and the 1971 Report addressed the question of whether the denial of matrimonial relief should be maintained. It concluded for a variety of reasons, including that this caused hardship, that as set out in the Report at : “there is no longer any justification for denying all forms of matrimonial relief to polygamously married persons resident in this country”.
90. The 1971 Report recommended (with one partial dissent) the abolition of the rule in Hyde v Hyde and that the full range of matrimonial relief be extended to parties to both potentially and actually polygamous marriages. This led to the enactment of the Matrimonial Proceedings (Polygamous Marriages) Act 1972 (“the MP(PM)A 1972”) which provided by section 1 that a court was not precluded from granting matrimonial relief by reason only that the marriage was actually or potentially polygamous. This provision became section 47 of the 1973 Act which, as initially enacted, provided:
“(1) A court in England and Wales shall not be precluded from granting matrimonial relief or making a declaration concerning the validity of a marriage by reason only that the marriage in question was entered into under a law which permits polygamy …
(4) This section has effect whether or not either party to the marriage in question has for the time being any spouse additional to the other party …”
91. In recommending the abolition of the rule in Hyde v Hyde, the Law Commission set out its reasons for doing so at some length. This included emphasising, at , the limited effect of what was being proposed which “would not result in the legalisation or recognition of something which has hitherto been forbidden or totally unrecognised in this country”. It would not “enable or encourage any person domiciled in England to enter into a polygamous marriage abroad: such a marriage would be void”.
92. Although not addressed in either the 1968 WP or the 1971 Report (as to which see paragraph 110 below), section 4 of the MP(PM)A 1972 amended section 1 of the NoMA 1971 by adding an additional ground on which a marriage would be void, namely:
“(d) in the case of a polygamous marriage entered into outside England and Wales, that either party was at the time domiciled in England and Wales.
For the purposes of paragraph (d) of this section a marriage may be polygamous although at its inception neither party has any spouse additional to the other.”
93. The 1971 Report (in Section 14) addressed the recognition of polygamous marriages for purposes other than matrimonial relief. The Report pointed out, at , that:
“In spite of Lord Penzance’s emphatic statement in Hyde v Hyde that his decision was limited to the question of matrimonial relief, there was for many years a tendency to assume that all polygamous marriages were wholly unrecognised by English law. However, since 1939 it has become clear that they are recognised for many purposes.”
The 1971 Report then considered “some situations in which polygamous marriages, valid by the law of the place of celebration and by the personal law of the parties, have been or may be considered by the courts”. These included legitimacy, succession and certain statutory provisions including section 17 of the Married Women’s Property Act 1882.
94. Section 15 of the 1971 Report addressed social security legislation. The Law Commission commented, at , that any “change in the social security position of parties to polygamous marriages will require further legislation”. This was because of the effect of a number of decisions by Commissioners under the National Insurance Acts, including R(G) 18/52, in which, as referred to above, it had been held “that the polygamously married wife of a contributor was not entitled in right of his contributions to benefits under” the National Insurance Acts.
“The reason given was that "the question whether the words 'marriage', 'husband', 'wife' and 'widow', when used in an Act of Parliament or statutory instrument are intended to include polygamous marriages and the parties thereto must be decided in the light of the language of the Act or instrument in question taken as a whole, and of its manifest scope and purpose”, and that for National Insurance purposes the claimant had never been the wife of the contributor. It was thought that it obviously could not have been the intention to allow several wives of one contributor each to claim benefits under the Acts.”
95. The 1968 WP, at , noted that, in response to the “obviously unjust” effect of the Commissioners’ decisions, the law had been changed (as referred to in paragraph 79 above). The 1971 Report summarised these changes, at :
“This interpretation was thought to cause injustice in the case of the one and only wife of a man who was compelled to pay contributions because of his employment in this country. Parliament went some way to meet this injustice by enacting section 3 of the Family Allowances and National Insurance Act 1956. This is now replaced by the National Insurance Act 1965, section 113(1), the National Insurance (Industrial Injuries) Act 1965, section 86(5) and the Family Allowances Act 1965, section 17(9) …”
96. The Law Commission questioned whether “the sections go far enough”, at  of the 1968 WP and at  of the 1971 Report, because of the requirement that the marriage be “at all times” monogamous. As a result, they “do not cover cases where the marriage was once actually polygamous, but is so no longer, for example, because the first wife died or was divorced before the parties came to England. Nor do they cover cases where the marriage is in fact polygamous at the time when the social security benefits are sought although only one wife is in this country”.
97. For the purposes of the issue in the present case, it is relevant to note that the 1968 WP went on to say, at :
“We appreciate that this position may only arise so long as the parties remain domiciled in e.g. Pakistan. For as soon as they acquire an English domicil, the marriage would become monogamous in law as well as fact” (per Ali v Ali)”
This would support Ms Leventhal’s submission that, in the section dealing with social security legislation, the Law Commission was addressing valid and not void polygamous marriages. Whilst not explicit, I would add that there is nothing in either the 1968 WP or the 1971 Report which would suggest that the Law Commission was proposing anything other than a development of the then legislation to address the requirement that the marriage be “at all times” monogamous. Conversely, there is nothing which would suggest that the Law Commission was proposing any more significant development which might have changed the underlying purpose of the legislation as identified in the National Insurance Advisory Committee Report.
98. In its analysis, the Law Commission referred to Iman Din v National Assistance Board  2 QB 213. The issue in that case was whether the fact that the marriage between the husband and the wife had been polygamous prevented the National Assistance Board from recovering sums from the husband in respect of “assistance” which the Board had paid for the wife and the children. The marriage had been contracted in Pakistan in 1948 when both parties were domiciled there and the husband was still married to his first wife, who died in 1949. The second wife and their children came to the UK in 1961.
99. The judgment of the Divisional Court was given by Salmon LJ, as he then was, (with whom Lord Parker CJ and Widgery J agreed) in the course of which he said, at p.218 E - p.218 A:
“It would perhaps be as remarkable as it would be unfortunate if a man coming from a country where he is lawfully married to a woman and is lawfully father of her children may bring them here and leave them destitute with impunity …
When a question arises of recognising a foreign marriage or of construing the word ‘wife’ in a statute, everything depends upon the purpose for which the marriage is to be recognised and upon the objects of the statute. I ask myself first of all: is there any good reason why the appellant's wife and children should not be recognised as his wife and children for the purpose of the National Assistance Act, 1948? I can find no such reason, and every reason in common sense and justice why they should be so recognised.
Mr Abbas sought to derive some assistance from Hyde v Hyde. That case and the long stream of authority that flows from it as in my judgment of no help to this appellant. All that it lays down is that parties to a polygamous marriage by their personal law and the law of the country in which it is celebrated, cannot obtain matrimonial relief against each other in the courts of this country.”
He then went on to consider Baindail v Baindail and the Sinha Peerage Claim. His conclusions, at p. 220 E/F were as follows:
“So it is plain from the authorities to which I have referred that there are purposes for which a polygamous marriage will be recognised as a valid marriage in this country, and also that in some statutes the word "wife" may be construed as covering a polygamously married wife.
The only question before us is whether, for the purposes of the National Assistance Act, 1948, this court should recognise the polygamous marriage, and hold that the woman whom the appellant married polygamously in Lahore in 1948 is his wife for the purposes of the Act. I would unhesitatingly answer that question in the affirmative.”
100. Salmon LJ specifically rejected the argument that the court should apply the approach taken in respect of the National Insurance Acts (as referred to above). He did not consider, at p. 222 B/C, that those decisions “have any bearing upon whether or not she should be recognised as a wife for the purposes of the National Assistance Act, 1948” because, at p. 221 G/p. 222 A, they were based on the fact that “as the man paid only one lot of contributions, calculated on the basis of one wife at a time … It would clearly be wrong for (him) … to reap benefits in respect of perhaps three or four current wives”.
101. It is clear from Salmon LJ’s judgment that he was able to distinguish the approach taken in respect of the National Insurance Acts because of the different context he was addressing. It is also clear that his decision was in respect of a marriage which was valid under the rules of private international law.
102. The Law Commission recommended that the law should be changed. This was because, first, at , the “present law is too restrictive in denying all benefits unless there has never been more than one wife”. Taking the facts of Iman Din, the Law Commission observed, at  of the 1971 Report, that the wife in that case would not “have been entitled to any social security benefits payable to a wife or widow because her marriage was at one time actually polygamous” and then commented:
“We regard this as both unfortunate and anomalous because the second wife, having been admitted into this country as the wife of a permitted immigrant, should be treated just like any English wife if she was in fact her husband’s only wife throughout the period of their residence in England while the husband was paying contributions in England. It cannot, surely, be right to compel the husband to suffer deductions from his wages because of his employment in England, and then deny social security benefits to the woman who, throughout the period of those compulsory deductions, was his one and only wife, merely because at some earlier time before coming to England he had another wife.”
103. The Report noted the effect of Ali v Ali, which had decided that the acquisition by a husband of an English domicile converted a potentially polygamous marriage into a monogamous one. This would not assist the wife in Iman Din because her marriage had been actually polygamous. The social security legislation would also not assist because the marriage had not “at all times been monogamous”. Further, the Law Commission considered, at :
“There are, in addition, practical reasons why Ali v Ali should not apply for the purposes of social security. In claims for insurance benefit the detailed investigation of all the facts necessary to establish domicil should be avoided in possible.”
104. Secondly, in respect of marriages which were in fact polygamous when the benefits were sought, the Law Commission recognised, at , that “to deny social security benefits in such a case may involve hardship and injustice” but it had been unable to identify a solution “which we were sure would be both administratively workable and acceptable to public opinion”. The different social security schemes meant that, at , “no single test applying to them all is practicable”. As a result, it was “within the sphere of the Secretary of State for Social Services … to determine as a matter of policy the circumstances in which each particular benefit should be available in a polygamous situation”.
105. The Law Commission did, however, recommend, at , that the relevant provisions in the NIA 1965 (and the other similar provisions)
“should be replaced in each case by a provision that any person claiming a benefit in reliance on a marriage celebrated outside the United Kingdom under a law which permits polygamy should qualify for the benefit except where regulations otherwise provide.”
106. This recommendation, known as “regulating out”, was not adopted. Instead, the Secretary of State was given the power to make regulations specifying when a party to a polygamous marriage would be entitled to benefits. Section 12 of the National Insurance Act 1971 (“the NIA 1971”) provided:
“The Secretary of State may by regulations make provision for any purpose of the Insurance Act, the Industrial Injuries Act or the Family Allowances Act 1965 as to the circumstances in which a marriage celebrated under a law which permits polygamy is to be treated as having the same consequences as a marriage celebrated under a law which does not, and any such regulations may make different provision in relation to different enactments, purposes and circumstances.”
This led to the National Insurance, Industrial Injuries and Family Allowances (Polygamous Marriages) Regulations 1972 which were substantially, and in so far as relevant, in the same terms as the 1975 Regulations.
Legal Framework 1975-1995
107. The statutory provisions which replaced section 12 of the NIA 1971, namely section 162 of the SSA 1975 (as set out in paragraph 25 above) and section 12(2) of the Family Allowances Act 1965 (as substituted by paragraph 16 of Schedule 2 to the Social Security (Consequential Provisions) Act 1975), were not in the same terms. First, section 162(a) gave a power to make regulations dealing with voidable marriages. Secondly, section 162(b) made clear, what was plainly implicit in section 12 of the NIA 1971, namely that the power to make regulations extended to both potentially polygamous marriages (paragraph (i)) and actually polygamous marriages (paragraph (ii)). I set out the latter provisions again:
“Regulations may provide —
(b) as to the circumstances in which, for the purposes of this Act —
(i) a marriage celebrated under a law which permits polygamy, or
(ii) any marriage during the subsistence of which a party to it is at any time married to more than one person,
is to be treated as having, or not having, the consequences of a marriage celebrated under a law which does not permit polygamy …”
108. We were referred, through Helen Walker’s statement, to a number of largely internal Government documents including speaking notes. We were not addressed on the admissibility of these documents but, in any event, although they were relied on by Ms Leventhal, they did not provide much assistance as to the meaning of the 1975 Regulations or other statutory provisions. They provided no more than a broad overview with, for example, one document from April 1971 recording that, when seeking Cabinet Committee approval for the National Insurance Bill, the Minister stated that he was using “the opportunity of the Bill to take power to make regulations to admit to benefit persons who are excluded under the present rules relating to validity of marriage”.
109. A speaking note, which is reflected in what the Minister (Sir Keith Joseph) said in Parliament, referred to “what we have in mind for the regulations … is to cover those cases where a marriage was once actually polygamous but is so no more, for example, because the first wife died or was divorced before the parties came to this country”. This was written before the regulations had been drafted and which the note records “will not be easy regulations”, I assume, to draft. The example given was no more than an example and, in my view, cannot be taken as necessarily demonstrating the intended scope or purpose of the regulations, once drafted. However, there is certainly nothing which shows any, or any clear, intention to depart from the purpose of the existing legislation or the “mischief” as referred to by Underhill LJ (in paragraphs 249 and 250 below).
110. As referred to above (paragraph 92), the Law Commission’s recommendations in its 1970 Report on the Nullity of Marriage were implemented by the MP(PM)A 1972. These recommendations had not, however, included the provisions of section 4 (which amended section 1 of the NoMA 1971): this was “an additional provision … included in the Bill by its sponsors” (the 1982 WP at [2.10]). The 1982 Report commented, at [2.11], that it was not easy to glean the reasons for this (because of the absence of debate in Parliament) but that, “it appears from the report of the Committee stage of the Bill’s passage through the House of Lords that the clause was introduced because the rest of the law of nullity had recently been codified in the Nullity of Marriage Act 1971 and its inclusion was intended merely to embody in statutory form the observation from our  Report” (on polygamous marriages), at , that “if a person domiciled in England goes through a polygamous form of marriage abroad, that marriage will, under English law, be void, even if it was only potentially polygamous”.
111. Section 4 of the MP(PM)A 1972 amended section 1 of the NoMA 1971 by providing, additionally, that a marriage would be void:
“(d) in the case of a polygamous marriage entered into outside England and Wales, that either party was at the time of the marriage domiciled in England and Wales.
For the purposes of paragraph (d) of this subsection a marriage may be polygamous although at its inception neither party has any spouse additional to the other."”
These provisions became section 11 of the 1973 Act which, in so far as relevant, originally provided that a marriage would be void on the following grounds:
“(b) that at the time of the marriage either party was already lawfully married;
(d) in the case of a polygamous marriage entered into outside England and Wales, that either party was at the time domiciled in England and Wales.
For the purposes of paragraph (d) of this section a marriage may be polygamous although at its inception neither party has any spouse additional to the other.”
The effect of these provisions was thought to be, prior to Hussain v Hussain (see paragraph 113 below), that a marriage contracted by a person domiciled in England would be void if it was either actually or potentially polygamous.
112. The 1982 WP on polygamous marriages was followed by a joint report: the 1985 Report on Polygamous Marriages, Capacity to Contract a Polygamous Marriage and Related Issues (Law Com. No. 146 and Scot. Law Com. No. 96) (“the 1985 Report”). The 1985 Report’s recommendations are set out in Part V and included, in effect, that a marriage which was only potentially polygamous should not be invalid under either the law of England and Wales or Scotland.
113. The completion of the 1982 WP had coincided with the Court of Appeal’s decision in Hussain v Hussain which had interpreted the effect of section 11(d) of the 1973 Act contrary to the previously accepted view that a person domiciled in England and Wales could not contract a valid potentially polygamous marriage. In the judgment of the court, given by Ormrod LJ, it was said, at p. 32 F/G, that “a marriage can only be potentially polygamous if at least one of the spouses has the capacity to marry a second spouse”. In that case, the wife did not have capacity (under her personal law) of Pakistan to marry a second spouse and neither did the husband (under his) because, at p. 30 H, section 11(b) of the 1973 Act meant that, being domiciled in England, he was “incapable of contracting a valid marriage when he is already lawfully married”. Accordingly, the marriage was not within section 11(d) because it was not even potentially polygamous.
114. This decision was based, in part, on the conclusion, at p. 31 C/D, that the MP(PM)A 1972 had “radically altered the law relating to polygamous marriages”:
“Prior to this Act, the law was governed by the decision in Hyde v Hyde …, that a ‘marriage’ which did not create a monogamous union between a man and a woman was no marriage at all. Stated perhaps more precisely, the word ‘marriage’, where it appeared in matrimonial legislation, did not, as a matter of construction, include any kind of ceremony which did not create a monogamous relationship … .”
The MP(PM)A 1972 meant that, at p. 31 D/E, “the position is quite different”. This meant that, for jurisdiction purposes, the “only question is whether the marriage under consideration is valid by English law, which is a question of capacity”.
115. Ormrod LJ also pointed to the difference in wording between section 11(d) and section 47 of the MCA 1973, the former referring to “a polygamous marriage” and the latter to “a marriage entered into under a law which permits polygamy”. Following the revocation of the rule in Hyde v Hyde, “the question of the capacity of persons domiciled in England and Wales to enter into polygamous or potentially polygamous marriages had to be considered”, at p. 32 B/C. The “different formulations” supported the conclusion that section 11(d) was dealing with a polygamous marriage that at least one of the parties had the capacity to contract, because, at p. 32 A/B, if “the intention of Parliament (had) been to prevent persons domiciled in England and Wales from entering into marriages under … laws which ‘permit polygamy’, it would have been easy to say so in so many words”. In addition, Ormrod LJ said, at p. 32 G/H:
“On a broader view, it is difficult to conceive any reason why Parliament, in an increasingly pluralistic society, should have thought it necessary to prohibit persons, whose religious or cultural traditions accept polygamy, from marrying in their own manner abroad, simply because they are domiciled in England and Wales. On the other hand, it is obvious that Parliament, having decided to recognise polygamous marriages as marriages for the purposes of our matrimonial legislation, would think it right to preserve the principle of monogamy for persons domiciled here.”
116. This decision is relevant because it again confirms that the validity of a polygamous marriage depends on a party’s capacity to contract such a marriage, which is governed by English law for a person domiciled here, and because Ormrod LJ considered it “obvious that Parliament … would think it right to preserve the principle of monogamy for persons domiciled here” (my emphasis).
117. Despite the decision in Hussain v Hussain, the Law Commissions still recommended changes to the legislation. This was because, as summarised in the 1982 WP at [4.1], of “(c)riticisms of the present law relating to capacity to contract a marriage abroad in polygamous form” including “uncertainty as to what are at present the relevant choice of law rules” and, subject to Hussain v Hussain, “difficulties which have arisen in practice from the operation of the rule that a person domiciled in England and Wales cannot contract a valid marriage in polygamous form”.
118. The 1985 Report also identified defects in the then law which included, at [2.13], that Hussain v Hussain only applied to marriages after 31 July 1971. The example was given that the validity of a potentially polygamous marriage in Bangladesh before that date depended on whether the man was domiciled in England (it would be void) or in Bangladesh (it would be valid). The invalidity of the marriage created a “whole range of practical consequences” in respect of “matters such as succession, taxation, the provision of social security benefits, matrimonial relief, legitimacy, citizenship and immigration”.
119. Further, at [2.14], the effect of Hussain v Hussain was discriminatory against women because, pursuant to section 11 of the MCA 1973, the validity of a marriage which was potentially polygamous (by the law of the place of celebration) would depend on whether the woman was domiciled in England and Wales (the marriage would be invalid because the marriage was potentially polygamous) or the man (it would be valid because it was not potentially polygamous). It was also considered, at [2.15], preferable “that the rules governing capacity to enter all polygamous marriages should be placed beyond doubt by legislation”.
120. The 1985 Report recommended in respect of England and Wales:
“2.33 (a) We recommend that a marriage which is entered into by a man or woman domiciled in England and Wales should not (if English law is applicable thereto in accordance with English rules of private international law) be invalid by reason of the fact that the marriage is entered into under a law which permits polygamy, provided that neither party to the marriage is already married.”
And in respect of Scotland:
“2.34 (a) We recommend that a person domiciled in Scotland should not lack capacity to enter into a marriage by reason only that the marriage is entered into under a law which permits polygamy.”
121. In contrast, and importantly for the present case, the Report, at [4.1]-[4.8], did not propose any change to the “law governing the capacity of men and women domiciled in England and Wales to enter into an actually polygamous marriage abroad”. The effect was summarised, at [4.7]:
“… under the internal rules of English law, a man or a woman domiciled in England and Wales lacks capacity to enter into an actually polygamous marriage at common law, which governs marriages celebrated on or before 31 July 1971, and under section 11 of the Matrimonial Causes Act 1973, which applies to marriages which take place after that date.”
122. The recommendation in paragraph 2.33(a) was implemented, after some delay, through the Private International Law (Miscellaneous Provisions) Act 1995 (“the PIL(MP)A 1995”). This Act amended both section 11 and section 47 of the MCA 1973. It also, in section 5, dealt with the validity in English law of a potentially polygamous marriage, so as to codify the effect of Hussain v Hussain while also remedying its discriminatory consequence:
“5 Validity in English law of potentially polygamous marriages
(1) A marriage entered into outside England and Wales between parties neither of whom is already married is not void under the law of England and Wales on the ground that it is entered into under a law which permits polygamy and that either party is domiciled in England and Wales.
(2) This section does not affect the determination of the validity of a marriage by reference to the law of another country to the extent that it falls to be so determined in accordance with the rules of private international law.”
Section 11 of the 1973 Act was amended as follows:
“11. Grounds on which a marriage is void.
A marriage which takes place after the commencement of this Act shall be void on the following grounds only, that is to say –
(d) in the case of a polygamous marriage entered into outside England and Wales, that either party was at the time domiciled in England and Wales.
For the purposes of paragraph (d) of this section a marriage
may be is not polygamous if although at its inception neither party has any spouse additional to the other.”
As a result of this amendment, only actually polygamous marriages, and not potentially polygamous marriages, are void. Section 47 was amended as follows:
“47. Matrimonial relief and declarations of validity in respect of polygamous marriages.
A court in England and Wales shall not be precluded from granting matrimonial relief or making a declaration concerning the validity of a marriage by reason only that
the marriage in question was entered into under a law which permits polygamy either party to the marriage is, or has during the subsistence of the marriage been, married to more than one person.”
123. There are passages in the 1982 WP and the 1985 Report which support the case advanced on behalf of the SSWP. In the 1982 WP, at [4.25], reference is made to the change effected by 1975 Regulations so that “a polygamous marriage is treated for the purposes of the Social Security Act 1975 and the Child Benefit Act 1975 as having the same consequences as a monogamous marriage for any day throughout which it is in fact monogamous”. This is then followed, at [4.26], by a reference to the Department of Health and Social Security having indicated that “the rule whereby an English domiciliary lacked capacity to contract a marriage in polygamous form did not often give rise to problems, because in practice the Department operated on the basis of a very strong presumption of the validity of a marriage …”. The Department had, however, drawn attention to “some cases” in which the rule had caused difficulties and had “pointed out that the relevant system of adjudication below the level of Commissioner is not well suited to the investigation and determination of difficult questions of domicile”.
124. Of perhaps more significance is the specific reference, at [4.41] of the 1982 WP, to the 1975 Regulations as having provided “that a valid polygamous marriage shall be treated as a monogamous marriage for any day on which it is in fact monogamous” (my emphasis). This is picked up in the 1985 Report which deals, in Part III, with polygamous marriages “which our law regards as validly contracted”, at [3.1]. The 1985 Report then sets out, at [3.2], how, over “the last three to four decades … the attitude of English law to polygamous marriages has altered radically”. In that context, the Report mentioned, albeit very briefly at [3.4], the 1975 Regulations as being an example of when “recognition is accorded to a polygamous marriage provided that it is in fact monogamous”.
125. In addition, in contrast to the observations made about void marriages (referred to in paragraph 121 above), the 1985 Report made the following comment about potentially polygamous marriages, at [3.10]:
“3.10 Once the earlier recommendations in this report are implemented, both English law, by reason of the changes proposed to the rules for capacity to marry and of the developments in the law relating to potentially polygamous marriages over the last few decades, and Scots law, by reason of the recommendations in this report, will not discriminate between valid marriages contracted by spouses, neither of whom was already married, on the basis of the form of the marriage ceremony. It is, however, the case that some current legislation is drafted on the basis of the existence of such a distinction.”
This is helpful because it neatly summarises the legal position of potentially polygamous marriages including those contracted by a person domiciled in England and Wales. There would no longer be any difference in treatment between those marriages contracted by parties who were, and those who were not, domiciled in England and Wales purely because “of the form of the marriage ceremony”.
126. One of the consequential amendments made by the PIL(MP)A 1995 was to section 121 of the SSCBA 1992. The words in section 121(1)(b), dealing with potentially polygamous marriages (the same words as in section 162(b)(i) of the SSA 1975 Act), were removed and the amended subsection (b) provided simply as follows:
“(b) as to the circumstances in which, for the purposes of the enactments to which this section applies, a marriage during the subsistence of which a party to it is at any time married to more than one person is to be treated as having, or as not having, the same consequences as any other marriage.”
Section 147(5) of the SSCBA 1992 was similarly amended.
Social Security Authorities
127. I now turn to consider other decisions dealing with entitlement to social security benefits to which we were referred.
128. R(G) 1/94 (4 May 1993: CG/16/1992) provides, in my view, no assistance because the report is very brief and the case was decided solely on the basis that the relevant marriage was actually polygamous.
129. R(G) 1/93 (27 August 1993: CG/4/1991) concerned a claim to widow’s benefit made by one of the deceased’s wives. He had married his first wife (the claimant) in what was then East Pakistan in 1955. He married his second wife in what had by then become Bangladesh in 1981. Both wives had made claims and both had been disallowed on the basis that neither marriage was monogamous at the relevant time, namely the date of the deceased’s death. The first wife’s appeal to the tribunal had been dismissed and she appealed to a Social Security Commissioner.
130. Commissioner Johnson set aside the tribunal’s decision because, at , it had failed to consider the issue of the deceased’s domicile which, in his view, was the “crucial” issue in the case because the validity of the deceased’s second marriage depended on where he was domiciled at that date. He started his legal analysis by quoting from R(G) 1/70:
“’7. At paragraph 7 of R(G) 1/70 it was held (following R(G) 18/52) that:
‘… the word ‘marriage’ … and related words such as ‘husband’ and ‘widow’, denote a matrimonial relationship of a monogamous character and do not included polygamous relationships …’
It is therefore well settled and trite law that for social security purposes a widow is only entitled to widow’s benefit if, at the date of her husband’s death, in addition to his having the necessary contribution record, their marriage was in fact monogamous. That means that at that date there must be no other valid and subsisting marriage. Where a husband has entered into another marriage, or purported marriage, in order for his marriage to his first wife to be monogamous at the date of his death, it will have to be shown, among a number of possibilities which it is not necessary for me to go into, that the second wife was dead, the second marriage had been validly dissolved or, the issue in the instant case, was invalid at the date it was celebrated.”
The Commissioner determined that the deceased was domiciled in Bangladesh at the date of the second marriage. As a result, under English law, it was a valid and subsisting marriage. This was, at , “extremely unfortunate” for the first wife:
“Had I been able to find that (the deceased) had retained his English domicile of choice at the time of his marriage to (the second wife), then that marriage would be invalid in English law and (the first wife) would be the only lawful wife, and widow, for the purposes of the social security legislation.”
131. The next decision is that of CG/2611/03, an appeal on a question of law to the Social Security Commissioner from the Appeal Tribunal. This decision post-dates the amendments made by the PIL(MP)A 1995 including to section 11(d) of the MCA 1973 so that a potentially polygamous marriage by a person domiciled in England and Wales was no longer void. These amendments are not referred to in the decision perhaps because they did not directly arise in the case. However, the decision addressed the issue raised by the present appeal.
132. The case concerned a claim to widow’s benefit by the deceased’s second wife. The deceased had married three times in Bangladesh (or what became Bangladesh). Commissioner Howell referred to the SSCBA 1992 and the 1975 Regulations. He allowed the appeal because the tribunal had considered the issue of the deceased’s domicile at the date of his respective marriages to be irrelevant. The Commissioner disagreed because, in his view at , the second wife’s claim would fail if her marriage was void under English law, which it would be if the deceased was domiciled here at the date of their marriage. Alternatively, if the deceased only became domiciled in England after the date of his second marriage but before the date of his third marriage, the third marriage would be void “for all purposes” under English law, with the result that, at , the second wife would “be entitled to succeed as the only surviving wife who had a valid marriage still subsisting at her husband’s death”. His reasoning was expressed, at , as follows:
“A person seeking to claim widow’s benefit under (the SSCB Act 1992) has to be either the surviving member of a monogamous marriage recognised as valid under United Kingdom law or the surviving member of a valid marriage under a law which permits polygamy but in fact the only spouse of the deceased at the date of his death: section 121(1)(b), and regulation 2 of (the 1975 Regulations).”
The case was sent back for rehearing because the tribunal’s decision was internally inconsistent on the issues of domicile and whether there were two subsisting valid marriages at the date of the deceased’s death.
133. In a number of decisions prior to his judgment in the present appeal, the UTJ applied the same approach as in CG/2611/03, namely that a marriage would only come within the 1975 Regulations if it was valid under English law. It is not directly relevant for the purposes of the present case but, as referred to above, the earlier decisions including R(G) 18/52 and R(G) 1/70 excluded from entitlement to benefit the surviving widow of a potentially polygamous marriage even when it was valid under English law. These later decisions appear, however, to have departed from this probably because of the changes effected by the PIL(MP)A 1995. This appears in particular from SA v Secretary of State for Work and Pensions  UKUT 436 (AAC) in which the UTJ decided that a widow, from what would appear to have been a potentially polygamous marriage in Pakistan in 1975, was entitled to bereavement benefit because it was a valid marriage.
134. The other decisions to which we were referred, in which the UTJ applied what he described in the present case, at , as the “orthodox reading” were: SB v Secretary of State for Work and Pensions  UKUT 219 (AAC); and Secretary of State for Work and Pensions v N (BB)  UKUT 68 (AAC).
135. In the last of these decisions the UTJ stated, at , that regulation 2(1) of the 1975 Regulations:
“… does not have the effect of converting a void marriage into a valid one simply by virtue of the parties being in practice monogamously married immediately prior to one party’s death. Instead, it means that a valid polygamous marriage can be treated as a ‘monogamous marriage for any day … throughout which the polygamous marriage is in fact monogamous’” (emphasis in original).
He referred to CG/2611/2003 and went on to say, at , that a void marriage “could not be rescued by regulation 2. In effect, it never existed as a valid marriage for the purposes of social security law (see R(G) 3/59)”.
136. We were not directly referred to the Court of Appeal’s decision in Bibi v Chief Adjudication Officer  1 FLR 375 probably because it did not directly address the issues raised by this appeal. However, it is clear from Ward LJ’s judgment that the case had proceeded (in appeals to the social security appeal tribunal and, then, to the commissioner) on the basis that the claimant’s entitlement depended on her marriage (in 1966 in what became Bangladesh) being valid under English law and on whether her deceased husband’s second marriage (in 1969 also in Bangladesh) was valid or void because of his domicile at that date. This approach was not questioned in the Court of Appeal. In the course of his judgment, at pp. 378/379 Ward LJ said:
“The appeal tribunal made various findings of fact which were shortly challenged by Mr De Mello on behalf of the appellant. The thrust of his submissions would have led to the deceased having established his domicile in this country in 1961, but when it was recognised that, if that were so, he would have lacked the capacity as a domiciled Englishman to contract this marriage at all, the appeal against the findings of fact was abandoned.”
137. Finally, in respect of the meaning of the 1975 Regulations, we were not referred to Cheshire, North & Fawcett Private International Law (14th Ed) which refers, briefly at pp. 934/935, to polygamous marriages and social security legislation:
“Statutory recognition of polygamy is also provided by social security legislation. Regulations made under or preserved by the Social Security Contribution and Benefits Act 1992 now govern the present position in relation to benefits falling within these Acts, eg widow’s benefit, maternity benefit and child benefit. They allow a valid polygamous marriage to be treated as a monogamous marriage if it has either always been monogamous or for any day throughout which it was, in fact, monogamous” (my emphasis).
A number of National Insurance Decisions are listed (p. 934, n. 500) as well as R v Department of Health ex p Misra  1 FLR 128 (“Misra”).
138. In Misra, the deceased had been a doctor. He had married his first wife in 1949 and his second in 1952. Both his wives applied for a widow’s pension under the State scheme (section 24 of the SSA 1975) and for the widow’s entitlement under the NHS superannuation scheme. Latham J (as he then was) summarised the position, at pp. 129/130, as follows:
“Under the State scheme, neither would be entitled to a pension if both marriages were valid under Indian law. Under the NHS scheme, if both marriages were lawful, the pension entitlement would, by extra statutory concession, be divided in half between each of them; if only one marriage was lawful, then the whole entitlement would go to whoever had been party to that lawful marriage.”
It was, at p. 132, “common ground that the operation of general principles of law means that a woman can only be considered to have been 'widowed' if she was, before the death of her husband, a wife in a monogamous marriage”. After quoting regulation 2 of the 1975 Regulations, Latham J said, at p. 132:
“It follows that the SSAT correctly identified the issue to be determined as whether or not a ceremony of marriage had taken place on 26 December 1952 which was a lawful polygamous marriage at the time.”
139. I now turn to the legal framework on the issue of discrimination.
140. Article 14 of the ECHR provides:
“The enjoyment of the rights and freedoms set forth in this Convention shall be secured without discrimination on any ground such as sex, race, colour, language, religion, political or other opinion, national or social origin, association with a national minority, property, birth or other status.”
As set out by Baroness Hale in McLaughlin, at , the issue of whether there has been a breach of article 14 can be determined by addressing the following four questions:
“As is now well known, this raises four questions, although these are not rigidly compartmentalised:
(1) Do the circumstances “fall within the ambit” of one or more of the Convention rights?
(2) Has there been a difference of treatment between two persons who are in an analogous situation?
(3) Is that difference of treatment on the ground of one of the characteristics listed or “other status”?
(4) Is there an objective justification for that difference in treatment?”
141. On the second question, namely whether the persons being treated differently are in an analogous situation, Baroness Hale started with the following, at :
“However, as Lord Nicholls explained in R (Carson) v Secretary of State for Work and Pensions  1 AC 173, para 3:
“the essential question for the court is whether the alleged discrimination, that is, the difference in treatment of which complaint is made, can withstand scrutiny. Sometimes the answer to this question will be plain. There may be such an obvious, relevant difference between the claimant and those with whom he seeks to compare himself that their situations cannot be regarded as analogous. Sometimes, where the position is not so clear, a different approach is called for. Then the court’s scrutiny may best be directed at considering whether the differentiation has a legitimate aim and whether the means chosen to achieve the aim is appropriate and not disproportionate in its adverse impact.”
As was pointed out in AL (Serbia) v Secretary of State for the Home Department  1WLR 1434, there are few Strasbourg cases which have been decided on the basis that the situations are not analogous, rather than on the basis that the difference was justifiable. Often the two cannot be disentangled.”
142. Baroness Hale next referred, at , to a number of decisions in which the European Court of Human Rights had stated that “marriage conferred a special status” and that unmarried cohabitees were not in an analogous situation to a married couple: Lindsay v United Kingdom (1986) 9 EHRR CD 555; Shackell v United Kingdom (Dec) (App no 45851/99), 27 April 2000; and Burden v United Kingdom (2008) 47 EHRR 38.
143. The last of these concerned a claim by unmarried sisters, who had lived together all their lives, that the effect of inheritance tax discriminated against them. The Grand Chamber, at , considered “that marriage confers a special status” and endorsed the view set out in Shackell that unmarried and married cohabiting couples are not in an analogous situation. At , the Court made the following observations:
“As with marriage, the Grand Chamber considers that the legal consequences of civil partnership under the 2004 Act, which couples expressly and deliberately decide to incur, set these types of relationship apart from other forms of cohabitation. Rather than the length or the supportive nature of the relationship, what is determinative is the existence of a public undertaking, carrying with it a body of rights and obligations of a contractual nature. Just as there can be no analogy between married and Civil Partnership Act couples, on one hand, and heterosexual or homosexual couples who choose to live together but not to become husband and wife or civil partners, on the other hand, the absence of such a legally binding agreement between the applicants renders their relationship of cohabitation, despite its long duration, fundamentally different to that of a married or civil partnership couple.”
As a result, at , the Court concluded that “the applicants, as cohabiting sisters, cannot be compared for the purposes of Art. 14 with a married or Civil Partnership Act couple”.
144. Shackell dealt with a claim not only to widow’s payment but also to widowed mother’s allowance. What had been said by the ECtHR in respect of the latter was held in McLaughlin, at , to be “wrong or should not be followed”. Lord Mance (with whom Baroness Hale, Lord Kerr and Lady Black agreed) explained, at , that the reasoning in Shackell failed to address “the clear purpose of (widowed mother’s) allowance, namely to continue to cater, however broadly, for the interests of any relevant child”. I return to this decision below.
145. We were referred to Yiğit v Turkey (2011) 53 EHRR 25, also a Grand Chamber decision. In that case, the applicant had been refused a survivor’s pension because she and her deceased partner had only gone through a religious ceremony of marriage and not a civil ceremony. Under Turkish law, the former was not recognised as valid and did not create the status of husband and wife.
146. The Court, at , set out the general approach to marriage:
“With regard to art.12 of the Convention, the Court has already ruled that marriage is widely accepted as conferring a particular status and particular rights on those who enter it. The protection of marriage constitutes, in principle, an important and legitimate reason which may justify a difference in treatment between married and unmarried couples. Marriage is characterised by a corpus of rights and obligations that differentiate it markedly from the situation of a man and woman who cohabit. Thus, states have a certain margin of appreciation to treat differently married and unmarried couples, particularly in matters falling within the realm of social and fiscal policy such as taxation, pensions and social security.”
147. The Court did not directly address the question of whether the applicant was in an analogous situation to someone who had gone through a civil ceremony of marriage but appears to have treated this as established. The Court decided, at , that the applicant had a “status” which brought her within the scope of article 14. This was summarised as being because “the difference in treatment … was based solely on the non-civil nature of her marriage to her partner”.
148. The Court next addressed whether the difference in treatment was justified. The Court decided, at , that the difference pursued a legitimate aim. This was based on “the importance of the principle of secularism in Turkey” and that, in requiring “monogamous civil marriage as a prerequisite for any religious marriage, Turkey aimed to put an end to a marriage tradition which places women at a disadvantage”. This latter observation was a reference to the fact that a purely religious marriage was polygamous. As for proportionality, the Court decided for a number of reasons that the difference in treatment was proportionate to the legitimate aim pursued. The Court noted, at , that the rules “governing civil marriage are clear and accessible and the arrangements for contracting a civil marriage are straightforward and do not place an excessive burden on the persons concerned”.
149. We were referred to a number of domestic authorities. These included R (Carson) v Secretary of State for Work and Pensions  1 AC 173 in which Lord Nicholls, at , referred to the “essential question” (quoted by Baroness Hale in McLaughlin as set out above).
150. In R (RJM) v Secretary of State for Work and Pensions  AC 311, Lord Walker addressed, at , the issue of “personal characteristics”, or status, for the purposes of article 14. In his view, they “are more like a series of concentric circles” and the “more peripheral or debateable any suggested personal characteristic is, the less likely it is to come within the most sensitive area where discrimination is particularly difficult to justify”.
151. R (DA) v Secretary of State for Work and Pensions  1 WLR 3289 confirms that, when the court is determining whether a difference in treatment in respect of entitlement to welfare benefits is justified, the test to be applied is that set out by Lord Wilson, at :
“… the weight of authority in our court mandates inquiry into the justification of the adverse effects of rules of entitlement to welfare benefits by reference to whether they are manifestly without foundation.”
See also Lord Carnwath, at -. This issue has been revisited by the Supreme Court in R (on the application of SC, CB and 8 children) (Appellants) v Secretary of State for Work and Pensions and others (Respondents)  UKSC 26,  3 WLR 428, (“R (SC)”), in which Lord Reed (with whom the rest of the court agreed) refers, at , to the European court having “generally adopted a nuanced approach” to the issue of justification and, at , to the need “to avoid a mechanical approach to these matters, based simply on the categorisation of the ground of the difference in treatment”. The point made, at , is that proportionality and, consequently, “the intensity of the court’s scrutiny can be influenced by a wide range of factors”.
152. In McLaughlin, the Supreme Court dealt only with WPA because the claim in respect of BP had been dismissed at first instance and had not been appealed. Treacy J (as he then was) rejected the challenge to the lawfulness of BP: McLaughlin’s (Siobhan) Application  NIQB 11. He explained, at , under the heading “Bereavement Payment - Comparability”:
“66. Through marriage (or civil partnership) a couple regulates their relationship with each other and with the state through their public contract. The couple puts the state ‘on notice’ of their relationship. A cohabiting couple make no such public contract. This in itself is usually sufficient to make the two relationships sufficiently different in a material particular to lawfully treat the relationships differently in certain circumstances. By the act of marriage the couple ‘opt in’ to this different treatment - the treatment arises not by virtue of the quality of the relationship or the length of the relationship, but because the couple have made the contract and made the state aware of their changed circumstances.”
This was picked up by Baroness Hale, at , when she commented that Treacy J had been able to distinguish between Mrs McLaughlin’s claims for BP and WPA:
“In the case of the former, he held that the lack of a public contract between Ms McLaughlin and Mr Adams meant that her situation was not comparable with that of a widow and her claim must fail … That decision has not been appealed. In the case of the latter, he held that the relevant ‘facet of the relationship’ was not their public commitment but the co-raising of children. For that purpose marriage and cohabitation were analogous”.
Baroness Hale, at , endorsed Treacy J’s analysis, saying that, in her view, it was “correct”.
153. Baroness Hale returned to this, at , when addressing whether the difference in treatment for the purposes of WPA had a legitimate aim:
“ The legitimate aim put forward by the respondent is to promote the institutions of marriage and civil partnership by conferring eligibility to claim only on the spouse or civil partner of the person who made the contributions. There is no doubt that the promotion of marriage, and now civil partnership, is a legitimate aim: this was the reason why the denial of widow’s benefits to an unmarried partner was held justified in Shackell v United Kingdom CE:ECHR:2000: 0427DEC004585199; and why the preference given to civil over religious marriage was held justified in Yiğit v Turkey 53 EHRR 25.”
Lord Mance also commented, at , that a “policy in favour of marriage or civil partnership may constitute justification for differential treatment, when children are not involved”.
154. The distinct status of marriage has been addressed in a number of domestic authorities including, in particular, Akhter v Khan (Attorney General and others intervening)  2 WLR 1183. That case considered the legal effect of an Islamic marriage ceremony performed in London. It was decided, at , that what had occurred was a “non-qualifying ceremony” which was of no legal effect and “did not create a void marriage”.
155. The judgment of the court (Sir Terence Etherton MR, King and Moylan LJJ), referred, at , to the importance attached to the status of marriage: a “person’s marital status is important for them and for the state” because of the “specific rights and obligations” derived from that status “and not any other form of relationship”:
“It is, therefore, of considerable importance that when parties decide to marry in England and Wales that they, and the state, know whether what they have done creates a marriage which is recognised as legally valid.”
Although this observation was limited to England and Wales, it clearly also applies to marriages contracted elsewhere. The judgment also noted, at , that “(c)ertainty as to the existence of a marriage is in the interests of the parties to a ceremony and of the state” and, at :
“As referred to in para 9 above, marriage creates an important status, a status “of very great consequence”, per Lord Merrivale P in Kelly (orse Hyams) v Kelly (1932) 49 TLR 99, 101. Its importance as a matter of law derives from the significant legal rights and obligations it creates. It engages both the private interests of the parties to the marriage and the interests of the state. It is clearly in the private interests of the parties that they can prove that they are legally married and that they are, therefore, entitled to the rights consequent on their being married. It is also in the interests of the state that the creation of the status is both clearly defined and protected.”
156. On the issue of reading down the 1975 Regulations, we were referred to Ghaidan v Godin-Mendoza  2 AC 557 including what Lord Nicholls said, at -, about the extent of and limitations on the power to read down legislation pursuant to section 3 of the HRA 1998. The power is very broad but it does not extend, at , to adopting “a meaning inconsistent with a fundamental feature of legislation”. Lord Rodger made similar observations, at -, when he contrasted reading legislation in a manner which is “consistent with the scheme of the legislation” or which goes “with the grain of the legislation” and reading it in a manner which is “inconsistent with the scheme of the legislation or with its essential principles”. Accordingly, the proposed reading of the legislation must leave “intact” and not contradict the “essential principles and scope of the legislation”.
157. We were also referred to AR v Secretary of State for Work and Pensions  UKUT 165 (AAC). Much of what was said in that case would apply equally to the circumstances of the present case. The Upper Tribunal (a three-judge panel of the Administrative Appeals Chamber including Farbey J) addressed whether the term “spouse” in section 39A of the SSCBA 1992 could be read, pursuant to section 3 of the HRA 1998, as applying to the claimant to avoid an admitted breach of her Convention rights (in respect of WPA). The claimant and her deceased partner, at , had “entered into a religious marriage ceremony (Nikah) in accordance with Islamic principles” in England. It was argued on her behalf, at [9(a)], that the word “spouse” in section 39A should be construed as including someone in her position, “namely a person living with her partner having participated in a religious marriage ceremony according to the rites of that religion and subjectively believing herself (on objectively reasonable grounds) to be thereby married”. It was accepted by the claimant’s counsel, at , that, following McLaughlin, this would mean that the word “spouse” would not have the same meaning in sections 36 and 39A.
158. The argument was rejected, at , because “the grain of section 39A is that benefits should only be paid to a spouse married under English law”. The UT, at , identified “three main reasons for understanding the grain in this way: the genesis of these provisions; the legal and policy considerations relating to marriage; and the legislator’s intention as to conditions applicable to receipt of the benefit”.
159. As to the first, at , benefit provision for widows had commenced in 1925 and the UT had “not been taken to any material suggesting that, when the legislation referred to a ‘widow’, it meant anyone other than a woman who had been in a lawful marriage terminated by the death of her husband”. The UT, at , had also “been directed to nothing in the legislative scheme which would persuade us that, by adopting the term ‘spouse’ in the 1999 Act, Parliament intended to grant bereavement benefit to those not validly married as opposed to intending to remedy historic discrimination between men and women”.
160. As to the second, the UT referred to Akhter v Khan, noting that it “emphasises the state’s interest in knowing who is, and who is not, married” and concluded, at , that: “In our view, legal policy in relation to the value of marriage complying with legislative formalities is reflected in the draftsman’s decision to use the word “spouse” in section 39A to provide one of the gateway conditions to receipt of the benefit. It forms part of the grain of the legislation”.
161. As to the third, the UT referred, at , to provisions in the legislation “as to when the benefit will not be, or will cease to be, payable”. Section 36(2) and section 39A(5):
“each refer to a person who is (ex hypothesi) married to the deceased but at the time of the deceased’s death is ‘living together as a married couple’ with someone else. There is a clear distinction made by the legislator between the formal status of being married and the position of ‘living together as a married couple’. Mr Amos’s proposed reading of ‘spouse’ in section 39A is in our view incompatible with this distinction”.
This was followed, at , by reference to section 39A(4):
“which provides that ‘the surviving spouse shall not be entitled to the allowance for any period after she or he remarries or forms a civil partnership’. Entitlement is lost when these formal steps are taken. It is far from obvious that Parliament would have intended different levels of formality as regards marital status to have applied to accessing the benefit under section 39A(1) and to losing it under section 39A(4).”
162. Ms Leventhal accepted that the consequences for NA in the circumstances of this case make it a “hard case”. However, she submitted that this does not justify reaching a determination other than that which, in her submission, is required as a matter of law.
163. In respect of the scope of the 1975 Regulations, Ms Leventhal’s simple submission was that they only apply to a marriage which is valid under English law. They do not apply to a marriage which is void and, if required, words to that effect are a necessary implication. She relied on what Baroness Hale said in R (Black) v Secretary of State for Justice  AC 215, at :
“(3) The goal of all statutory interpretation is to discover the intention of the legislation.
(4) That intention is to be gathered from the words used by Parliament, considered in the light of their context and their purpose. In this context, it is clear that Lord Hobhouse of Woodborough's dictum in R (Morgan Grenfell & Co Ltd) v Special Comr of Income Tax  1 AC 563, 616, para 45, that “A necessary implication is one which necessarily follows from the express provisions of the statute construed in their context” must be modified to include the purpose, as well as the context, of the legislation.”
164. In her submission, the starting point must be that the words “spouse” and “marriage”, when used in the SSA 1975 Act, the SSCBA 1992 and the 1975 Regulations, mean a party to a marriage which is valid under English law and a marriage which is similarly valid. Accordingly, a person is only a “spouse” if they are a party to a marriage which is legally valid; a party to a void marriage is not a spouse. Equally, a void marriage is not a “marriage”.
165. Ms Leventhal submitted that there is nothing in the text or structure of the 1975 Regulations to suggest that these terms are intended to have any broader or different meaning and, in particular, to suggest that they include a party to, or, a void marriage. Indeed, she submitted that it is clear from regulation 2(2)(a) that they only apply to a spouse because, otherwise, they could not have “a spouse additional to the other”. In order to be a “spouse” they must, she submitted, have been party to a marriage valid under English law.
166. Looking at the legislative history, as referred to above, she submitted that the purpose of the 1975 Regulations can be seen to have been limited to remedying the effect of the Commissioner decisions in the 1950s and 1960s, namely that the National Insurance legislation did not apply to a party to a valid polygamous marriage, including one that was only potentially polygamous. They were not intended to change the law as to the effect of a polygamous marriage contracted by a party who was domiciled in England and were not considering or addressing a polygamous marriage which was, as a result, void.
167. Ms Leventhal pointed to the express power given by section 162(a) of the SSA 1975 for regulations to provide that “a voidable marriage” be treated “as if it had been a valid marriage”, but no similar provision in respect of void marriages. She also pointed to section 16 of the MCA 1973 which expressly provides that a decree of nullity (granted after 31 July 1971) in respect of a voidable marriage is prospective only in its effect and “the marriage shall, notwithstanding the decree, be treated as if it had existed up to that time”. There is no similar provision in respect of a void marriage because it is of no effect at all.
168. Ms Leventhal also relied on the consistent interpretation of the 1975 Regulations in the decisions referred to above and on the fact that most of these were by specialist tribunals.
169. It was, of course, accepted by Ms Leventhal that, consequent on the Supreme Court’s decision in McLaughlin, the legislation in respect of WPA was incompatible with the ECHR. It was also accepted that the claims to both BP and WPA fall within the ambit of A1P1 and that article 14 is, therefore, engaged. Otherwise, she challenged on a number of grounds the UTJ’s decision that the legislation was discriminatory and that the 1975 Regulations could be read down so as to include NA.
170. First, she submitted that, contrary to the UTJ’s decision at , NA was not in an analogous situation to that of a “lawful” widow and that the UTJ had been wrong to reject the binary nature of the distinction between a void and a valid marriage by reference to there being a “spectrum of potentially analogous situations” and a “kaleidoscope of other types of quasi-matrimonial relationship”. The fact that the marriage in this case was “an overseas religious marriage” which was valid under the law of Pakistan did not make it analogous to a marriage valid under English law. Ms Leventhal submitted that there is an “obvious and relevant difference” between parties to a marriage which is valid under English law and parties to a void marriage. There is a critical difference in the status conferred by the law when the requirements of a valid marriage are met.
171. Ms Leventhal relied on the decisions of the ECtHR and domestically, referred to above, as establishing that marriage has a special status. She also relied on the fact that the claim in respect of BP had been dismissed by the High Court in McLaughlin, a decision which had not been appealed and which had been endorsed by Baroness Hale, at . In her submission, NA was in an analogous position to the surviving partner of an unmarried cohabiting couple.
172. Secondly, Ms Leventhal accepted that, if NA was in an analogous situation to a lawful widow, she would have an “other status” for the purposes of article 14.
173. Thirdly, Ms Leventhal submitted that, when the court is considering, as here, a measure of general social or economic policy, the question is whether the disputed measure is manifestly without reasonable foundation, as established by R (DA) v SSWP. The UTJ had rightly, at , referred to this test but had failed properly to apply it. She also relied on the additional factor that the court should be “very slow to substitute its view for that of the executive, especially” when the asserted discrimination “is not one of the express, or primary grounds”: Lord Neuberger, at , in R (RJM) v SSWP referring to R (Carson) v SSWP.
174. On the issue of justification, Ms Leventhal relied on the same elements as had been advanced below (see paragraph 44 above) and submitted that the UTJ’s approach to them was flawed. Her overarching submission was that it is plainly not without reasonable foundation to treat the surviving partner of a void polygamous marriage in the same way as others who are not lawfully married rather than as the surviving spouse of a valid marriage. The UTJ had not applied this test but had substituted his own view of the strength of each aspect of the asserted justification for that of the SSWP. The six points advanced by Ms Leventhal were as follows.
175. (i) The justifiable primacy given to legal marriage is, she submitted, not diminished by the matters referred to by the UTJ. The fact that the marriage might be valid in Pakistan does not detract from the importance ascribed under English law to the distinction between a valid and a void marriage. The “privileges that UK law affords marriages” cannot be separated from “the circumstances in which the law recognises marriage (and divorce)”.
176. (ii) Ms Leventhal relied on what she submitted is the UK’s “strong and longstanding public policy against polygamous marriages”. The fact that for some, limited purposes, the law recognises valid polygamous marriages, does not undermine this as part of the justification for differentiating between valid and void marriages which are polygamous.
177. (iii) There is a need for a bright line and, contrary to the UTJ’s approach, the distinction between a valid and a void marriage does provide a bright line. (iv) Ms Leventhal contended that, in respect of the Beveridge contributory principle, the fact that the right to rely on a deceased partner’s national insurance record had been extended to widowers and civil partners provided no support for extending the right to the surviving partner of a marriage which was void. (v) Ms Leventhal relied on the SSWP’s evidence that there would be additional administrative difficulties; and (vi) she submitted that Yiğit v Turkey provided support for the legislative scheme being within the state’s margin of appreciation.
178. Ms Leventhal also challenged the UTJ’s approach to reading down the 1975 Regulations. She submitted that it goes against the grain of the substantive legislation. She also submitted that, contrary to the UTJ’s approach at , the meaning of “spouse” is central to the interpretation of the 1975 Regulations because the question of whether a marriage is “in fact monogamous” turns on whether, she emphasised, the person has a “spouse additional to the other”.
179. I can summarise Ms Rooney’s submissions more briefly in part because she seeks to uphold the UTJ’s decision essentially for the reasons he gave. Ms Rooney submitted that the UTJ’s decision is worthy of particular respect; he is a judge with extensive expertise in the field of social security sitting in a specialist tribunal.
180. Ms Rooney relied on the following facts. NA’s marriage was “inadvertently” polygamous because, at the date of their marriage, she and Mr A had mistakenly believed that he had divorced his first wife. From the date of Mr A’s divorce from his first wife, which preceded NA’s coming to the UK, their marriage had been monogamous. NA was granted entry clearance as a spouse. Mr A had paid national insurance contributions which were deducted from his salary. At the date of Mr A’s death, NA was the only claimant for BP and WPA as she was the sole survivor of a marriage to Mr A.
181. Ms Rooney submitted that, contrary to the SSWP’s submissions, the effect of the UTJ’s decision is not that a void marriage is transformed into a lawful one nor that the distinction between a valid and an invalid marriage would be undermined. In her submission, the 1975 Regulations simply contain deeming provisions which provide when a polygamous marriage is to be treated as having the same consequences as a monogamous one. This is not a decision which undermines the institution of marriage or which would lead to the promotion or endorsement of polygamy. It is, she submitted, about a discrete piece of legislation which contains a deeming provision for very specific purposes.
182. The 1975 Regulations do not include an express validity requirement and, Ms Rooney submitted, there is no justification by way of necessary implication or otherwise for importing into them a requirement that a marriage will only be “in fact monogamous” when it is valid.
183. Ms Rooney submitted that, as a matter of ordinary construction, the 1975 Regulations apply to NA because the natural reading of the words “in fact monogamous” is that they apply when, as a matter of fact (rather than law), there is no other spouse. In her submission, this is the effect of the definition in regulation 2(2)(a), which stipulates that “a polygamous marriage is referred to as being in fact monogamous when neither party to it has any spouse additional to the other”.
184. The key element in regulation 2, on which Ms Rooney focused, was the expression “in fact monogamous” which, she submitted, does not require the court to interpret the meaning of “spouse” or “marriage” as submitted by the SSWP. Ms Rooney distinguished AR v SSWP because that case was not concerned with a polygamous marriage and did not, therefore, have to consider the 1975 Regulations. Further, she submitted that section 162(b) of the SSA 1975, under which the 1975 Regulations were made, referred in sub-paragraph (ii) to “any marriage during the subsistence of which a party to it is at any time married to another person” (emphasis added). There was no attempt to limit the scope of any regulations made under those provisions to valid marriages when it must have been foreseeable that marriages within either section 162(b)(i) or, in particular, section 162(b)(ii), would often be invalid. The potential scope of the regulations permitted by section 162 was, therefore, very broad.
185. Looking at the context for the 1975 Regulations, Ms Rooney submitted that this strongly suggested that their purpose, and the purpose of the enabling legislation, was to remedy the potential injustices in entitlement to social security benefits caused by “the binary restrictive rules that governed the nature, capacity and validity of marriage”. She referred to parts of the 1968 WP and the 1971 Report including the conclusion in the former, at , that it was “obviously unjust to deny such benefits to the one and only wife of a man who was compelled to pay contributions because of his employment in this country, simply because his marriage was potentially polygamous” and the view expressed in the latter, at , that the law was “too restrictive in denying all benefits unless there has never been more than one wife”.
186. Ms Rooney submitted that the UTJ had been right to determine that NA had been discriminated against in breach of article 14. NA, as the sole surviving widow of a religious marriage, was in an analogous position to that of a “lawful” widow under a marriage recognised by the law of England and Wales. Ms Rooney also emphasised that this decision was based on a detailed consideration of NA’s position by reference to the “highly specific facts of this case and the particular benefits in question”. The UTJ had rightly rejected the submission on behalf of the SSWP that there was a binary distinction between the widows of a lawful marriage and a void one. As Lord Walker had said in R (Carson) v SSWP, at , “Some analogies are close, others are more distant”. In this case, as the UTJ noted, NA and Mr A had entered into a public contract which was valid in Pakistan and which made their position markedly different from that of a cohabiting couple.
187. As for the issue of justification, Ms Rooney submitted that the UTJ had expressly applied the test of manifestly without reasonable foundation and had been right to conclude that the discriminatory effect of the legislation was unjustified. She accepted that there may well be “good reasons for a public policy against polygamy”. However, she submitted, the UK has not in fact adopted such a policy consistently. She pointed to section 2 of the Immigration Act 1988, which provides that “no wife has the right of abode … on the basis of a polygamous marriage … if there is another woman living who is the wife or widow of the husband” and to the fact that some benefits are paid to the spouses of polygamous marriages. The latter are addressed in the House of Commons Briefing Paper: Polygamy, 20 November 2018.
188. Ms Rooney also submitted that the UTJ had recognised the potential validity of a bright line rule between lawful marriages and other relationships for social security purposes but had rightly concluded that the difficulties in establishing domicile made the line in the present circumstances “distinctly dim”.
189. The UTJ had also been right to read down the 1975 Regulations in the manner which he did. This “gave effect to the ordinary meaning of the words”, “in fact monogamous”, and “did no violence to the relevant provisions”. Accordingly, Ms Rooney submitted, it was not only permissible but mandatory relying on Ghaidan v Godin-Mendoza, at -.
190. I propose to address, in turn, each of the issues set out in paragraph 12 above.
191. The first issue is:
(i) Do the 1975 Regulations, with sections 36 and 39A of the SSCBA 1992, only apply if the relevant marriage is valid under English law and, specifically, not void under section 11 of the MCA 1973 and, accordingly, do they not apply to NA.
This depends on the proper construction of the legislation. Does the legislation apply only to polygamous marriages which are valid under English law, as submitted by Ms Leventhal, or does it include a party to a religious marriage valid in the place of celebration but void under English law, as submitted by Ms Rooney?
192. I make clear that, at the conclusion of the hearing, I was inclined to accept Ms Rooney’s submission that the 1975 Regulations were applicable to a void marriage such as NA’s. This was significantly because it appeared that, if they did not apply to void marriages, they would be of limited, if any, effect.
193. Having spent longer than I should have done since the hearing analysing how a void marriage could come within the legislative scheme, I have come to the clear conclusion that the 1975 Regulations only make sense if they do not apply to marriages which are void under English law. I set out my reasons for this below but I first set out why this would not, as it previously appeared to me, deprive them of any substantive effect.
194. The primary social security legislation (initially the NIA 1946) was interpreted as not applying to a party to a polygamous marriage, including only a potentially polygamous marriage, even when it was otherwise valid under the general rules of English private international law. This was the position following the decision in R(G) 18/52 dealing with section 17 of the NIA 1946. It was repeated in respect of section 26 of the NIA 1965 in R(G) 1/70 in which it was said, at , that the issue was “not whether in general the claimant’s marriage … was valid under English law, but the narrow and technical question whether her marriage is valid for the purposes of the 1965 Act”. It was this “obviously unjust” effect (1968 WP at ) which was addressed, first, by section 3 of the NIA 1956 in respect of potentially polygamous marriages which had “at all times been monogamous”. The 1972 and the 1975 Regulations extended this to actually polygamous marriages “for any day … throughout which the polygamous marriage is in fact monogamous”.
195. Accordingly, as Ms Leventhal submitted in her Post-Hearing Note, the premise on which the 1975 Regulations were made was that a party to a valid polygamous marriage was not a widow within the meaning of the substantive social security legislation. The 1975 Regulations were, therefore, effective even if they applied only to valid polygamous marriages.
196. I would also add that, significantly, NA’s marriage was void not only because of section 11(d) of the MCA 1973 but also because of section 11(b). The latter provides that a marriage is void if either party was already lawfully married or a civil partner. This is a question of capacity and, accordingly, determined by the law of the party’s ante-nuptial domicile. Mr A was already lawfully married and, accordingly, he did not have capacity to contract a second marriage, as explained in Baindail v Baindail and Hussain v Hussain. The result is that, in English law terms, his marriage to NA was bigamous and, therefore, void. This would apply wherever the second ceremony of marriage took place.
197. I now turn to my reasons for concluding that the 1975 Regulations do not apply to NA.
198. As referred to above, the primary legislation was interpreted as not applying to a widow from a polygamous marriage, even one which was only potentially polygamous and was recognised as valid under English conflict of laws rules. As a result, absent the 1975 Regulations, a “widow” for the purposes of sections 36 and 39A of the SSCBA 1992 (and its predecessors) did not include a widow from either a potentially or an actually polygamous marriage.
199. Perhaps because of the 1975 Regulations, the effect of the changes made by the MP(PM)A 1972 and the PIL(MP)A 1995 do not appear to have been directly considered in the context of the social security legislation. As referred to above, the latter amended section 11(d) of the MCA 1973 so that a marriage entered into by a person domiciled in England and Wales which was only potentially polygamous was no longer void. The PIL(MP)A 1995 also amended section 121 of the SSCBA 1992 by removing the reference to potentially polygamous marriages but the 1975 Regulations were not, themselves, amended. These changes would certainly seem to me to suggest that the previous interpretation of the social security legislation, as for example in R(G) 18/52 and R(G) 1/70, which excluded a party to a potentially polygamous marriage which was valid under English private international law rules, would no longer apply. I have referred to later decisions which may reflect this (in particular, paragraph 133). In any event, it would be very hard to interpret the legislation as applying to a party to a potentially polygamous marriage who was domiciled in England and Wales (and which is no longer void as a result of the amendment made to section 11(d) of the MCA 1973 by the PIL(MP)A 1995) but as not applying to a party to a potentially polygamous marriage contracted abroad.
200. However, the issue in the present case is the meaning of the word “spouse” in the SSCBA 1992 and whether it can be interpreted as including a party to a marriage which is void under English law.
201. The SSCBA 1992, as originally enacted, continued to base the entitlement to benefits on a person being a widow. Section 26 was replaced, and section 39A was added, by the Welfare Reform and Pensions Act 1999 (“the WFPA 1999”). These conferred entitlement to the relevant benefits on a “spouse”. As was noted by Baroness Hale in McLaughlin, at , the reforms effected by the WFPA 1999 were “part of a general package of welfare and pension reforms introduced by the 1997 Labour Government. But a major spur to their changes to bereavement benefits was that it had become inevitable that widows’ benefits would be successfully challenged for discriminating against men”.
202. The word “spouse” is not defined in the primary legislation. In the absence of any alternative definition, it is clear to me that the word “spouse” cannot be interpreted as meaning a party to a marriage which is void under English law. However, although I put it this way for the purposes of this case, I would also agree with Ms Leventhal’s submission that, because of the developments referred to in paragraph 125 ([3.10] of the 1985 Report) and as explained in paragraph 199 above, “spouse” should be interpreted as meaning a party to a marriage recognised as valid under English law. As a result, I agree with the UTJ when he said, at : “In the absence of any other definition of ‘spouse’ in the SSCBA 1992, one must fall back on the understanding supplied by matrimonial legislation”.
203. Accordingly, in my view, the word “spouse” cannot mean a party to a marriage which is void under English law, for the simple reason that a party to a void marriage is not a spouse. There would have to be some express, or possibly implied, provision which makes it clear that the conventional construction does not apply. There is nothing in the primary legislation which would support this conclusion. In particular, there is nothing to suggest that the introduction of the term spouse (in place of widow) was intended to include a party to a void marriage. Further, as Ms Leventhal submitted, this conclusion is supported by the contrast drawn in sections 36 and 39A of the SSCBA 1992 between a spouse or civil partner and a person living together with another person, to whom they are not married or in a civil partnership, “as if they were a married couple or civil partners”: section 39A(5)(b).
204. That this is the effect of the primary legislation in this case was not significantly disputed. The focus of Ms Rooney’s submissions was on the meaning of the 1975 Regulations and the question of whether they had broadened the scope of the benefits provisions so as to include those who would otherwise be excluded; in particular, whether they applied to a party to a religious marriage which was polygamous.
205. I would first note that, as with the primary legislation, the 1975 Regulations do not contain any definition of the word “spouse”. Accordingly, absent some alternative necessary implication, in my view, it must mean a party to a marriage which is, at least, not void under English law.
206. At first sight, the 1975 Regulations could apply to a party to a religious marriage which was polygamous, as submitted by Ms Rooney, because, as referred to above, they do not make a void marriage valid. It is a deeming provision and provides only that a polygamous marriage is “treated as having the same consequences as a monogamous marriage” (my emphasis).
207. The marriage in this case could also potentially come within the definition of “polygamous marriage” in regulation 1(2). I agree with Ms Leventhal’s submissions about the meaning of “marriage” more generally but this provision refers to a marriage “celebrated under a law which, as it applies to the particular ceremony and to the parties thereto, permits polygamy”. As set out above (paragraph 79), in referring to a law, the regulation is clearly referring to the law of the country where the marriage took place.
208. However, in my view, regulation 2 makes it clear that a void marriage is not included within the scope of the 1975 Regulations.
209. First, considering regulation 2(1) on its own, without regard to regulation 2(2), the only way of determining whether a marriage is “in fact monogamous” would be by determining whether any other purported or alleged marriage was valid under English law. In this respect, contrary to what I understood to be Ms Rooney’s submission, I consider that the expression “in fact monogamous” must mean as a matter of law. This is because, in order for a marriage to be “in fact” monogamous, there would have to be, as a matter law, no other valid marriage. If there is another valid marriage, it cannot be monogamous, regardless of how the parties are living. The determination of the factual question depends on the determination of the legal issue. Accordingly, I disagree with the UTJ’s conclusion, at , that NA’s marriage was in fact monogamous because Mr A “only ever lived with one spouse”.
210. Secondly, and more importantly, regulation 2 contains an express provision as to the meaning of the expression “in fact monogamous”. Regulation 2(2)(a) provides:
“(a) a polygamous marriage is referred to as being in fact monogamous when neither party to it has any spouse additional to the other”.
Because a polygamous marriage will only be in fact monogamous “when neither party to it has any spouse additional to the other”, I can see no way round Ms Leventhal’s submission that this provision only works if NA is a spouse. Otherwise, as she submitted, these words make no sense. This is because both the person seeking to come within regulation 2 and the other party to the relevant polygamous marriage have to be spouses. The question as to whether neither of them has “any spouse additional to the other” necessarily requires them to be spouses. As well as there being no definition of spouse in the 1975 Regulations, there is, in my view, nothing to suggest that it is to be interpreted differently from the primary legislation and other than in accordance with whether the marriage is valid, or not void, under English law.
211. Accordingly, the critical requirement for a polygamous marriage to be within regulation 2(1) “for any day” is that on that day “neither party has any spouse additional to the other”. It is only if this condition is satisfied that the marriage is to be treated as having the same consequences as a monogamous marriage. This is clearly critical because, if regulation 2(1) provided simply that the marriage was to be treated as having the same consequences as a monogamous marriage, one of those consequences would be that the parties to it would be treated as spouses. However, regulation 2(1) does not have this effect. It only applies if the requirements of regulation 2(2)(a) are satisfied.
212. In summary, therefore, a person is only to be treated as a spouse (because their polygamous marriage is treated as having the same consequences as a monogamous marriage) if they come within the scope of the 1975 Regulations which require both parties, for the relevant day or days, to be spouses and to have no other spouse than each other.
213. An additional difficulty is, if someone in NA’s position is to be treated as a spouse, how do you determine whether there is any “additional” spouse? What if the other marriage is also a void polygamous marriage under English law? The parties to that marriage would not be spouses so there would be no additional “spouse”. But, if NA’s marriage was within the scope of the 1975 Regulations, the other marriage would also be a polygamous marriage within their scope, applied separately to it. There would, therefore, be two marriages within the scope of the Regulations. This would seem to require that the words “any spouse additional to the other” would have to include a party to a void polygamous marriage, otherwise two widows would independently qualify. However, I do not see how the word “spouse” can be interpreted as applying to a party to a void marriage so this is another obstacle to the interpretation proposed by Ms Rooney.
214. I would add that, in my view, there is nothing in the legislative history, including that prior to the 1975 Regulations and the preceding 1972 Regulations, nor in the jurisprudence which supports an alternative conclusion. Indeed, the whole history, as referred to above, supports the conclusion that the 1975 Regulations do not include a marriage, or a party to a marriage, which is void under English law.
215. Finally, I would repeat that NA’s marriage was void under both section 11(b) and section 11(d) of the MCA 1973. There is nothing in the 1975 Regulations which suggests that they are intended to apply to marriages which are void because, under English law, they are bigamous.
216. The answer therefore to (i) is that the 1975 Regulations, with sections 36 and 39A of the SSCBA 1992, do not apply to a party to a marriage which is void under English law and, accordingly, NA is not a spouse within the scope of section 36 or section 39A of the SSCBA 1992.
217. I now turn to the next issue:
(ii) Do the 1975 Regulations breach NA’s rights under the ECHR, specifically her rights within A1PI (article 1 of Protocol 1) together with article 14.
218. As the provisions in respect of WPA have already been found to breach article 14, it is only necessary to consider this issue in respect of BP.
219. It is agreed that NA’s claim falls within the scope or ambit of A1P1 and, therefore, that article 14 is engaged.
220. I next deal with the issue of comparability or analogous situation. As Baroness Hale said in McLaughlin, at , this issue has to be addressed “in the context of the measure in question and its purpose”. The measure in question is the grant of a bereavement payment. Its purpose can be seen to be providing financial assistance following the death of a husband, wife or civil partner. As the UTJ asked, the essential question is whether NA’s position is analogous to that of a surviving spouse or civil partner.
221. Ms Rooney argued that NA’s position is analogous to that of a surviving spouse because she and Mr A had gone through a religious ceremony of marriage in Pakistan which was valid under the law of Pakistan. It is clearly arguable that NA’s position is closer to that of a surviving spouse than to a surviving cohabitant. However, I do not consider that, as the UTJ did, there is a “spectrum” of relationships in this context. There is, in my view, “an obvious and relevant difference”, namely the difference between those who have contracted a marriage which is valid under English law and those who have not. Marriages can be void for a number of reasons and I do not see how the position can vary or depend on the reason for the marriage being void. The focus in McLaughlin was on the “public contract” because the court was analysing the difference between a married couple and a couple who had not entered into “the act of marriage”. In all cases involving void marriages, the parties will inevitably have undertaken some act or ceremony. This will very probably be a public act or ceremony but the critical distinction is that it will not be an effective “public contract”.
222. Accordingly, in my view, NA’s position as a party to a religious marriage which is void in English law is not analogous to a party to valid marriage. A religious ceremony of marriage performed in England and Wales might create a valid marriage, a voidable marriage, a void marriage or it might be a non-qualifying ceremony. Taking the facts of the present case, in my view a party to a religious marriage performed in another country which is void because it is bigamous is in an analogous position to a party to a religious marriage performed in England which is void because it is bigamous. It is the bigamous nature of the marriage which is the relevant and important feature not that the marriage was polygamous nor that the marriage was a religious ceremony.
223. Accordingly, contrary to the UTJ’s decision, I do not consider that NA is in an analogous position to a party to a marriage which is valid, or not void, under English law. It is, in my view, a clear distinction of the nature identified by Lord Nicholls in R (Carson) v SSWP, namely “an obvious, relevant difference”.
224. Another way at looking at the question is by focusing on the effect of the 1975 Regulations. It could be argued that what are being compared are a surviving spouse of a valid polygamous marriage and a surviving party of a void polygamous marriage. Are they in an analogous situation? Again, in my view, they are not because it is not possible to focus only on the reason, in this case, for the marriage being void. As Ms Leventhal submitted, there a number of reasons why a marriage might be void. I do not consider that the issue of analogy can be phrased in any way other than by reference to one marriage being void and the other valid. If the 1975 Regulations did not exist at all, and if the law had not developed as it has, I could see a strong argument for a surviving spouse from a valid polygamous marriage being in an analogous situation to a surviving spouse from a valid monogamous marriage. However, that is not this case.
225. Despite my conclusion on the above question, I nevertheless propose to deal with the other matters relevant to this issue.
226. I next deal with the issue of justification because, if NA was in a relevantly analogous position to a widow entitled to BP, then she would be likely to have a status within the scope of article 14.
227. The issue in the present case, as agreed at the hearing, was whether the justification advanced by the SSWP, as supporting the exclusion of someone in NA’s position from entitlement to BP, was “manifestly without reasonable foundation”. In this respect also, I disagree with the UTJ’s conclusions. I should add that I do not consider that the more nuanced approach identified by the Supreme Court in R (SC) significantly impacts on the approach to be taken in this case. This is because, as Lord Reed said, at , “the courts should generally be very slow to intervene in areas of social and economic policy such as housing and social security”. This is not a case in which there are factors which would require a higher degree of justification. Or, to put it the other way, this is a case in which, adopting what Lord Reed said, at , “the ordinary approach to proportionality will accord the same margin to the decision-maker as the ‘manifestly without reasonable foundation’ formulation” because the circumstances are such that a particularly wide margin is appropriate. I would also add that, even if a smaller margin was appropriate, I have no doubt, for the reasons set out below, that the difference of treatment is justified.
228. As Baroness Hale noted in McLaughlin, at , marriage has been recognised by the ECtHR as conferring a “special status”. She also said, at , that “the promotion of marriage, and now civil partnership, is a legitimate aim”.
229. In my view, it is difficult to see how the objective of supporting marriage could be achieved other than by providing that BP is available only to those who are legally married as a matter of English law. The UTJ considered, at , that this was a circular argument. I agree that the reasons why the particular marriage is not regarded as a valid marriage cannot be ignored but, subject to that, confining entitlement to those recognised as spouses under English law provides, in my view, powerful justification for the effect of the legislation.
230. Further, I can see no logical basis for drawing the line so as to include some of those who are parties to a void marriage or, indeed, those who are parties to a non-qualifying ceremony. It is also relevant to note, as submitted by Ms Leventhal, that the legislation provides for a voidable marriage to be treated as having been a valid marriage terminated by divorce. In my view, the only logical place to draw the line is between valid and void marriages. The benefit provides support for those whose marriage is recognised as valid and, as a result, have the status of spouse (or civil partner). A marriage with this legal effect is clearly distinguishable from a ceremony which does not create a valid marriage and, in particular, one which creates a void marriage (a tautologous expression because, as referred to above, it is of no effect at all).
231. The UTJ considered, at , that the public policy reasons were “primarily framed in … terms of the state favouring marriage over cohabitation”. That may be one aspect of the difference but there are also strong public policy reasons supporting differentiating between valid and void marriages. As explained in Akhter v Khan, at , it is in the interests of the state to know “whether what [the parties] have done creates a marriage which is recognised as legally valid”; and , “It is also in the interests of the state that the creation of the status is both clearly defined and protected”.
232. This is a bright line based on what, I would suggest, is a clear distinction. The UTJ concluded that the line was “distinctly dim” because domicile “can be notoriously difficult to assess” and that, accordingly, section 11(d) of the MCA 1973 “fails in practice to provide legal certainty”. I do not agree with this conclusion. In my view, the UTJ focused too narrowly on the issue of domicile and did not consider the effect of section 11 as a whole.
233. Section 11 provides a clear legal line in that it makes clear when a marriage will be void. I do not consider that evidential difficulties which might arise in the determination of the issue of domicile undermines or diminishes the bright line established by section 11, anymore than might difficulties in determining whether the marriage is void under section 11(a)(iii) because it is not valid “under the Marriage Acts 1949 to 1986” because the parties have married “in disregard of certain requirements as to the formation of marriage”. Such potential difficulties, in my view, do not diminish the effect of the bright line relied on for the purposes of the present case, namely the bright line between those marriages which are void and those which are not.
234. This line is also not impacted by the effect of the 1975 Regulations because, if I am right that they only apply to marriages which are not void, the bright line is maintained. In addition, I do not consider the fact that some polygamous marriages are recognised as valid undermines the SSWP’s submissions on the issue of justification. First, because it is the fact that they are valid under English law which is significant. Secondly, as explained below, there are sound reasons for providing that actually polygamous, and bigamous, marriages by those domiciled in England and Wales are void.
235. In my view, there are powerful public policy reasons which support differentiating between void and valid marriages including between valid polygamous marriages and polygamous marriages which are void because, under English law, they are bigamous under section 11(b). I do not consider that there is any basis on which the ground on which the latter are void could be successfully challenged. I consider it clearly justifiable to require those domiciled in England and Wales, a strong connecting factor, to have the capacity validly to contract the relevant marriage which does not include an actually polygamous marriage. This law applies to all marriages wherever performed and to everyone domiciled in England and Wales.
236. Further, although Ms Leventhal’s submissions focused on public policy being antithetical towards polygamy, the real issue in this case is that there are powerful public policy reasons for prohibiting marriages which are actually polygamous (i.e. bigamous). The approach to potentially polygamous marriages has changed significantly over the years, as set out above, including with the changes effected by the PIL(MP)A 1995. However, in my view, maintaining a legal structure which discourages bigamous marriages and makes bigamous marriages void for those domiciled in England and Wales cannot be said to be without reasonable foundation, let alone manifestly without reasonable foundation. Indeed, as set out above, I consider that the structure and effect of the legislation is clearly justified.
237. Finally, I would note that this case is different from the situation considered by the Grand Chamber in Yiğit v Turkey. The present case is not about the recognition of marriages but about the circumstances in which a marriage is void. The fact that the 1975 Regulations extend the entitlement to benefit to valid polygamous marriages does not undermine the exclusion of a party to a marriage which is void under English law. In any event, there is nothing in Yiğit v Turkey which suggests that the distinction between a valid actually polygamous marriage and a void actually polygamous marriage conflicts with article 14.
238. Accordingly, in my view, the answer to issue (ii) is that the legislation does not breach NA’s rights under the ECHR.
239. The last issue is:
(iii) If the legislation does breach NA’s rights, can the 1975 Regulations be read down so as to apply to NA.
240. This issue does not arise because of my conclusions in respect of issues (i) and (ii). I propose, therefore, only to say that, in my view, to read the 1975 Regulations as applying to void polygamous marriages would be contrary to the “grain” of sections 36 and 39A of the SSCBA 1992 and, indeed, would be inconsistent with a fundamental feature of the legislation, including the 1975 Regulations, namely that it applies only to spouses and civil partners.
241. For the reasons set out above, in my view, this appeal must be allowed and the UTJ’s determination set aside. NA is not entitled to benefits under either section 36 or section 39A of the SSCBA 1992. Her exclusion from entitlement to BP under section 36 is not discriminatory. Her exclusion from entitlement to WPA under section 39A is discriminatory in accordance with the decision in McLaughlin. A declaration of incompatibility has already been made in respect of the latter.
LADY JUSTICE MACUR:
242. I agree that the appeal should be allowed although I respectfully disagree with some of the reasons given by my Lord, Moylan LJ in his interpretation of the 1975 Regulations in respect of issue (i). I am in full agreement with my Lord, Moylan LJ, in relation to issues (ii) and (iii). I adopt the abbreviations used above for the purpose of the short judgment that follows.
243. Specifically, in so far as the interpretation of the 1975 Regulations is concerned, I would agree with the UTJ at  that the ‘meaning of spouse is not central’ to their interpretation, having recognised as he did that, for the purpose of SSCBA 1992, the term is to be understood in accordance with matrimonial legislation. I also agree with the UTJ that the term ‘in fact monogamous’ in the context of the 1975 Regulations refers to the factual status of the parties’ union. Nevertheless, this does not lead me, as it did not lead the UTJ, away from the ‘conventional or orthodox’ reading of the 1975 Regulations to only apply to valid polygamous marriages.
244. If ever my reason to take this different route to the same outcome in relation to issue (i) becomes relevant, it is precisely because section 162(b) of the SSA 1975 and the 1975 Regulations do not seek to legitimise a void polygamous marriage that I disagree with my Lord, Moylan LJ’s justification of his interpretation of Regulation 2 by reference to a valid polygamous, or monogamous, marriage. I think it is unnecessary to make the comparison. That is, if, which I do not accept, the effect of Regulation 1(2) of the 1975 Regulations was to deem void polygamous marriages as acceptable for the purpose of entitlement to social security benefits, then I do not see why the ‘wives’ of those ‘marriages’ would not be deemed ‘spouses’ pursuant to Regulation 2. As such, I cannot foresee any circumstances in which more than one wife would be able to qualify for benefit under the terms of Regulation 2, for if both are treated as spouses then there would be an additional spouse.
245. However, I interpret Regulation 1(2) of 1975 Regulations to refer to a valid polygamous marriage since I regard the underlined part of the text, ‘a law which, as it applies to the particular ceremony and to the parties thereto’, must be a reference to the personal capacity of the parties rather than the formalities of the ceremony. If either party is subject to a law which does not permit polygamy, as in this case, then the ceremony of marriage they undergo will not lead to a polygamous marriage for the purpose of the 1975 Regulations and is unable to be deemed a monogamous marriage for the purpose of the relevant social security benefit. Notably, the underlined words do not appear in the text of section 162(b) of the SSA 1975, that Regulations may provide - “as to the circumstances in which for the purposes of this Act - (i) a marriage celebrated under a law which permits polygamy, … is to be treated as having or not having the consequences of a marriage celebrated under a law which does not permit polygamy” which connotes to me that the 1975 Regulations refer to an additional requirement to that of the formalities of the ceremony. I do not see the fact that Mr A’s wedding to NA was also bigamous is significant to the issues in the appeal.
246. Finally, in support of my interpretation of the Regulations, I agree with my Lord, Moylan LJ, that there is nothing in the legislative history prior to the Regulations nor in the jurisprudence, so comprehensively and carefully detailed in his judgment above, which supports the conclusion that the 1975 Regulations are intended to deem a void polygamous marriage as a monogamous marriage for the purpose of the relevant benefit. I do not doubt that the mischief at which the 1975 Regulations was aimed was the failure to recognise, for the purpose of social security benefits, the valid polygamous marriages celebrated in accordance with proper formalities by those with capacity to enter into such unions prior to acquiring domicile in the UK.
LORD JUSTICE UNDERHILL:
247. I agree that this appeal should be allowed, but since I have not found the case easy I will shortly summarise my reasons in my own words. I will adopt Moylan LJ’s abbreviations.
248. As regards “issue (i)”, I was initially attracted by the straightforward argument that the effect of regulation 1 of the 1975 Regulations was that all that was necessary to attract the operation of regulation 2 was that the polygamous marriage in question was valid according to the law under which it was celebrated. Since Mr and Mrs Akhtar’s marriage was “celebrated under a law which … permits polygamy”, namely the law of Pakistan, “as it applie[d] to the particular ceremony and to the parties”, it was irrelevant that (because of Mr Akhtar’s British domicile) it was void as a matter of English law: the effect of regulation 2 was that the only thing that mattered was whether it was in fact monogamous at the relevant date, i.e. that of Mr Akhtar’s death.
249. In the end, however, I have been persuaded by Moylan LJ’s clear and painstaking analysis of the legislative history (which I should say is rooted in counsel’s and Helen Walker’s thorough researches) that that approach does not correspond to Parliament’s intention in enacting section 12 of the NIA 1971 or section 162 of the SSA 1975, nor therefore to the intention of the Secretary of State in making, first, the National Insurance, Industrial Injuries and Family Allowances (Polygamous Marriages) Regulations 1972 (“the 1972 Regulations”) or, then, the 1975 Regulations. It is necessary to start with section 3 of the NIA 1956. As Moylan LJ demonstrates at paras. 76-78 above, the Report of the National Insurance Advisory Committee, which was the genesis of that provision, was concerned specifically with the problem of marriages which were only potentially polygamous and was plainly directed at the line of decisions which began with R (G) 18/52, which was a case of that kind; and that is reflected in the terms of the section as enacted. Importantly, as he also demonstrates, an essential part of the Committee’s thinking was that such marriages were regarded in English law as valid for some purposes (Baindail v Baindail), albeit not for the purpose of matrimonial relief (Hyde v Hyde), so that all that was being done was to require that for national insurance purposes they be treated as being in the former rather than the latter category. Of course the position was then modified by the 1972 Regulations, but it seems clear that they proceeded on the same basis, even if the thinking in the 1968 Working Paper and the 1971 Report is not quite so explicit: see paras. 93-97 above. Section 12 of the 1971 Act was then replaced by section 162 of the 1975 Act, but although, as Moylan LJ explains in para. 107, there was some difference in the drafting, that was not the result of any reconsideration of policy as regards the current issue; and, as he says, the 1972 and 1975 Regulations are in identical terms. In short, the legislation in its various iterations was only directed at the treatment for social security purposes of kinds of marriage which were already recognised as valid for some purposes and not to marriages which were regarded in English law as void.
250. Against that background, I do not think that it is possible to treat either section as intended to empower the Secretary of State to take the very significant step of permitting a marriage that would otherwise be regarded in English law as definitively void to be treated as having the same consequences for social security purposes as a valid marriage. Legislation must always be construed with regard to the mischief to which it is directed. (I should say that I have not myself thought it right to have regard to developments in the legislation and case-law after 1975, since I do not think that they can affect the construction of the Regulations as enacted.)
251. I do not think that there is any difficulty in reading the phrase “polygamous marriage” in regulations 1 and 2, in order to give effect to the evident statutory intention, as referring only to such a marriage which is not regarded as void as a matter of English law. Ms Rooney submitted that that involved implying words into the 1975 Regulations. Even if in one sense that may be so, I do not think that it is a real objection. Other things being equal, a reference in domestic legislation to persons being married would naturally be understood as a reference to a marriage that was valid as a matter of domestic law.
252. Although my essential reasoning is as stated above, I should say that my conclusion is reinforced by the point made by Moylan LJ at para. 213.
253. I have read Macur LJ’s judgment. I am, with respect, not persuaded by the alternative route to the same result which she takes at para. 245, because I read regulation 1 (1) (a) as proceeding on the basis that the marriage in question is celebrated under a single system of law - “a law” - which permits polygamy. The purpose of the phrase “as it applies to the ceremony and to the parties thereto” is to make clear that the law in question must permit the particular marriage between those particular parties: it is not, as I read it, to import into the definition the capacity requirements of a different system of law. I think the more straightforward route is, as suggested above, simply to construe the term “polygamous marriage” in the context of the legislative history and the mischief to which the statutory power was directed. I agree, however, with her observation at para. 243 that the use of the term “spouse” in sections 36 and 39A of the SSCBA 1992 is not of central importance.
254. As regards issue (ii), my reasons are, I believe, substantially the same as Moylan LJ’s. In bare outline, if, as I believe, the intention of the legislation is to distinguish between cases where the parties were and were not validly married as a matter of domestic law, that is a legitimate distinction in the context of entitlement to a benefit of this character. Lawful marriage (or civil partnership) is a well-recognised status of fundamental importance in our society and one which it is entirely appropriate should be defined by formal rules. It is a reasonable legislative choice to limit entitlement to bereavement payments only to the surviving party to a marriage or partnership which is formally valid, even if there may be occasional hard cases where the validity of a marriage is vitiated by a defect of which the surviving party was unaware. That argument can be expressed equally as going to “analogous position” or to justification: those questions typically overlap in the article 14 context. I agree with Moylan LJ that McLaughlin is distinguishable for the reasons that he gives.
255. On that basis issue (iii) does not arise. If it did, I am not sure that I would agree with Moylan LJ that it would be impossible to read the Regulations down in the way that the Upper Tribunal did; but it is unnecessary to consider the point further.