Forthcoming in E. Brems and E. Desmet, Integrated Human Rights in Practice: Rewriting Human Rights Decisions (Edward Elgar, 2017). Revisiting Joslin v New Zealand: Same-Sex Marriage in Polarised Times Malcolm Langford* Abstract. In Joslin v. New Zealand (2002), the UN Human Rights Committee rejected the claim that marriage equality could be grounded in the International Covenant on Civil and Political Rights (ICCPR). Some scholars have argued that emerging state practice and a proper understanding of the drafting of the convention now justify the argument for marriage equality. Instead, this paper argues that a project of human rights integration reveals, paradoxically, additional legal challenges. Taking a departure point in relevant regional and national judgments, the chapter sets out a three-tiered cumulative argument for a right to same-sex marriage in Article 23(2) of the ICCPR. The crux of the argument is that the Committee will be able to recognise marriage equality when the ordinary meaning of ‘marriage’ becomes ambiguous as to the gender identity of spouses. 1. Introduction In Joslin v. New Zealand (2002), the UN Human Rights Committee was confronted with the question of same-sex marriage. Could marriage equality be grounded in the International Covenant on Civil and Political Rights?1 The Committee replied in the negative. It held that the Covenant recognized marriage as ‘only the union between a man and a woman wishing to marry each other’.2 In its familiar skeletal, oracular and somewhat cryptic style, the Committee reasoned that as Article 23 exclusively governed questions of marriage and the provision’s language was gendered (the right of ‘men and women’), the Covenant provided no space for a broader right to marriage equality. * Associate Professor, Faculty of Law, University of Oslo; Co-Director, Centre on Law and Social Transformation, CMI and University of Bergen; and Affiliate Researcher, Pluricourts Centre of Excellence, University of Oslo. Email: malcolm.langford@jus.uio.no This paper was supported by research grants on Sexual and Reproductive Rights Lawfare: Global Battles and PluriCourts - The Legitimate Roles of the Judiciary in the Global Order (both Research Council of Norway). The author thanks the editors, Geir Ulfstein, Claire Poppelwell-Scevak and Paul Johnson for their comments on an earlier version. 1 Joslin et al. v. New Zealand, Communication No. 902/1999, U.N. Doc. A/57/40 at 214 (2002). 2 Ibid. Para. 8.2. For progressive advocates, the decision was perhaps more of a disappointment than a surprise. Textually, the gendered nature of the provision appeared to provide a literalist roadblock to purposive forms of interpretation. Jurisprudentially, the Committee had scant legal practice upon which to build an affirmative answer. This can be contrasted with the Committee’s ground-breaking decision in Toonen v. Australia (1994), concerning the criminalization of same-sex relations. 3 The Committee was able to stand on the shoulders of the European Court of Human Rights’s well-established precedent of Dudgeon v. United Kingdom. 4 Finally, and politically, the Committee risked wading into the waters of a highly contested question, both globally and nationally. Faced at the time with a relatively modest or poor compliance record on its decisions, 5 exits by two states in protest at some decisions,6 and non-ratification of the Optional Protocol by a third of state parties,7 the Committee might have understandably not wished to risk its sociological legitimacy by staking out a strong course on this issue. In light of this background, the initial and muted scholarly response to Joslin – pointing out possible rather than actual flaws in the reasoning – was perhaps all that could be expected.8 However, much water has passed under the bridge since Joslin was decided 15 years ago. Only two years after the decision, came the first full judicial recognition of the right to same-sex marriage. In 2004, in Goodridge v. Dept. of Public Health, the Massachusetts Supreme Judicial Court found a denial of same-sex marriage unconstitutional.9 The prohibition violated the equal protection and due process clause of the state constitution as it failed to meet a rationality test: 3 Communication No. 488/1992, U.N. Doc CCPR/C/50/D/488/1992 (1994). Application no. 7525/76 (1981). 5 See C. Heyns and F. Viljoen, The Impact of the United Nations Human Rights Treaties on the Domestic Level, 23 HUM. RTS. Q. 483, 517–18 (2001). 6 In 23 January 1998, Jamaica denounced the Optional Protocol, followed by Trinidad and Tobago on 27 June 2000. Both states faced a large number of complaints on the use of the death penalty and were displeased with t the Committee’s approach. For a discussion, see R. Smith, Text and Materials on International Human Rights (London: Routledge, 2013), 90. 7 See underlying data at http://indicators.ohchr.org/ 8 See, e.g., Ignacio Saiz, 'Bracketing Sexuality: Human Rights and Sexual Orientation: A Decade of Development and Denial at the UN ' (2004) 7 Health and Human Rights 48, 54 and Michael O'Flaherty and John Fisher, 'Sexual Orientation, Gennder Identity and International Human Rights Law ' (2008) 8 Human Rights Law Review 207, 217. 9 Goodridge v. Dept. of Public Health, 798 N.E.2d 94; 440 Mass 309 (Mass. 2003) (‘Goodridge’). The first successful state-level decision on same-sex marriage was in Hawaii (1996) but was reversed in 1999: See Baehr v. Miike, No. 91-1394 First Circuit Court, Hawaii (1996) and Baehr v. Miike, No. 20371 (Supreme Court of Hawaii 1999-12-09). 4 2 The department [responsible for issue of marriage licenses] has offered purported justifications for the civil marriage restriction that are starkly at odds with the comprehensive network of vigorous, gender-neutral laws promoting stable families and the best interests of children. It has failed to identify any relevant characteristic that would justify shutting the door to civil marriage to a person who wishes to marry someone of the same sex.10 Notably, the Court finds that the denial of same-sex marriage fails to meet a rationality test. As this is the weakest standard of review in US constitutional law, this holding suggested that denial of marriage equality elsewhere might be equally vulnerable to judicial scrutiny. Since Goodridge, multiple national courts have ruled that same-sex marriage deserves constitutional protection. In 2005, the Canadian Supreme Court and the South African Constitutional Court both ruled that the exclusion of same-sex couples violated various constitutional rights;11 and a decade later similar decisions were made by the Constitutional Court of Colombia12 and the Supreme Court of the United States.13 The latter finding attracted enormous global attention in its wake. Contemporaneous with these judicially-driven developments, came legislative initiatives. Beginning with the Netherlands in 2001, same-sex marriage has been legislated as right in Belgium (2003), Spain (2005), Norway (2007), Sweden (2007), Portugal (2010), Iceland (2010), Argentina (2010), Denmark (2012), Caribbean Netherlands (2012), Brazil (2013), France (2013), Uruguay (2013), New Zealand (2013), England, Wales and Scotland (2014), Luxembourg (2015), Guam (2015), Ireland (2015), Greenland (2016), Isle of Man (2016) and Finland (2017). LGBT-rights are also regularly litigated at the international level. As Figure 1 shows, the number of cases filed with all international courts and quasi-judicial human rights bodies rose dramatically in the early 1990s and has exploded since 2010 (with approximately nine cases filed each per year). The bulk of this litigation has proceeded before the European Court of Human Rights but advocates have also approached a range of regional courts and commissions as well as UN human rights 10 Goodridge, p. 342. Reference Re Same-Sex Marriage [2004] 3 S.C.R. 698, 2004 SCC 79; Minister of Home Affairs and Another v Fourie and Another; Lesbian and Gay Equality Project and Others v Minister of Home Affairs and Others, [2005] ZACC 19. 12 28 April 2016. 13 Obergefell v. Hodges, 576 U.S. ___ (2015). 11 3 treaty bodies. While early cases concerned criminal legislation, age of consent claims and transsexual identity rights, the salient new issues are family rights, with questions over inheritance, adoption, and marriage. As Figure 2 shows, success rates have also risen dramatically. 4 FIGURE 1: INTERNATIONAL PETITIONS FOR LGBT-RIGHTS IN 5-YEAR INTERVALS14 50 40 30 20 Prog 10 Cons 0 FIGURE 2: PERCENTAGE OF SUCCESSFUL INTERNATIONAL LGBT-PETITIONS IN 5-YEAR INTERVALS15 70 60 50 40 30 20 10 0 While no international adjudicative body has yet recognised the right to same-sex marriage, this new context suggests Joslin should be reconsidered. After a decade and a half of jurisprudential advancement, legal reflection and political development, we can ask whether the case now represents bad law. Thus, is it time to revisit Joslin and argue that the ICCPR does recognize same-sex marriage? Some scholars believe the answer is a clear yes. In an extended analysis, Gerber, Tay and Sifris argued in 2012 that Joslin is ‘no longer good law’.16 Contesting the textual 14 Extracted from V. Vibe, M. Langford and T. Kirekbø, Litigating Queer Rights in the South, presented at Latin American Studies Association, June 2016. The data is based on our Sexual and Reproductive Rights Lawfare Database. The blue line is for progressive cases (seeking to advance LGBTI rights); the red line is for cases trying to restrict them. 15 Source: Ibid. Note that success rates apply to both sets of cases. 5 construction of the Committee and pointing to changing state practice and the right to equal treatment, they argue that the ‘right to marry should be interpreted in a nondiscriminatory manner’. They also predict that the Committee will soon align with this view, endorsing Joseph’s statement that it is only a question of time: ‘At some point I expect a global tipping point to be reached, where the HRC will decide a similar case differently.’17 I disagree. The legal reasoning in Joslin is certainly problematic but I am far from certain that the reasons offered by Gerber, Tay and Sifris should, or would, produce a different outcome. These authors fail particularly to consider the long-standing samesex marriage jurisprudence of the European Court of Human Rights 18 and they overstate the nature of global trends. Moreover, LGBTI rights now form a faultline in global and domestic ‘culture wars’. The Committee might be prudent to consider the consequences, including backlash, of a positive decision on same-sex marriage. As I will argue, the Joslin decision should be rewritten but only to provide a conditional right to same-sex marriage as well as provide the basis for advancing a range of other LGBTI rights on a global basis. This chapter sets out a three-tiered cumulative argument for a right to same-sex marriage in Article 23(2) of the ICCPR.19 First, I recall and reiterate that the provision is grammatically open to same-sex marriage and highlight the inferential nature of the Committee’s reasoning. Second, I argue that the Committee’s reference to the general understanding of the terms is susceptible to changes in state practice. Third, I claim that when there is no longer an international consensus on the definition of marriage (universal ambiguity), the Committee can draw on cross-cutting obligations (nondiscrimination) and legal sources (‘object and purpose’) to recognise a right to sameP. Gerber, K. Tay and A. Sifris, ’Marriage: A Human Right for All?’, 36 Sydney Law Review (2012), 642-667. 17 S. Joseph, Latest Case Law Trends: the International Covenant on Civil and Political Rights, 28 October 2013, Comment, available at https://castancentre.com/2013/10/28/latest-case-law-trends-theinternational-covenant-on-civil-and-political-rights/ 18 See sections 2 and 3. 19 In doing so, it is accepted, for now, that the Committee’s argument that Article 23(2) exclusively governs the question of the right to marry in the context of the ICCPR. It is of course possible and reasonable to argue that the free-standing provision on non-discrimination, Article 26, also covers the right to marry. However, this route does not seem particularly fruitful for two reasons. First, one is still left with an interpretive problem: how to interpret Article 26 in the ‘context’ of Article 23(2). The Vienna Convention on the Law of Treaties does not let one escape so easily. Second, the Committee and the European Court of Human Rights have been crystal clear on this point: see Joslin para. 8.2 and Schalk and Kopf v. Austria Application no. 30141/04 (ECHR), para. 54 (‘Schalk and Kopf’). There seems little to be gained in waving this flag. 16 6 sex marriage. The chapter concludes with an overall legal and political assessment and the rewritten judgment is annexed. Before commencing the analysis, a comment on theory and method is warranted. The over-arching ambition of this volume is to explore ‘human rights integration’ within a legal paradigm. The editors invite reflection on their view of the human rights project, an enterprise that sees human rights as ‘one and indivisible’ and pursues legal unity despite the fragmentation of human rights in multiple legal sources. My approach could be described as pluralistic and dialogical. By pluralistic, I mean a broad and multi-level approach to identifying relevant legal sources for re-imagining the content of a particular human rights decision. Thus, international, regional and national decisions and judgments should all be viewed as relevant legal sources together with soft law standards or general developments in state practice. In some cases, such sources might be directly relevant to interpretation, forming ‘subsequent practice’ or ‘relevant rules of international law’ for the purposes of Article 31 of the Vienna Convention on the Law of Treaties - and it is arguable that such the pace of such integration is slower than expected.20 However, pluralism goes beyond simply locating the relevant authoritative or persuasive precedents from another jurisdiction. Other jurisprudence and legal sources should be sought in order to seek inspiration and test the defensibility of a prevailing interpretive approach – even if it cannot be a formal legal source. The aim is to avoid the risks of parochialism by, as Adam Smith puts it, obtaining the ‘view from elsewhere’.21 However, this departure point is not a finishing point. The point of lifting one’s eyes up from a single jurisdiction is to put diverse legal sources into dialogue with each other. The aim is not necessarily to always unify legal ambition. This seems somewhat utopic and, in certain circumstances, possibly inappropriate. For a start, only a few human rights are designed to apply globally in the same manner. The prohibition against torture would be the most obvious candidate. Yet, even with this particular right, caveats apply. The relevant standards for the positive obligations 20 Compare, e.g., the optimism of M. Andenæs and E. Bjørge (ed.) A Farewell to Fragmentation Reassertion and Convergence in International Law (Cambridge University Press, 2015) with Malcolm Langford, 'The New Apologists: The International Court of Justice and Human Rights', Retfærd, 28 (2015), 49-78. 21 The notion was originally articulated by Adam Smith. It has been revived by Amartya Sen, The Idea of Justice (Cambridge M.A.: Harvard University Press, 2009) in the context of obtaining reasons from those living beyond our societies. 7 concerning torture (such as prevent, investigate and remedy) are relative and require contextual construction and application. Thus, convergence towards unity is less necessary as one moves from abstract rights articulation towards concrete rights application. Jurisprudential pluralism might therefore be a legitimate result; and one that does no violence to the idea of the universality of human rights. Moreover, a singular ambition of convergence could risk deflating the opportunities for experimentalism and learning. Human rights jurisprudence has often advanced through interdisciplinary borrowings 22 and local pressures and experiments. 23 Integration projects that simply seek to unify could risk dampening such experimentalism, resulting in ‘equalising down’ as much as ‘equalising up’. The result might be to drag down progress on human rights as much as advance it. It should be noted that this dialogical approach also requires the rigours of comparative method. The great risk in trawling jurisprudence and legal sources from ‘elsewhere’ is cherry-picking. Only sources that conform with one’s favoured understanding of rights (of whatever colour) might be sought after, selected and/or highlighted. Such an approach is, of course, legitimate if the relevant question is one of feasibility. For example, if one has a question concerning the mere justiciability of a right – e.g., the right to adequate housing or same-sex marriage – a legitimate response might be to simply point to ‘positive’ cases. However, if the question concerns the actual meaning of a right or other legal provision, comparative method becomes more demanding. Both ‘positive’ and ‘negative’ cases should be discussed, particularly if the negative cases contain substantive or applicable reasoning. As will be argued in this chapter, caselaw from other jurisdictions both advances and restricts the possibilities of advancing the right to marriage equality. M. Langford, ‘Interdisciplinarity and Multimethodism’, in B.A. Andreassen, H.O. Sano and S. McIerney-Lankford, Human Rights Research Methods, Edward Elgar, Cheltenham, 2017, Ch. 8. 23 A premier example of this is the evolutionary patchwork development of the right to education jurisprudence across fifty states in the USA between 1973 and the present. See: Michael Heise, 'State Constitutions, School Finance Litigation, and the Third Wave: From Equity to Adequacy, Heise, Michael ' (1995) 68 Temple Law Review 1151. At the global level, the creation of the proportionality and minimum core tests by the German Federal Constitutional Court have been adopted in various ways by international, regional and national bodies and debate on their appropriate refinement and development in light of their application continues. See S. Tsakyrakis, ‘Proportionality: An assault on human rights? A rejoinder to Madhav Khosla’ (2010) 8 International Journal of Constitutional Law 307; and K. Young, 'The Minimum Core of Economic and Social Rights: A Concept in Search of Content' (2008) 33 Yale Journal of International Law 113. 22 8 2. Linguistic construction Turning to Joslin, at the core of the Committee’s decision was the finding that the express and gendered language of the right excludes same-sex couplings.24 Article 23(2) reads as follows, ‘The right of men and women of marriageable age to marry and to found a family shall be recognized.’ Following Article 31(1) of the Vienna Convention on the Law of Treaties, the ‘ordinary meaning’ of the provision might seem rather straightforward and close off any other interpretation. Yet, on a straight-forward reading, this literal approach by the Committee is highly questionable. Men and women are referred to in the plural not the singular.25 Thus, there seems no literal barrier in Article 23 to a man marrying a man - both are included in the word ‘men’. This follows because the first mention of the word ‘and’ in Article 23(2) cannot be read as a strict conjunctive. This would entail multipolygamy: multiple men marrying multiple women. Put another way, Article 23(2) does not read that ‘the right of a man and woman to marry each other shall be recognized’. Such a closed formulation in the singular would clearly rule out samesex marriage. Indeed, in the analogous case of Schalk and Kopf v. Austria (2010), the European Court of Human Rights appears to concede this linguistic point: ‘The Court observes that, looked at in isolation, the wording of Article 12 [the equivalent provision in the European Convention of Human Rights] might be interpreted so as not to exclude the marriage between two men or two women.’26 However, this grammatical approach to marriage clauses has been balanced against a contextual reading. The ‘context’ of the treaty is drawn on in order to provide an 24 Para. 8.2. Some might read the European Court of Human Rights as finding that the French version is in the singular. In Schalk and Kopf, the Court noted that The French version provides that “l’homme et la femme ont le droit de se marier” (para. 54). However, the Court does not say this or discuss this and the genitive singular form is used, which can also be understood in the plural. For the ICCPR, there are five authentic languages. The Chinese version reflects the English (“men and women”) while the Spanish and Russian use the genitive singular (“the man and the woman”). In the event that this viewed as a conflict, Article 33 of the Vienna Convention on the Law of Treaties provides that “when a comparison of the authentic texts discloses a difference of meaning which the application of articles 31 and 32 does not remove, the meaning which best reconciles the texts, having regard to the object and purpose of the treaty, shall be adopted.” In this case, a good faith reading could suggest that the English and Chinese versions supply the correct meaning given the Spanish. French and Russian are ambiguous. Alternatively, a reading of the travaux préparatoires in accordance with Article 32 of the Vienna Convention (to be discussed in section 3) shows fairly clearly that the phrase was introduced to protect a women’s rights to marry. 26 Application no. 30141/04, para. 54 (‘Schalk and Kopf’). 25 9 alternative and unambiguous ordinary meaning. The common approach is to focus on the incidence of use of gendered language in the treaty as a whole. Article 23(2) is seemingly unique in the ICCPR for the use of gendered language. Thus, the Committee states that it ‘is the only substantive provision in the Covenant which defines a right by using the term “men and women”, rather than “every human being”, “everyone” and “all persons”.’27 It is at this point that Gerber, Tay and Sifris find fault with the Committee. They point out that the words ‘men and women’ are also contained in Article 3 of the ICCPR, which requires States Parties to ‘ensure the equal right of men and women’ to enjoy all Covenant rights. 28 Factually, these authors are correct. Yet, legally, I see no implication. Article 3 concerns inter-gender rather than intra-gender comparisons. Instead, the language of Article 3 rather supports the Committee’s interpretation rather than diminishes it and the scholars’ arguments become counter-productive. This line of attack is thus rather unfortunate. In my view, a better approach is to simply follow and emphasise the linguistic trail I outlined above. It reveals ambiguity. At the least, it does not totally prevent a different interpretation. This approach, outlined in para. 8.3 of the rewritten judgment, makes clear that an ordinary reading of Article 23(2) on its own lends support to a right to same-sex marriage: there is no literal barrier. Furthermore, it underlines that the Committee (and the European Court) avoided this implication by relying on an additional source of interpretation, namely context (other treaty provisions). While the Committee’s approach may be legitimate, it is not exclusive. If other sources of law can be mobilised, then Article 23(2) might be read differently. It is a provision waiting to be interpreted not simply found. 3. General understanding The UN Human Rights Committee does not rest, however, on this point. It adds a secondary argument concerning the ordinary meaning. It alleges that its interpretation is backed by evidence: 27 28 Joslin, para. 8.2 Gerber, Tay and Sifris (n 13 above), 646. 10 Use of the term "men and women", rather than the general terms used elsewhere in Part III of the Covenant, has been consistently and uniformly understood as indicating that the treaty obligation of States parties stemming from article 23, paragraph 2, of the Covenant is to recognize as marriage only the union between a man and a woman wishing to marry each other. Notably, the Committee provides no reference as for this statement. Despite peppering the rest of the decision with footnotes, the Committee is curiously silent on its most important empirical claim. Even more perplexing is its content. What does the Committee mean when it states ‘consistently and uniformly understood’? It does not identify who holds this understanding, what legal method is being employed, and when did this understanding begin. To say the sentence is cryptic would be an understatement. Gerber, Tay and Sifris engage with this declaratory statement by assuming that the Committee is referring to the travaux préparatoires. The respondent State party, New Zealand, had raised it in argument, pointing to references to ‘husband and wife’ during the drafting of the treaty. However, trawling the travaux préparatoires, Gerber, Tay and Sifris identify a greater level of ambiguity. They note and cite a number of comments by states which emphasized ‘equality’ rather than ‘gender’. A recent article by Johnson goes further and confirms their conclusions. 29 Johnson takes particular issue with the forceful statement of the European Court of Human Rights in Schalk and Kopf that: ‘The choice of wording in Article 12 must thus be regarded as deliberate’.30 After comprehensively searching the travaux préparatoires, he finds nothing to suggest that ‘men and women’ was intended to exclude the possibility of same-sex marriage. The phrase was introduced during the drafting of the Universal Declaration of Human Rights in order to protects the rights of women to freely marry. However, I am less sure that the Committee was referring to the travaux préparatoires. A closer examination of other judgments on same-sex marriage reveals that the Committee could be referring to the ‘quotidian’ or everyday meaning of the term marriage. In other words, the meaning at the time of the drafting of the 29 P. Johnson, ’”The choice of wording must be regarded as deliberate”: same-sex marriage and Article 12 of the European Convention on Human Rights’, European Law Review (2015) 207. 30 Schalk and Kopf, para. 55. 11 convention or the decision, or both. The Committee thus interprets the ‘ordinary’, in the criteria of ‘ordinary meaning’, to mean popular usage. This approach can be found in the Goodridge decision in 2004. The Massachusetts Supreme Judicial Court rejected the applicant’s definitional arguments, stating that the reliance on the ‘common-law and quotidian meaning’ of marriage was a perfectly reasonable starting point. 31 Likewise, the European Court of Human Rights in 2010 stated that, ‘Moreover, regard must be had to the historical context in which the Convention was adopted. In the 1950s marriage was clearly understood in the traditional sense of being a union between partners of different sex.’ Importantly, the Court includes this sentence after and separately from their claim that the choice of the terms ‘men and women’ in the Covenant was ‘deliberate’. While Johnson is correct to claim that his analysis of the travaux préparatoires eviscerates this claim (as the drafters of the European Convention consciously adopted the UN text), he does not address the additional idea that marriage had an accepted meaning at the time of the adoption of the European Convention. Alternatively, the Committee could be referring to some sort of treaty-related practice, whether its own or the practice of states. Article 31(3)(b) refers to ‘subsequent practice’ which ‘establishes the agreement of the parties’. Unlike the European Court of Human Rights, the Committee refers to the understanding of particular treaty terms rather than the general understanding of the term ‘marriage’. The phraseology conjures up an image of Committee members and states proceeding on some sort of working assumption that ‘men and women’ is interpreted in the singular. Divining the Committee’s meaning is clearly a challenge. Possibly the best approach is simply to proceed on the basis that the Committee (like others) is resting its interpretation of a generally accepted meaning of the term ‘marriage’ with some preference given to legal expressions of that meaning. Even if the identity of the relevant interpretive community is vague, it is the general understanding of the term that counts. Yet, in my view, it is precisely here that the next opening for same-sex marriage emerges. Can the meaning of these terms change? On its face, the relevant terms in 31 Goodridge, 319. 12 Article 23(2) would appear to be open to dynamic interpretation. As set out by the ICJ in Costa Rica v. Nicaragua: there are situations in which the parties’ intent upon conclusion of the treaty was, or may be presumed to have been, to give the terms used — or some of them — a meaning or content capable of evolving, not one fixed once and for all, so as to make allowance for, among other things, developments in international law. In such instances it is indeed in order to respect the parties’ common intention at the time the treaty was concluded, not to depart from it, that account should be taken of the meaning acquired by the terms in question upon each occasion on which the treaty is to be applied. 32 Importantly, the Court in Costa Rica v. Nicaragua affirms that dynamic interpretation is not contingent on a change in practice amongst state parties along the lines of Article 31(2)(b) of the Vienna Convention on the Law of Treaties – i.e., an express or tacit agreement to alter the original meaning. The Court expressly treats Article 31(2)(b) as a distinct and separate ground for a change in meaning. 33 Instead, as the Court clarified after providing an illustration, dynamic interpretation: is founded on the idea that, where the parties have used generic terms in a treaty, the parties necessarily having been aware that the meaning of the terms was likely to evolve over time, and where the treaty has been entered into for a very long period or is “of continuing duration”, the parties must be presumed, as a general rule, to have intended those terms to have an evolving meaning. In the present case, we must first ask whether ‘marriage’ is a ‘generic term’ which the parties ‘intended’ to have an ‘evolving meaning’. An examination of the travaux préparatoires would suggest that the idea of marriage was far from fixed. There was disagreement, for example, over who had the right to marry. States were particularly concerned with removing barriers that prevented marriages between certain groups of persons or placed restrictions on them. However, they differed over which restrictions were more valid. For instance, Egypt stated: In Egypt, as in almost all Moslem countries, certain restrictions and limitations existed regarding the marriage of Moslem women with persons belonging to another faith. Those limitations were of a religious character, sprung from the very spirit of the Moslem religion, and therefore could not be ignored. They did 32 Dispute regarding Navigational and Related Rights. (Costa Rica v. Nicaragua), Judgment, I.C.J. Reports 2009, p. 213, para. 64. 33 Ibid. Para. 64. 13 not, however, shock the universal conscience, as did, for instance, the restrictions based on nationality, race or colour, which existed in certain countries and which were not only condemned, but unknown in Egypt.34 Despite this disagreement, and the broader context in which struggles for marriage rights were occurring, the drafters opted for a very general right with an open nondiscrimination clause. It is hard not to conclude that the term is of a generic nature.35 The key question is whether the meaning of critical terms in Article 23(2) has evolved. Gerber, Tay and Sifris begin with a similar idea of a ‘living instrument’ and argue that the changing times allow the Human Rights Committee to revisit its understanding of Article 23(2)(b), particularly its construction of ‘men and women’. These authors point to instances where the Human Rights Committee has reversed its earlier understanding by referring to general legislative trends. For example, the Committee in LTK v Finland (1985) found that there was no right to conscientious objection in the ICCPR36 but retreated from this position in its later General Comment No. 22 (1993). There it stated: Many individuals have claimed the right to refuse to perform military service (conscientious objection) on the basis that such right derives from their freedoms under article 18. In response to such claims, a growing number of States have in their laws exempted from compulsory military service citizens who genuinely hold religious or other beliefs that forbid the performance of military service and replaced it with alternative national service. The Covenant does not explicitly refer to a right to conscientious objection, but the Committee believes that such a right can be derived from article 18, inasmuch as the obligation to use lethal 34 United Nations, General Assembly Hundred and Eighty-Third Plenary Meeting (December 10, 1948) A/PV.183, p.912, cited in Johnson, 7. 35 In this sense, it somewhat compared with the evolving meaning of “privacy” with which the European Court of Human Rights first used in finding that it protected individuals from criminalisation of same-sex relations. In Dudgeon v. United Kingdom, it stated, “As compared with the era when that legislation was enacted, there is now a better understanding, and in consequence an increased tolerance, of homosexual behaviour” (para. 60). In this respect, Gerber, Tay and Sifris (n 13 above), 648 also refer to General Comment No. 15 on the Right to Water of the Committee on Economic, Social and Cultural Rights (CESCR) as a good and relevant example of a dynamic approach. There the Committee noted that the drafters in the mid-20th Century had not contemplated the relevance of such a right as part of the open-textured ‘right to an adequate standing, including food, clothing and housing’ in Article 11 of the International Covenant on Economic, Social and Cultural Rights. However, as I argue below, General Comment No. 15 also drew on contemporary sources. 36 L. T. K. v. Finland, Communication No. 185/1984, Declared inadmissible 9 July 1985, U.N. Doc. CCPR/C/OP/2 at 61 (1990), para. 5.2. 14 force may seriously conflict with the freedom of conscience and the right to manifest one's religion or belief.37 In the case of same-sex marriage, Gerber, Tay and Sifris argue that Article 23 should be likewise ‘interpreted in a modern context’.38 Drawing on the Committee’s aboutturn on the right to conscientious objection, they paint ‘a global context’ in which a ‘growing number of states’ are recognising a right to marriage equality. While Gerber, Tay and Sifris are correct to point to the Committee’s prior practice and the evolution of state policy, they neglect two important points. The first is that the European Court of Human Rights has repeatedly considered this argument and reached a different conclusion. Notably, this is despite the greater acceptance of same-sex marriage in that region of the world. In Schalk and Kopf, the applicants had argued that ‘Article 12 should, in the light of present-day conditions, be read as granting same-sex couples access to marriage’.39 However, the Court responded that it was: not persuaded by the applicants’ argument. Although, as it noted in Christine Goodwin (cited above), the institution of marriage has undergone major social changes since the adoption of the Convention, the Court notes that there is no European consensus regarding same-sex marriage. At present no more than six out of forty-seven Convention States allow same-sex marriage (see paragraph 27 above).40 In Hämäläinen v Finland (2014)41 and Oliari and Others v. Italy (2015),42 the Court confirmed this argument. Despite the number of European states recognizing samesex marriage growing to thirteen, the Court remained resolute that the European consensus on the definition of marriage in Article 12 had not shifted.43 Indeed, some have argued that there can only be a right to same-sex marriage in the ECHR when a 37 Human Rights Committee, General Comment 22, Article 18 (Forty-eighth session, 1993). Compilation of General Comments and General Recommendations Adopted by Human Rights Treaty Bodies, U.N. Doc. HRI/GEN/1/Rev.1 at 35 (1994). 38 Gerber, Tay and Sifris (n 13 above), p. 648. 39 Schalk and Kopf, para. 57. 40 Ibid. para. 58. 41 Hämäläinen v Finland [GC] Application no 37359/09, 16 July 2014. 42 Oliari and Others v. Italy, Applications Nos. 18766/11 & 36030/11 (July 21, 2015) (‘Oliari’). 43 The furthest the Court is prepared to go is same-sex unions, at least in the case of Italy. In Oliari, it held that in the absence of marriage for same-sex couples in Italy, the right to family life obliged the state to offer other options, such as civil unions or registered partnerships. 15 majority of 24 states change their laws. 44 If we apply the Court’s approach to the global level, the argument for using changing legislation becomes thinner. Indeed, Gerber, Tay and Sifirs concede elsewhere in their article that there is no global consensus and that the number of states is small and dominated by Western states.45 The second challenge is that the description of the Committee’s approach in conscientious objection needs greater context. The Committee’s approach to state practice is not so straight-forward as presented. First, the Committee in its General Comment No. 22 in 1993 only recognised the possibility that the right could be found in the ICCPR: ‘believes that such a right can be derived from article 18’.46 It took another fourteen years, in the case of Yoon and Choi v. South Korea, 47 until the Committee felt brave enough to declare that the right was actually recognised in the Covenant. Indeed, in her dissent, Wedgewood makes clear that the Committee in its case law and concluding observations had never stated that the derivation of the right was ‘required by the Covenant’.48 Second, the reasoning of the Committee in its General Comment No. 22 has been subject to methodological criticism. Peters argues a practice amongst a minority of states was insufficient to create a consensus and there was likewise a lack of consensus amongst the international community in various standards.49 She notes that: In Yoon-Choi the HRC neither undertook any greater substantive assessment of the practice of the states parties to the ICCPR, nor assessed the relevant international practice as a basis for concluding the general acceptance and consensus of the parties on the right to conscientious objection. Do these critical points imply that the Committee should follow the European Court of Human Rights and expect a universal consensus, for example a majority in national state practice or a clearly expressed consensus at the international law? In my view, 44 C. Poppelwell-Scevak, The European Court of Human Rights and Same-sex Marriage: The Consensus Approach, LL.M Thesis, Faculty of Law, University of Oslo, 15 May 2016. 45 Gerber, Tay and Sifris (n 13 above), 644. 46 Emphasis added. 47 Choi and Yoon v. Republic of Korea, Communication Nos. 1321/2004 and 1322/2004, 23 January 2007, UN Docs.CCPR/C/88/D/1321–1322/2004. 48 Choi and Yoon v. Republic of Korea, Communication Nos. 1321/2004 and 1322/2004, 23 January 2007, UN Docs.CCPR/C/88/D/1321–1322/2004, Dissenting Opinion by Committee member Ms Ruth Wedgewood, 2. 49 See discussion in B. Peters, ‘Aspects of Human Rights Interpretation by the UN Treaty Bodies’, in H. Keller and G. Ulfstein (eds.), UN Human Rights Treaty Bodies, Law and Legitmacy (Cambridge University Press, 2012), 261-319. 16 the answer is no. This approach is too simplistic in interpreting Article 23(2) as it ignores the other grounds for justifying same-sex marriage. Following the proposed cumulative approach, the relevant question is whether the meaning of marriage has become ambiguous amongst states parties or at the international level. If it remains heteronormative – a man and a woman – then the analysis stops. If the meaning has become unclear, then we must rely on other methods of interpretation to provide a meaning. As the ordinary meaning is also unclear (see section 2), we must turn to other legal sources (object and purpose). However, before doing so it is important to ask whether we have reached that stage. Is the understanding of marriage now ambiguous? In my opinion, no. The current general understanding still falls on the side of heteronormativity – as is set out in para. 8.4 of the rewritten judgment. Indeed, it is not changing particularly fast. Same-sex marriage rights have been recognised in almost all of North America (partial in Mexico), most of Western Europe and a significant portion of South America. However, besides South Africa in Africa and New Zealand in the Pacific, same-sex marriage has not been recognised in the rest of the world, namely Eastern Europe, Central America, Africa, Middle East, Asia and the Pacific. Yet, the key test should be not a requirement of a simple majority of states – that seems artificial. Rather the question is whether the term of marriage has a settled meaning at the global level. This leads us to ask what would define a state of ambiguity? To that, with irony noted, there is no clear-cut answer. Ambiguity might arise with only a minority of states or require a clear majority of states. It depends on the character of that minority or majority. The litmus test is global meaning. For example, a good indicator of rising ambiguity might be found in the content of international declarations.50 If terminology around marriage becomes less clear - e.g. regular references to ‘spouses’ by states when discussing the right to marry or noninter-gendered uses of the terms ‘men and women’ – or various marriage-related rights of same-sex couples become increasingly recognised, we could argue that a state of ambiguity has emerged. 50 In General Comment No. 15, the CESCR refers to international standards that demonstrated that the right to water could be legally derived and that on two instances the international community had recognised a right to water. These texts were not primarily used as evidence of consensus but rather as important interpretive tools. See discussion in Malcolm Langford, 'Ambition that overleaps itself? A Response to Stephen Tully’s ‘Critique’ of the General Comment on the Right to Water' (2006) 26 Netherlands Quarterly of Human Rights 433. 17 A recent decision by the General Assembly is notable in this regard. It not only erodes the traditional understanding of marriage, but points towards recognition of same-sex marriage – which are two sides of the same issue, and thus suggests evolutive interpretation. By a majority, the UN General Assembly voted to deny a challenge to a UN-Secretary-General administrative directive on spousal entitlements. It would extend benefits to UN employees that are in legally-recognized same-sex unions, including those employees from states where same-sex marriage was not legal.51 The Russian Federation challenged this occlusion of consideration of national law but its resolution only attracted 43 votes (with 80 against and 37 abstentions). On one hand, this decision is indicative of a changing international understanding of ‘marriage’. The Secretary-General’ staff defended the directive on the basis of the ‘Organization’s values’.52 On the other hand, a number of abstaining and supportive states defended the directive the basis that the UN Secretary-General had administrative discretion in this area. For example, in voting against the Russian challenge, the representative of Mexico noted that the directive ‘promoted the elimination of discrimination’ but also stated that the UN directive: (i) ‘did not encroach upon the competencies of national laws regulating those types of acts’ and (ii) ‘did not establish an international standard for marriages, domestic partnerships, or similar types of unions.’53 4. Substantive interpretation If the meaning of ‘men and women’ or ‘marriage’ is ambiguous, then the Committee would have greater freedom to recognise a right to same-sex marriage. The Committee could proceed in two mutually supportive directions. The first is to argue that the object and purpose of the provision is to ensure that all couples are able to enjoy the right to marry, regardless of their gender identity or registered gender. Indeed, Article 23(1) begins with the overarching recognition that ‘the family is the natural and fundamental group unit of society and is entitled to protection by society and the State’. If same-sex couples are families, a finding the European Court of 51 GA Fifth Committee, 69th Session, 33rd Meeting (AM), 24 March 2015, UN. Doc, GA/AB/4150. The UN directive is ST/SGB/2004/13/Rev.1. 52 Budget Committee Rejects Draft Decision to Withdraw Current Personal Status Rules for Determining Staff Benefits and Entitlements, available at https://www.un.org/press/en/2015/gaab4150.doc.htm 53 Ibid. 18 Human Rights made in Schalk and Kopf,54 then it would follow that marriage rights should ensue to all couples in order to realise the overriding objective of Article 23. The majority decision of the US Supreme Court in Obergefell nicely captures many of the positive reasons why the principles fundamental to affirming a right to oppositesex marriage (long-established in its jurisprudence) apply equally to same-sex couples. For example, Justice Kennedy notes that the ‘right to personal choice regarding marriage is inherent in the concept of individual autonomy.’ 55 Indeed, he echoes the language of the ICCPR when he further reasons, ‘it would be contradictory "to recognize a right of privacy with respect to other matters of family life and not with respect to the decision to enter the relationship that is the foundation of the family in our society”’.56 The second is that the right to marry is fundamental because it supports a ‘two-person union unlike any other in its importance to the committed individuals’.57 The third is that it ‘safeguards children and families and thus draws meaning from related rights of childrearing, procreation, and education’ although Kennedy is quick to note that ‘the right to marry is not less meaningful for those who do not or cannot have children’.58 The second approach is to apply the non-discrimination provision in Article 2(1) to Article 23(2). This method has been central in all national court decisions affirming the right to marriage equality. Moreover, in the concurring individual opinions of Lallah and Scheinin in Joslin, these two members noted the ongoing relevance of non-discrimination to other LGBTI-related rights and the well-established and relevant test of reasonable and objective justification for any distinctions based on protected characteristics: Contrary to what was asserted by the State party (para. 4.12), it is the established view of the Committee that the prohibition against discrimination on grounds of ‘sex’ in article 26 comprises also discrimination based on sexual orientation. And when the Committee has held that certain differences in the treatment of ‘In view of this evolution, the Court considers it artificial to maintain the view that, in contrast to a different-sex couple, a same-sex couple cannot enjoy “family life” for the purposes of Article 8. Consequently, the relationship of the applicants, a cohabiting same-sex couple living in a stable de facto partnership, falls within the notion of “family life”, just as the relationship of a different-sex couple in the same situation would.’: Schalk and Kopf, para. 94. 55 Obergefell v. Hodges, Justice Kennedy, 12. 56 Ibid, quoting from Zablocki v. Redhail, 434 U. S. 374, 384 (1978), at 386. Emphasis added. 57 Ibid. 14. 58 Ibid. 54 19 married couples and unmarried heterosexual couples were based on reasonable and objective criteria and hence not discriminatory, the rationale of this approach was in the ability of the couples in question to choose whether to marry or not to marry, with all the entailing consequences. No such possibility of choice exists for same-sex couples in countries where the law does not allow for same-sex marriage or other type of recognized same-sex partnership with consequences similar to or identical with those of marriage. Therefore, a denial of certain rights or benefits to same-sex couples that are available to married couples may amount to discrimination prohibited under article 26, unless otherwise justified on reasonable and objective criteria. If the meaning of marriage in Article 23(2) is ambiguous, then one can consider whether any national prohibition on same-sex marriage or failure to recognise it in law and practice can be justified on reasonable and objective criteria. After more than a decade of litigation on this question, it is difficult to see how heteronormative approaches of marriage could survive this test. Multiple judgments have found that prohibitions of same-sex marriage do not even meet a rationality test let alone a reasonableness test. This is because the various objectives or aims associated with opposite-sex marriage do not make logical or basic evidential sense. Indeed, it is appropriate to compare the scrupulous 50-plus page empirical evaluation by the district trial judge in Perry v. Schwarznegger (2010) of the claims that same-sex marriage would have harmful effects on society and the curt approach of the Supreme Court five years later. The majority dismisses such arguments in a mere paragraph: The respondents also argue allowing same-sex couples to wed will harm marriage as an institution by leading to fewer opposite-sex marriages. This may occur, the respondents contend, because licensing same-sex marriage severs the connection between natural procreation and marriage. That argument, however, rests on a counterintuitive view of opposite-sex couple’s decisionmaking processes regarding marriage and parenthood. Decisions about whether to marry and raise children are based on many personal, romantic, and practical considerations; and it is unrealistic to conclude that an opposite-sex couple would choose not to marry simply because same-sex couples may do so…. The respondents have not shown a foundation for the conclusion that allowing samesex marriage will cause the harmful outcomes they describe. Indeed, with respect to this asserted basis for excluding same-sex couples from the right to marry, it is appropriate to observe these cases involve only the rights of two consenting 20 adults whose marriages would pose no risk of harm to themselves or third parties.59 Given these types of findings, it is difficult to imagine how the Committee would find a prohibition on same-sex marriage to be ‘reasonable and objective’. It would take a large dose of adjudicative creativity and state deferentialism. 5. Conclusion This chapter has argued that Joslin v New Zealand might be bad law under certain conditions. Following a three-step argument, it was first asserted that a treaty-centric reading of Article 23(2) results in ambiguous meaning. On its face, the provision is clearly open to a right to same-sex marriage but a contextual interpretation in light of other provisions in the treaty points in the other direction. Secondly, it was argued that the Committee’s reliance on the general and traditional understanding of marriage might be legitimate but that the relevant terms were open to dynamic interpretation. If these terms become globally ambiguous (through significant changes in state practice or a shift in the nature of statements by the international community), then one can turn to other legal sources and methods. However, it was argued that this step has not yet been fulfilled. Third, it was argued that the introduction of new legal sources and methods (such as a purposive approach and/or the obligation of non-discrimination) would clearly favour a right to same-sex marriage. Pertinently, it would be difficult to find reasonable and objective criteria for denial of the right in light of existing national jurisprudence. Of course, this conclusion can be attacked as too progressive and too conservative. In these concluding remarks, I will focus on the potential latter critique. Even if a more progressive approach might be legally justifiable, it is unlikely to be politically and practically advisable. As foreshadowed, LGBTI rights currently represent a dividing line in global politics. On the one hand, support for these rights, particularly decriminalisation, has become arguably a new ‘civilisation norm’ for many (Western) states and even international institutions – witness the rare foray of the World Bank into human rights with its suspension of loans to Uganda for its enhanced criminalisation of same-sex relations. On the other hand, many states in the world 59 Obergefell v. Hodges, Justice Kennedy, 26. 21 have embraced repressive measures against LGTBI persons as a fundamental cornerstone of their expressed desire to protect community values and identities. The potential for backlash from these states against a pro-same-sex marriage decision could be significant and the Committee would be unlikely to get much support from major Western states (such as the USA, United Kingdom, and France) who tend to be critical to any signs of excessive adjudicative interpretive activism by international bodies. More importantly, the Committee has significant work to do in simply enforcing its existing decisions on LGBTI rights. For a start, there are 31 African, Asian and Latin American countries which both criminalise homosexuality and are subject to the Committee’s complaints procedure. However, no individual or actor in these countries has brought a case challenging these laws. Preliminary investigation of this phenomenon suggests that domestic advocates are concerned about their internal legitimacy and the effectiveness of any decision from the Committee.60 Moreover, the Committee itself has been very tardy in questioning states over LGBTI rights in their regular reporting. The shift does not begin until the early 2000s despite the issuance of the Toonen decision in 1994.61 This trend is also confirmed in a recent study by Gerber and Gory (2014) of the period 2003-2013. They find a marked increase in comments and questions by Committee members about LGBTI rights (for example four times in 2004 against twelve times in 2013.) Beyond preparing the legal ground for a future decision on same-sex marriage, the Committee and advocates might be well-advised to see how the Committee’s jurisprudence and effectiveness on LGBTI rights can be first solidified. M. Langford, V. Vibe and T. Kirkebø, ‘International Legal Mobilisation: The Rise of Sexual and Reproductive Rights’, paper presented at Law and Society Association Annual Conference, Seattle, 1-5 June. 61 M. Langford and C. Creamer, The Toonen Decision: Domestic and International Impact, European Society of International Law Interest Group on Human Rights Seminar, Oslo, 3 September. 60 22 Annex. Rewritten Decision Ms. Juliet Joslin et al. v. New Zealand, Communication No. 902/1999, U.N. Doc. A/57/40 at 214 (2002). Consideration of the merits 8.1 The Human Rights Committee has considered the present communication in the light of all the information made available to it by the parties, as provided in article 5, paragraph 1 of the Optional Protocol. 8.2 The authors' essential claim is that the Covenant obligates States parties to confer upon homosexual couples the capacity to marry and that by denying the authors this capacity the State party violates their rights under articles 16, 17, 23, paragraphs 1 and 2, and 26 of the Covenant. The Committee notes that article 23, paragraph 2, of the Covenant expressly addresses the issue of the right to marry. Given the existence of a specific provision in the Covenant on the right to marriage, any claim that this right has been violated must be considered in the light of this provision. 8.3 Examined in isolation, the wording of Article 23(2) can be interpreted so as not to exclude the marriage between two men or between two women.62 However, Article 23, paragraph 2, of the Covenant is the only substantive provision in the Covenant which defines a right by using the term "men and women", rather than "every human being", "everyone" and "all persons". This treaty context suggests that the specific use of gendered language in Article 23(2) could be regarded as ‘deliberate’63 and that the reference is to a union of ‘a man’ with ‘a woman’. However, a review of the travaux préparatoires suggest that the phrase ‘men and women’ was included in order to protect the rights of women.64 The question of same-sex marriage was not discussed during the drafting. Moreover, on its face, the language could also refer to the gendered nature of intimate relations which might be opposite-sex or same-sex. 8.4. The broader context of the understanding of the relevant terms must therefore be considered. Use of the term "men and women", rather than the general terms used elsewhere in Part III of the Covenant, has been consistently and uniformly understood as indicating that the treaty obligation of States parties stemming from article 23, paragraph 2, of the Covenant is to recognize as marriage only the union between a man and a woman wishing to marry each other. Moreover, regard must be had to the historical context in which the Convention was adopted. Marriage was clearly understood then in the ‘traditional sense of being a union between partners of different sex’.65 However, the meaning of both ‘men and women’ and ‘marriage’ is not fixed. As the International Court of Justice has stated, ‘there are situations in which the parties’ intent upon conclusion of the treaty was, or may be presumed to have been, to give the terms used — or some of them — a meaning or content capable of evolving, not one fixed once and for all, so as to make allowance for, among other things, developments in 62 Schalk and Kopf’ v. Austria, Application no. 30141/04 (2010), European Court of Human Rights, para. 54. 63 Schalk and Kopf, para. 54. 64 See overview in 64 P. Johnson, ’”The choice of wording must be regarded as deliberate”: same-sex marriage and Article 12 of the European Convention on Human Rights’, European Law Review (2015) 207. 65 Ibid. para. 54. 23 international law’. 66 If developments in state practice or international law led to a situation where the meaning of the term was, at least, ambiguous then the Committee would arguably have the freedom to consider whether the denial of same-sex marriage was in violation of Article 23(2). The Committee would be free to consider whether such a denial was consistent with the object and purpose of the provision as well as whether it was consistent with the obligations of states not to discriminate on the basis of sexual orientation and the state can justify such exclusion by reference to reasonable and objective criteria. Referring to the Committee’s treatment of the right to conscientious objection in General Comment No. 22(1993), the applicants submitted they need only show that ‘a growing number of states’ have recognised the right to same-sex marriage.67 However, the Committee recalls that in General Comment No. 22, the changing state practice only prompted the Committee to remark that the right to conscientious objection ‘can be derived’ from the Covenant. 68 It was only in 2007 that the Committee affirmed the recognition of this right.69 In the present case, we note that only a small minority of states have changed their laws to recognise same-sex marriage (a total of 24) and there has been no recognition of same-sex marriage by states in international forums. Nonetheless, a continuance of this trend might render the meaning of ‘men and women’ or ‘marriage’ in Article 23(2) ambiguous in the future. 8.5. The Committee also reminds that it is the established view of the Committee that the prohibition against discrimination on grounds of ‘sex’ in article 26 comprises discrimination based on sexual orientation. And when the Committee has held that certain differences in the treatment of married couples and unmarried heterosexual couples were based on reasonable and objective criteria and hence not discriminatory, the rationale of this approach was in the ability of the couples in question to choose whether to marry or not to marry, with all the entailing consequences. No such possibility of choice exists for same-sex couples in countries where the law does not allow for same-sex marriage or other type of recognized same-sex partnership with consequences similar to or identical with those of marriage. Therefore, a denial of certain rights or benefits to same-sex couples that are available to married couples may amount to discrimination prohibited under article 26, unless otherwise justified on reasonable and objective criteria. 70 The obligation of non-discrimination (Article 2(1)) also applies to Article 23, including sub-paragraph 1, which states ‘The family is the natural and fundamental group unit of society and is entitled to protection by society and the State.’ It may be arguable that the meaning of the term ‘family’ has evolved to included same-sex couples.71 In this respect, we note that the UN General Assembly voted by a majority to deny a challenge 66 Dispute regarding Navigational and Related Rights. (Costa Rica v. Nicaragua), Judgment, I.C.J. Reports 2009, p. 213, para. 64. 67 Human Rights Committee, General Comment 22, Article 18 (Forty-eighth session, 1993). Compilation of General Comments and General Recommendations Adopted by Human Rights Treaty Bodies, U.N. Doc. HRI/GEN/1/Rev.1 at 35 (1994). 68 Ibid. Para. 11. ’The Covenant does not explicitly refer to a right to conscientious objection, but the Committee believes that such a right can be derived from article 18’. 69 Choi and Yoon v. Republic of Korea, Communication Nos. 1321/2004 and 1322/2004, 23 January 2007, UN Docs.CCPR/C/88/D/1321–1322/2004. 70 *Note: this paragraph is taken from the original individual concurring opinion of Lallah and Scheinin in Joslin v New Zealand (2002). 71 See Schalk and Kopf, para. 94. ‘In view of this evolution, the Court considers it artificial to maintain the view that, in contrast to a different-sex couple, a same-sex couple cannot enjoy “family life” for the purposes of Article 8. Consequently, the relationship of the applicants, a cohabiting same-sex couple living in a stable de facto partnership, falls within the notion of “family life”, just as the relationship of a different-sex couple in the same situation would.’ 24 to the UN-Secretary-General’s administrative directive to extend entitlement benefits to UN employees that are in legally-recognized same-sex unions, not just those from states where same-sex marriages are legal.72 8.6 [Previous 8.3] In light of the scope of the right to marry under article 23, paragraph 2, of the Covenant, the Committee cannot find that by mere refusal to provide for marriage between homosexual couples, the State party has violated the rights of the authors under articles 16, 17, 23, paragraphs 1 and 2, or 26 of the Covenant. 9. The Human Rights Committee, acting under article 5, paragraph 4, of the Optional Protocol, is of the view that the facts before it do not disclose a violation of any provision of the International Covenant on Civil and Political Rights. 72 GA Fifth Committee, 69th Session, 33rd Meeting (AM), 24 March 2015, UN. Doc, GA/AB/4150. The UN directive is ST/SGB/2004/13/Rev.1. 25