The Government’s proposal to introduce same-sex marriage seems to rest on reasons of equality, stability and convenience. But on closer inspection, these are respectively incomplete, speculative and negligible. As currently defined, marriage secures the equal value of men and women. It also promotes the welfare of children. By contrast, the new definition of marriage will unavoidably call into question its exclusivity, its permanence and even its sexual nature. Such an unravelling of marriage is too high a price to pay for a proposal which fulfils no practical legal need.
On 15 March 2012 the British Government announced proposals to allow two people of the same sex to enter into marriage by way of civil ceremony. The responses of major Protestant churches to the proposal have depended substantially on arguments rooted in Scripture and Christian tradition. There is nothing wrong with this. In a multi-faith democracy we should be encouraged to disclose our deepest convictions; and where a majority of the population still identify themselves as Christian, one can expect Christian convictions to carry some weight. However, plural democracy will only survive if we also offer each other reasons we can expect each other to share. This is all the more important in an area which shows signs of collapsing into a ‘culture war’ in which mutual hostility takes the place of collective rational deliberation. If the only reasons against (or for) same-sex marriage are ‘ideological’ or a matter of gut reaction then all we are left with is mutual incomprehension.
The purpose of this paper is to set out a non-religious case for retaining the current legal definition of marriage. It does not seek to question the morality of same-sex relationships, the provision of civil partnerships, or the current law prohibiting discrimination on grounds of sexual orientation. Rather it seeks to answer the fundamental question: should we think of marriage as the sort of arrangement two people of the same sex can enter into? Clearly, a same-sex partnership is like a marriage in some respects, but is it indistinguishable? What, then, is marriage?
The common law has always defined marriage as the voluntary union for life of one man and one woman to the exclusion of all others. Parliament has altered the age at which people may marry, the family relationships within which marriage is prohibited, the formalities by which marriage may be commenced, the proprietary and financial consequences of marriage, the extent to which aspects of the common life of husband and wife may be enforced by law, and the circumstances and process by which a marriage may be dissolved. But the essential definition of marriage itself has remained unchanged.
At root, the meaning of marriage is socially, not legally, defined. When a couple decide to get married, they do not start negotiating the clauses of an open-ended contract. Rather, they assume that there is a pre-existing and familiar type of relationship they are about to enter. Marriage has socially-given expectations, purposes or goods, which are intrinsic to the relationship. This is what we mean by saying that marriage is a social institution. The legal definition of marriage reflects and supports this social consensus, and it seems fair to assume a rather close relationship between legal form and social expectation. Changing the legal definition of marriage will likewise reflect and support a different view of what marriage is and what it is for.
The Government was therefore right to accept that everyone has a stake in its proposal, but disingenuous in distinguishing between ‘civil’ and ‘religious’ marriage. Legally, the distinction is non-existent: marriage may be entered into by way of civil or religious ceremony, but the status and implications are identical. Socially, as well, we don’t think of people as ‘religiously married’ or ‘civilly married’. They are just ‘married’. The debate about same-sex marriage is a debate about using law to change the meaning of the social institution of marriage. And that affects everyone.
The Government’s arguments
The argument from equality
Ideas of equality dominate the Government’s proposal, which also speaks in terms of ‘fairness’ and ‘rights’. In simple terms: the current law of marriage discriminates on grounds of sexual orientation. Discrimination on grounds of sexual orientation is unjust, so the current definition is unjust. The same idea underlies the suggestion that same-sex partners are ‘banned’ from marrying.
This argument assumes what needs to be proved. Any law which sets criteria for anything discriminates. When the law discriminates on grounds of a ‘protected characteristic’ (such as sex, race, age, religion or sexual orientation) there is, of course, a presumption that the discrimination is wrongful. In most public contexts, such as political life, business or employment, these distinctions should not be drawn. But sometimes it is right to draw distinctions even on these grounds. For example, the law of marriage discriminates on grounds of age. One must be at least 16 years old. In our society, we do not consider this discrimination to be unacceptable, because it is justified. It is right that children should not marry. The law of marriage also discriminates on grounds of kindred and affinity: one cannot marry a parent, sibling or child, nor a range of more distant relations. But once again, we think this is justified – marriage between close blood relatives is medically unwise and in any case sexualised families are abusive and oppressive – so we do not think of it as discrimination. It is just good moral sense. We need to watch out for unjustified slippage between ‘discrimination’ as the drawing of any distinction and ‘discrimination’ as the drawing of an inappropriate distinction.
If anything, ‘equality’ proves too much. Whatever its causes, human sexuality is complex and variable. The commonest form of minority sexual orientation would appear to be bisexuality. This is also a matter of degree. Many people have had at least some fleeting experience of attraction to someone of the same sex. There are people not sexually attracted to others, or who wish to enter long-term stable but non-sexual relationships of companionship. There are also people who only wish for brief sexual encounters; some of whom are prepared to pay for them. And there are those who enjoy the challenge of maintaining multiple sexual relationships simultaneously. Self-awareness, desire, social convention, choice, behaviour and identity are all intertwined in complex ways. The normal condition of human sexuality seems diverse and even chaotic at times.
It follows that for marriage to be ‘equal’ on grounds of sexual orientation, the law should not be restricted to just one type of sexually-intimate companionship. Why can’t a man marry two wives? Why isn’t prostitution treated as a form of short-term marriage? Why, for that matter, should single people be deprived of the chance to pass on their pension rights to a best friend? There may or may not be reasons for drawing the legal boundaries in any particular place, but until those reasons are stated, the argument from equality is incomplete.
The argument from stability
The argument from stability figures less prominently in the proposal. But the argument that marriage is a relatively stable social institution, and that stability is good, plays a supporting role. By allowing two people of the same sex to marry, society will support the stability of that relationship.
It is fair to assume that, practically speaking, no same-sex couple would be willing to marry who are not already willing to enter into a civil partnership. So the argument depends on the relative stability of marriage as opposed to civil partnership. As far as the law is concerned, the difference is slight. One enters into marriage by making a public declaration of intent; one can enter into civil partnership by merely signing the civil partnership document. But the truth is that we simply do not know the relative stability of marriage and civil partnership. Civil partnership has only been available since 2005. The data so far might even imply that civil partnerships are more stable than marriages.
Even if we could show that marriages have a greater longevity than civil partnerships, we would not have shown that allowing civil partners to marry would make any difference. Whichever way the data goes, it will always be more likely that relative stability is an effect of the nature of the relationship and its social context rather than its legal form. Furthermore, the argument is undercut by the Government’s proposal to allow a choice between civil partnership and marriage. The logic of the argument requires all parties to take vows, not to be given a choice of legal forms. In short, the argument from stability is speculative.
The argument from convenience
The difficulties of a small group of people are emphasised at some length by the Government. Currently, the marriage or civil partnership of a transsexual person is automatically dissolved on receiving their newly assigned gender. If their spouse or partner is willing to continue the relationship, a new civil partnership or marriage must be entered into. By making marriage gender-blind, this will no longer be necessary. Rather, the parties could choose to affirm and continue their marriage.
This problem affects only a tiny number of people and the inconvenience of entering into a new civil partnership or marriage must be minor compared to the trauma of gender dysphoria and medical and psychological treatment for reassignment. In any case, civil partnerships would still have to be dissolved in cases of gender reassignment, which shows that again the Government does not take its own argument seriously. The argument from convenience is negligible.
The Government’s proposal relies on three arguments for changing the legal definition of marriage. They are, respectively: incomplete, speculative and negligible. If anything, they suggest that there should be only one legal arrangement, ‘marriage’, which should be available to a much wider range of relationships than the proposal envisages. The root of the problem is that the Government fails to address the fundamental question of what a marriage is, and thus it fails to identify and defend the boundaries of any new definition.
Two arguments against same-sex marriage
Marriage secures the equal value of men and women
One of the goods of marriage is that it confers social recognition on a relationship which is dependent on the gendered ‘other’. Thankfully, we live in an age and society which has done more than most to ensure that gender roles are fluid, that men and women are equally able to access jobs, careers and other social opportunities, as well as taking up domestic responsibilities. Yet we still recognise that men and women are in various ways different. The point about equality for men and women is not that the difference is irrelevant, but that both are equally valuable and necessary. We may struggle to identify all the dimensions of that difference, and disagree about their significance, but as many feminist writers have rightly recognised, ‘gender-blindness’ is not the answer. Gender-blindness runs the risk of entrenching norms and practices which typically favour men and are oppressive towards women. The fact of difference has to be acknowledged and valued if we are to secure equality. That is why we are right to worry about the small number of women MPs or CEOs.
Marriage as currently defined is the central social institution which expresses the idea that men and women are equally valuable as men and women. It is only marriage which harnesses gender difference to the purposes of social cooperation. Almost all other ways in which difference is acknowledged – from sports teams to public lavatories – depend on segregation. Sexual union in marriage reinforces a comprehensive ‘together-in-otherness’ of male and female. Yet marriage is also open to a wide range of views on the practical outworking of this difference. It would be ironic if after having reformed the patriarchal consequences of marriage the institution itself should become gender-blind.
The grounding of interdependence in marriage has wider social consequences. It anchors our understanding that it is normal for all social institutions to rely on this mutual interdependence. The excessive individualism of modern Western society, as well as the collapse of participation in all forms of organised social action, has often been observed. Redefining marriage to be indifferent to sexual identity reinforces this individualistic tendency because it turns human society – from marriage outwards – into a matter of individual inclination and choice. Marriage will cease to be an institution which reflects the necessary and natural interdependence of men and women. Every successful relationship depends on some degree of mutual dependency, but marriage is distinctive in recognising and valuing a difference that is deeper than those of character and circumstance. The man-and-woman criterion not only secures gender interdependence in marriage, it also models and promotes this gender interdependence more widely across society.
Marriage promotes the welfare of children
While the connection between the current definition of marriage and the equal value of men and women is quite clear, the implications for children are less obvious. Marriage and childbearing have become disconnected in historically unprecedented ways. More precisely, childbearing has become a matter of unprecedented choice. The availability of effective contraception makes decisions to avoid childbearing more reliable. Medical advances have increased the possibilities of childbearing for infertile and older couples. Individuals and same-sex couples can now take steps to acquire a child by adoption or by making medical arrangements.
In spite of these developments, there are still connections between marriage as currently defined and the bearing and rearing of children. Most married couples of childbearing age will be able to have children and will have to take steps to avoid having children. By contrast, individuals and same-sex couples have to take active steps to acquire a child, at some point involving another party. Far fewer do. So there is a social and practical presumption connecting marriage with children. But the connections go deeper than this.
A child represents the combined genetic inheritance of his or her parents. She is the embodiment of both, a living testimony to their intimacy and a bearer of the identity of earlier generations. A child is likely to outlive her parents, and so this embodiment symbolises the permanence of marriage as well as its heterosexual nature. We know that these multiple intertwined relationships are important, both on account of the strong natural desire people have to know their genetic parentage as an aspect of their identity, and because of the pain a child experiences on divorce. The child can feel almost literally torn apart. This means that there is a distinction between sexual union and sexual intimacy. A sexual relationship can be called a ‘union’ not because it is intense, but because it can be embodied in a new human person. Only a man and a woman can form the biological unit capable of pro-creating another being ‘free and equal in dignity and rights’. No new human being can exist as a living expression of the intimacy of a same-sex couple.
Redefining marriage to include same-sex partnerships will have two distinct effects. As far as same-sex couples are concerned, it will close off the question of the impact of same-sex parenting on children. But more importantly, it will sever the presumptive connections between marriage, childbearing and kinship for everyone.
As far as the first point is concerned, there is an inconclusive and controversial debate about the impact of same-sex parenting on children. Quite apart from the high political stakes, there are the problems of what one is comparing same-sex parenting with (single parenthood, step-parenthood, institutional care, natural parenting?) and what counts as a negative outcome. Throw in the distorting factors of small sample sizes, social class, education, other disrupting life-events, and the scope for contestation is enormous. The prevailing view is that there is no significant deficit in same-sex parenting, although a recent major study has called this into question. However, we do know that the distinctive gender roles of a father and a mother are important in the psychological development of children. It must be at least possible that having two ‘fathers’ or ‘mothers’ will not compensate for the absent mother/father-figure. The data is inevitably recent, and we will not know with certainty for some considerable time what the effects of same-sex parenting are. Recent changes already allow for same-sex partners jointly to acquire children; redefining marriage will render these developments immune from reconsideration. Such confidence seems premature.
But the more important effect is on the place of children within marriage generally. Every child has a moral claim on her natural father and mother, grounded in the fact that they brought her into being and that it is in principle good for every child to be brought up by her natural parents committed in relationship to each other and to her. Redefining marriage breaks the necessary connection with childbearing, in the sense that marriage will no longer mean ‘the relationship which is normally and naturally productive of children and thus a nexus of kinship.’ Its intrinsic purpose will be reduced to sexually-intimate companionship. And it is bad for children if their parents come to believe that all that being married means is sexually-intimate companionship. Every child needs his or her parents to think of their marriage as embodied in their children. Children benefit from strong wider networks of kinship, not least grandparental care, and it is unrealistic to suppose that these can be created merely by contract. Breaking the intrinsic connections between marriage, childbearing and kinship risks the further commodification of children, in which children become ‘ultimate accessories’ – means to the ends of their parents, and ultimately subject to their agendas, rather than persons of equal worth, with an equal stake in the success of the marriage.
What marriage will mean
The Government’s proposals fail to define the substance of marriage and thus fail to distinguish rationally between relationships and arrangements which are and are not to be treated as marriage in law. Presumably, it is assumed that marriage will mean an exclusive and permanent sexual companionship between any two adults. But if the proposals are implemented, what will marriage come to mean?
The unspoken rationale for changing the meaning of marriage is most plausibly based on radical choice.Given that civil partnership already provides an equivalent legal framework with similar practical effect to marriage, the campaign for same-sex marriage is best seen as an attempt to confer full social recognition on minority sexual identities. The basis for that recognition lies in the view that no social disadvantages should attach to different sexual choices. Rather, individual decisions in this area should merit equal public respect. The problem is that this rationale extends far beyond exclusive and permanent sexual companionships. It calls into question other definitional boundaries as well.
The exclusivity of marriage will be undermined
The proposal to extend marriage to same-sex partners assumes that marriage is exclusive to two people. Why should this not be a matter of personal choice? From a purely pragmatic and political point of view, if a minority defined by sexual orientation can get marriage redefined to suit their wishes, why can’t other minorities? Traditional Islam allows a man up to four wives. Polyamorists argue for no limit to the number of partners one may choose. These are real options in the social ordering of human sexuality.
Sexual interdependence expresses the exclusivity of marriage. The fact that the human race is relevantly male and female in this context makes sense of our intuition that sexual exclusivity matters. Between them, a man and a woman represent humanity and its future in a way that two people of the same sex do not. If two people of the same sex can enter into marriage on the grounds of sexual intimacy and companionship, then why not three?
The permanence of marriage will be undermined
In spite of the relatively easy availability of divorce, marriage is still lifelong in intention. Permanence serves important personal, social and economic goals. The fragility of marriage is a major cause of harm in twenty-first-century British society. The problem is that stability requires the genie of autonomous choice to be kept firmly inside its bottle. Why should one be so foolish as to sign away one’s future choices for the rest of one’s life? If one is allowed to choose the terms on which one enters marriage, why should one not choose a less permanent form of relationship?
The permanence of marriage is symbolised by the life of the child who embodies it. But if marriage includes the choice of a relationship which has no intrinsic connection with procreation, why should it not also include the choice of a time-limited relationship? The costs for children are unlikely to prevail against the trump card of autonomy. Again, this is not a fanciful suggestion. Other societies have known forms of fixed-term bonding, and campaigns are now under way in countries which have accepted same-sex marriage to render this element a matter of choice also.
The sexual dimension of marriage will be undermined
In law, marriage is a sexual relationship. Incapacity and wilful refusal to consummate a marriage are grounds for annulment, and adultery is one of the five facts which demonstrate irretrievable breakdown. While in theory one could imagine courts trying to identify same-sex analogies, in practice the law will have to draw a veil over the sexual dimension of the relationship, subjecting disputes to the broad test of ‘unreasonable behaviour’. The Civil Partnership Act 2004 certainly does not attempt to fix the sexual dimension of a same-sex partnership, which is thus treated as a matter of choice and agreement. A close same-sex companionship need not be sexually active, so marriage and sex will be similarly disconnected.
Given that civil partnership need not be a sexually-active relationship, the refusal to allow close relatives to enter into that status is illogical. Of course, the state has an interest in preventing children being born of incestuous unions, but this is already covered by the criminal law of sexual offences. It should have no further interest in preventing formal companionships between unmarried family members. Marriage will become like civil partnership in this respect, taking on the same illogicality. If marriage includes relationships which are not necessarily sexual, the companionship of any adults who aspire to formal recognition and its other advantages could be called a marriage.
The moral weight of marriage will be undermined
As indicated at the start of this paper, we cannot expect political debate in a plural democracy to rely solely on theological arguments. However, marriage as currently defined derives at least part of its social significance from its historic location in Christian theology. Only in the last generation or so has the Church of England ceased to play a central role in marriage/divorce law reform. Redefining marriage will disconnect it from this heritage and undermine its moral weight and social significance.
In traditional Christian teaching, God created human beings male and female. In marriage, the two halves of humanity come together to share their life and work, and as procreators of yet more human beings. The insufficiency of the individual points also to human dependency on God, just as the joint capacity to ‘procreate’ children is testimony to our creation in his image. Furthermore, marriage is a symbol of the mutually loving and dependent relationships between the different persons of the Trinity. It also pictures the relationship between Jesus Christ and his bride, the church. The exclusivity and permanence of marriage represent the faithfulness of the one true God, and our commitment to him. The pleasure of sexual union points to the greater joy of spiritual union with God. Marriage is part of the created order which is capable both of considerable social distortion and also of positive transformation in the light of the deepest theological symbolism. This is why so many Christians take complementarity, exclusivity, permanence and sexuality in marriage so seriously.
It is likely that this intense theological symbolism is a subliminal part of why we all think marriage matters so much. One does not need to share the Christian worldview to accept that the social recognition which the label of marriage currently confers in British society depends to a large extent on this submerged theological memory. But redefining marriage to include same-sex partnerships undercuts that memory, and thus also eventually the social recognition to which movements for various forms of sexual liberation aspire.
The proposal to change the current definition of marriage depends on a sense that the man–woman criterion confers no distinctive social goods and represents an arbitrary limitation. But this is not the case. Marriage affirms the equal value of men and women, and promotes the welfare of children. Moreover, the logic of equal recognition and radical choice means that the boundaries of any new definition will be far more vulnerable. Challenges to its exclusivity, its permanence and even its sexual nature will be unavoidable. Marriage risks becoming any formalised domestic arrangement between any number of people for any length of time. On such a trajectory, marriage will eventually unravel altogether.
Julian Rivers has contributed regularly to Cambridge Papers for the 20 years of their existence. He has published widely in constitutional law, legal theory, and law and religion studies. He is Professor of Jurisprudence at the University of Bristol Law School, an editor-in-chief of the Oxford Journal of Law and Religion, and a member of the advisory board of the Ecclesiastical Law Journal.
 Government Equalities Office, Equal civil marriage: a consultation, March 2012. Since the law of marriage is a devolved matter, the proposal relates only to England and Wales. On 25 July 2012, the Scottish Government announced its intention to enact similar legislation.
 See A Response to the Government Equalities Office – “Equal Civil Marriage” – from the Church of England (June 2012); Response from the Methodist Church in Britain to the consultation on “equal civil marriage” (June 2012). By contrast, the Response from the Catholic Bishops’ Conference of England and Wales to the Government Consultation on “Equal Civil Marriage” (June 2012) relies on a ‘common and instinctive understanding of the meaning of marriage shared by people of any religion and none’.
 Whether a majority is still ‘Christian’ depends on how one phrases the question. In the 2001 National Census, 71.8% stated that they were ‘Christian’ (Office for National Statistics, Focus on Religion (2004), Table 1). By contrast, the British Social Attitudes Survey records a steady drop from 66% (1983) to 44% (2010) in respondents who regard themselves as ‘belonging’ to the Christian religion. The 2010 survey was the first in which 50% stated that they ‘belonged’ to no religion. (NatCen Social Research, British Social Attitudes 28 (2011–12), Table 12.1).
 Jürgen Habermas writes critically of the pressures on religious believers to be ‘polyglottal’ in a plural democracy (‘Religion in the Public Sphere’, in European Journal of Philosophy, 14:1, 2006, pp.1–25), but it is hard to envisage practical alternatives.
 One suspects that Hugo Rifkind speaks for many in his view that the only arguments are theological or visceral: ‘Eeeeuw is no argument against gay marriage’, The Times, 9 March 2012.
 Hyde v Hyde (1865–9) L.R. 1 P. & D. 130. The laws of King Cnut (11th century) stated that no woman could be compelled to marry a man she did not like, or sold for money, and that no Christian man should ever have more wives than one. See R. H. Helmholz, The Oxford History of the Laws of England, Vol. I, OUP, 2004, p.46 and pp.44–51 for the impact of canon law on pre-Christian Anglo-Saxon practice more generally.
 Stephen Cretney, Family Law in the Twentieth Century, OUP, 2003, is the definitive account.
 A minister may refuse to marry certain couples who may otherwise marry by civil ceremony, including divorcees, persons related by a certain degree of affinity and post-operative transsexual people: Matrimonial Causes Act 1965, s. 8; Marriage Act 1949, ss. 5A and 5B.
 In Hall v Bull  2 All E.R. 1017, the Court of Appeal found that the criterion of marriage directly discriminated on grounds of sexual orientation. The case is on appeal to the Supreme Court, which may find indirect discrimination instead.
 Marriage Act 1949, s. 2. In England, Wales and Northern Ireland (but not Scotland) parties under 18 need parental, or equivalent, consent.
 Marriage Act 1949, s. 1.
 National Survey of Sexual Attitudes and Lifestyles II (April 2003): 5.3% men and 9.7% women reported that they were attracted at least once to the same sex, although more often to the opposite sex. Only 0.9% men and 0.2% women reported exclusive attraction to the same sex. The next National Survey is due in April 2013. One recent US study, which draws on several surveys, finds that 11% of adults acknowledge at least some same-sex attraction, 8.2% report having engaged in same-sex sexual behaviour, 1.8% identify as bisexual and 1.7% as lesbian or gay. See Gary J. Gates, ‘How many people are lesbian, gay, bisexual and transgender?’, The Williams Institute, UCLA, April 2011.
 Note that in law, ‘sexual orientation’ embraces the entire spectrum of self-awareness and lifestyle: see R (Amicus) v Secretary of State for Trade and Industry  I.C.R. 1176.
 Recent attempts to map the range of sexualities include the Klein Sexual Orientation Grid and the Sell Assessment of Sexual Orientation.
 The Prime Minister has also appealed to stability on a number of occasions, most notably his speech to the Conservative Party conference on 5 October 2011.
 Compare Marriage Act 1949, s. 44 with Civil Partnership Act 2004, s. 2. It would seem that mutual vows are optionally used in civil partnership ceremonies, although no religious service is to be used in a register office or other approved civil premises.
 5.5% of marriages entered into in 2005 had been dissolved by their fourth anniversary, as compared with only 2.5% civil partnerships. ONS, Civil Partnerships Five Years On (Population Trends Nr. 145, Autumn 2011). This may reflect a ‘backlog’ of stable longstanding informal partnerships. One major Scandinavian study suggests that over time same-sex civil partnerships have higher levels of instability: Andersson, Noack, Seierstad and Weedon-Fekjær, ‘The Demographics of Same-Sex Marriages in Norway and Sweden’, Demography 43.1, 2006, pp.79–98.
 Gender Recognition Act 2004, s. 4(3)–(4).
 260 gender recognition certificates were issued in the year 2010–2011. In only four cases was an interim certificate issued, because a marriage or civil partnership needed to be dissolved: Ministry of Justice, Gender Recognition Certificate Statistics Bulletin, 29 March 2012.
 Carol Gilligan, In a Different Voice, Harvard University Press, 1982, was a seminal work. See also Robin West, ‘Jurisprudence and Gender’, University of Chicago Law Rev. 55, 1988, pp.1–72.
 Many historic legal consequences of marriage did not reflect gender equality; but it is the core definition which is now at stake.
 Robert Putnam’s ‘Bowling Alone: America’s Declining Social Capital’ (1995) is now a classic study.
 Joint adoption for same-sex couples has been available in the UK since 2005; joint registration after fertility treatment since 2009: ONS, Civil Partnerships Five Years On, Population Trends Nr. 145, Autumn 2011, p.19.
 In 2011, 8.5% of civil partners and 4.8% of cohabiting same-sex couples had dependent children, amounting to 0.1% of all dependent children. ONS, Families and Households 2001 to 2011, January 2012, Table 1. Given the relatively recent introduction of adoption and assisted reproduction, these children are likely to be from previous relationships.
 Of numerous studies of the negative impact of divorce, see the older meta-analysis of 92 studies by Paul R. Amato and Bruce Keith: ‘Parental Divorce and the Well-being of Children’, Psychological Bulletin 110, 1991, pp.26–46. Judith S. Wallerstein and Julia Lewis, ‘The long-term impact of divorce on children’, Family Court Review, 36, 1998, pp.368–383, present a focused and qualitative study.
 Universal Declaration of Human Rights 1948, art. 1.
 See the American Psychological Association policy statement on Sexual Orientation, Parents and Children (2004): http://www.apa.org/about/policy/parenting.aspx (accessed 30 August 2012).
 A recent issue of Social Science Research contained a debate on this issue. Loren Marks challenged the reliability of the studies underlying the conventional wisdom: ‘Gay and Lesbian Parenting: the State of the Science’, Social Science Research, 41, 2012, pp.735–751, and Mark Regnerus presented the findings of the largest survey yet conducted: ‘[The survey] also clearly reveals that children appear most apt to succeed well as adults—on multiple counts and across a variety of domains—when they spend their entire childhood with their married mother and father, and especially when the parents remain married to the present day.’ ‘How different are the adult children of parents who have same-sex relationships?’, Social Science Research, 41, 2012, pp.752–770. Subsequent expert commentators accept that the research has at least put a question mark against the conventional wisdom, but caution generally against using social science research to solve constitutional questions around same-sex marriage.
 See Christiane Olivier, Jocasta’s Children: The Imprint of the Mother, Routledge, 1989, and at a more popular level, Robin Skynner and John Cleese, Families and How to Survive Them, Cedar Books, 1993, ch. 5.
 The right of the child to biological parenting is apparent from the UN Convention on the Rights of the Child (1990), articles 7–9, which contain rights to know and be cared for by both parents, to preserve family identity, not to be separated from parents against their will, as well as a right to maintain personal relations and direct contact even if the parents are separated. Article 3 of the UN Declaration on Social and Legal Principles relating to the Protection and Welfare of Children, with special reference to Foster Placement and Adoption Nationally and Internationally (GA Res. 41/85 of 3 Dec. 1986) states that the first priority for a child is to be cared for by his or her own (i.e. biological) parents.
 On this cultural development, see Anthony Giddens, The Transformation of Intimacy: Sexuality, Love and Eroticism in Modern Societies, Polity Press, 1993.
 Matthew Parris expresses this ideal with customary clarity: ‘The day that the battle for homosexual equality is won and over will be the day when a man, straight or gay, can boast that he chose.’, The Times, 21 April 2012.
 See, e.g., the arguments of Ronald Dworkin in Sovereign Virtue, Harvard University Press, 2000, ch. 14. Dworkin makes clear that he considers ‘prohibitions’ on same-sex marriage to breach each person’s foundational right to ethical independence: Justice for Hedgehogs, Harvard University Press, 2011, p.369.
 Around fifty, mostly Muslim majority, countries recognise polygamous marriages, either generally or for Muslims under their personal law.
 The growing construction of an ‘identity’ among the small British community of ‘polyamorous people’ can be seen at http://www.polyamory.org.uk/. The first goal of the (US) Polyamory Society includes ‘to support, defend and promote the social institution of polyamory’: http://www.polyamorysociety.org/mission.html (both accessed 30 August 2012).
 Philip L. Kilbride, Plural Marriage for our Times: a reinvented option?, Bergin & Garvey, 1994. See also the manifesto, ‘Beyond Same-Sex Marriage’ at http://www.beyondmarriage.org/full_statement.html (accessed on 17 September 2012).
 Jean Hannah Edelstein, ‘Why shouldn’t three people get married?’, The Guardian, 6 September 2012.
 Quite apart from the emotional impact, the cost of family breakdown to the public purse must be phenomenal. The obvious methodological caveats notwithstanding, The Relationships Foundation has recently calculated a figure of £41.7bn a year: Counting the Cost of Family Failure (2010).
 E.g. the nikah al-mut’ah of traditional Shi’ite Islamic law.
 ‘Mexico City considers fixed-term marriage licences’, The Guardian, 30 September 2011.
 Matrimonial Causes Act 1973, s. 12(a) and (b); s. 1(2)(a).
 Civil Partnership Act 2004, s. 44, only has ‘unreasonable behaviour’, separation and desertion as the facts establishing breakdown.
 The European Court of Human Rights upheld this restriction by a majority in Burden v United Kingdom (2008) 47 E.H.R.R. 38, but the logic of the dissenting judgements is impeccable.
 Sexual Offences Act 2003, ss. 64–65.
 Except, of course, a significant financial one. But that begs all the questions.
 On the political role of the Church of England in 1960s divorce reform, see Cretney, Family Law in the Twentieth Century, OUP, 2003, n. 7, ch. 9.
 From the substantial literature see Christopher Ash, Marriage: Sex in the Service of God, IVP, 2005, and Tim and Kathy Keller, The Meaning of Marriage, Hodder & Stoughton, 2011.
 And it is why the Government’s proposal also gives rise to more serious unresolved problems of conscientious objection than did civil partnership.
 Grace Davie draws on the works of French sociologists Danièle Hervieu-Léger and Maurice Halbwachs to analyse religion in terms of various forms of collective memory: see Religion in Modern Europe: A Memory Mutates, OUP, 2000.