Enhancing LGBTI Rights by Changing the Interpretation of the Convention on the Elimination of All Forms of Discrimination Against Women?

ABSTRACT At first sight, it seems clear that the Convention on the Elimination of All Forms of Discrimination against Women (CEDAW) aims to eliminate discrimination against (only) women. However, a legal analysis of the object and purpose of the Convention reveals that CEDAW, in particular article 5a (which requires modification of ‘cultural patterns of conduct’), could be instrumental in addressing at least some aspects of LGBTI rights. Nevertheless, an analysis of Concluding Observations adopted by the Committee since 2010 reveals that the Committee, the body entrusted both with monitoring the implementation of the Convention and acting as its principal interpreter, does not yet use the possibility offered under article 5a to interpret the Convention in such a way. This article argues that the most realistic (albeit slow) way to enhance LGBTI rights within the framework of the CEDAW Convention seems to be to encourage state parties and non-governmental organisations to include discussions of discrimination against LGBTI persons in their Country Reports and Shadow Reports to the Committee, thereby inviting the Committee to reflect on LGBTI discrimination.


I. Introduction
In contrast to other non-discrimination grounds like sex, race/ethnicity, and disability, there is no one particular international legal instrument that expressly and explicitly obliges state parties to respect the human rights of lesbian, gay, bisexual, transgender, and intersex (LGBTI) persons and prohibits all forms of discrimination against these persons. 1 How then can the enhancement of the human rights of LGBTI persons be advanced, and, in particular, their rights to sexual freedom and gender identity? This question was at the heart of the discussions at the conference held in Oslo in December 2014. 2 The right to sexual freedom and gender identity are of crucial importance for each and every human being, regardless of his or her sexual orientation or gender identity, but are most fiercely contested when it concerns LGBTI persons. 3 Sexual freedom refers to the freedom of people to choose with which other person(s) they want to establish relationships of a sexual nature and whether or not they can or should formalise these relations in any legal construction provided for by the state. 4 Gender identity, on the other hand, refers to the freedom of people to express the way they experience a particular gendered identity. 5 Gendered, in this context, means that the sense of self (identity) is co-determined (among many other factors) by the biological or genetic 'sex' features that persons have, in interaction with the socially and culturally constructed meanings and normative regulations of sex and sexuality that are attached to this 'sex'. Sex, sexuality, and gender, therefore, come together in the notion of gender identity. The question we address in our contribution to this special issue is whether the UN Convention on the Elimination of All Forms of Discrimination against Women (CEDAW) can in some way be instrumental to the goal of enhancing the rights to sexual freedom and gender identity of LGBTI persons. We shall approach this question primarily from a positivist legal perspective. Section II contains a legal analysis of the object and purpose of the Convention in so far as it relates to the issue of LGBTI rights. Wr argue that, in particular, article 5a, which requires modification of 'cultural patterns of conduct', could in theory be instrumental in addressing at least some aspects of LGBTI rights. 6 But, as will be shown in section III on the basis of a systematic analysis of Concluding Observations communicated since 2010, while the CEDAW Committee gives considerable attention to LGBTI issues, article 5a has thus far not been the road taken by the Committee in its authoritative interpretations of the Convention to expand the interpretation therein. Although some people have set high hopes on the CEDAW as a possible way of enhancing the human rights of LGBTI %20Gender%20Identity%20Rights%20within%20the%20United%20Nations%20Framework%20and%20System%20Past %2C%20Present%20and%20Future.pdf. 2 University of Oslo, 'Sexual Freedom and the Right to Gender Identity as a Site of Legal and Political Struggle', Oslo, House of Literature, 11-12 December 2014. 3 Sexual freedom and gender identity are not catalogued in any of the existing international human rights treaties and therefore formally do not exist as explicitly guaranteed and established human rights. Rather, as the title of the Oslo Conference indicates, these (normative) concepts are sites of legal and political debate, over their very existence as rights, over their meaning, as well as over their consequences. 4 In so far as this freedom is recognised as a right, it is immediately restricted to being only applicable to adult human beings who (can) give their free and conscious consent to having particular sexual relationships. Pedophilia, bestiality, sexual assault, and rape are not deemed to be protected by this (presumed) freedom to have sexual relationships of one's own choice. It is also contested whether persons have the freedom to choose to engage in sexual relationships of a purely commercial or economic nature, either as a provider, as a customer, or both. 5 The right to gender identity, therefore, is not only about something that people experience 'internally', but in fact is most contested when people express their gender identity, in particular, when this identity deviates from what are perceived as 'normal' expressions along the bipolar axes of the 'male' and 'female' sexes and of heterosexuality. 6 In this article, we have chosen to focus on the potential of article 5a CEDAW. Other substantive articles in this Convention could (and should) also be investigated as regards their potential to contribute to the enhancement of the human rights of LGBTI persons. The same applies for the potential of other international and regional human rights instruments, eg the UN ICCPR and the European Convention on Human Rights, which include substantive rights, like the right to marriage and the right to private life that have great relevance for LGTI persons. However, time and space are too limited to include all such possibilities for progressive interpretation and implementation of these instruments in this article.
people, we are therefore not particularly optimistic about the Convention's actual contribution in this respect. The question raised at the Conference in Oslo also calls for a theoretical and strategic analysis of the pros and cons, or the opportunities and obstacles, of using this Convention to enhance LGBTI (human) rights. 7 In the concluding section IV, we briefly raise some of the most important concerns regarding these problems. We argue that the most realistic (albeit slow) way forward seems to be to urge state parties and NGOs to include discussions of discrimination against LGBTI persons in Country Reports and Shadow Reports to the Committee and thus invite the Committee to reflect on LGBTI discrimination.

II. The Object and Purpose of the CEDAW Convention: A Holistic Approach to Equality 8
In order to answer the question of whether CEDAW can be instrumental in enhancing LGBTI rights, it is necessary to consider the object and purpose of the Convention. At first sight it seems clear that the Convention aims to eliminate discrimination against women, ie discrimination on the ground of the female sex (and not sexuality or gender identity). The object and purpose of the Convention are, however, less clear-cut and more contested than may be expected on the basis of its name. In this section, we shall consider whether there are any theoretical pegs on which to hang claims that move beyond that understanding of the Convention as exclusively aimed at women's equality. We first discuss the main object of the Convention, which is to provide international legal standards to combat all forms of discrimination against women. Subsequently, the Convention's definition of discrimination against women is considered. Thirdly, we shall analyse the three main objectives of CEDAW and assess whether there isat least in theoryspace for including LGBTI rights within the scope of this Convention.

II.i. For women only? CEDAW's asymmetrical approach to equality
The preamble of CEDAW expressly recognises that, despite the international obligations to put an end to discrimination between men and women that were laid down in the previously adopted human rights treaties, 9 extensive discrimination against women continues to exist. The drafters recognised that in reality it is predominantly women who suffer from sex discrimination and that therefore a Convention specifically aiming at the elimination of this discrimination was necessary. 10 In the view of some commentators, CEDAW's asymmetrical nature is seen as a great advantage over sex-equality laws that are symmetrical, ie those which protect both sexes equally from differential treatment. 11 In the words of Brown, referring to McKinnon: 'the more gender-neutral or gender-blind a particular 7 These issues are also addressed in Dianne Otto's contribution to this issue. 8  right (or any law or public policy) is, the more likely it is to enhance the privilege of men and eclipse the needs of women as subordinates.' 12 This is also applicable to the right to equality: 'The person who has the power of definition, who succeeds at defining discrimination against women as sex discrimination, takes the sting out of the matter and at the same time does not have to fear much from it anymore.' 13 CEDAW's asymmetrical approach fits well with its substantive approach to equality: 14 it means that the ultimate aim of non-discrimination law is not only to put an end to unfair or unjustifiable classifications of individuals on the basis of a particular characteristic, but also to put an end to oppression and exclusion of groups that are subordinated in society. 15 Although the Convention in most of its substantive articles provides that women have rights of 'on a basis of equality between men and women', it clearly includes the obligation of state parties to enhance de facto equality of women by means of both special measures and eradicating the causes of women's oppression and violence against women. 16 The women-only approach of the CEDAW Convention has also been fiercely criticised. First, it is argued that the Convention falls into the trap of essentialising 'women', thereby neglecting the fact that the unequal relations between the sexes are socially constructed. The Convention, in other words, (supposedly) fails to address or redress gender inequality. Secondly, it is argued that the effect of the asymmetrical approach is that women are further stigmatised as victims per se of discrimination and of (sexual) violence. Thirdly, the approach is said to disregard the possibility that men too may be victims of gender discrimination, such as gender stereotypes and fixed parental roles. 17 The solution to these problems would, at least in theory, be to expand the scope of the Convention to also cover men. While some arguments have been presented in favour of such an interpretation, it is still commonly understood that it is a 'women's only' Convention. 18 14 In other words, that the general abstract equality of all human beings is not the starting point for the construction of legal norms, but the fact that, in reality of social life, large differences in power and differences in the actual situation and capacities of human beings exist. The goal of realising more de facto equality is therefore the core of the substantive approach to legal equality. 15 This refers to the difference between non-discrimination as an anti-classification or as an anti-subordination principle. See  The Convention's focus on women (instead of the neutral category 'sex') is expressed in its title and in the definition of discrimination in article 1. See A Byrnes, 'Article 1', in M Freeman et al (eds) (n 16). For an argument in favour of expanding the Convention's scope to include men, also see T Loenen, 'Het discriminatiebegrip', in A. Heringa et al (eds), Het Vrouwenverdrag. Een beeld van een verdrag (Maklu 1994) 3. 'Nothing seems to stand in the way of men invoking CEDAW, provided that they are equally disadvantaged by something that qualifies as discrimination against women' (translation RH). Loenen gives the example of granting pensions only to widows, which prevents a change of roles between men and women and as such is also detrimental for women (ie discriminatory for women).
The asymmetrical approach of CEDAW is also debated in relation to LGBTI rights. In 2011, Rosenblum proposed that the Convention should be 'unisexed', ie that it would be better to prohibit all discrimination on the ground of sex and/or gender, instead of discrimination against women. 19 Rosenblum presents an important argument in favour of such an amendment of the Convention: the word 'sex' might be understood to include not only the male and female sex but all kinds of sexes, including transgender, intersex and other differently-sexed and gendered people. 20 We shall return to this discussion in section IV, where we briefly address some strategic aspects of whether or not to explicitly include LGBTI people's rights under the scope of CEDAW.

II.ii. Defining discrimination under the Convention
Article 1 of the Convention defines discrimination against women as: Any distinction, exclusion or restriction made on the basis of sex which has the effect or purpose of impairing or nullifying the recognition, enjoyment or exercise by women, irrespective of their marital status, on a basis of equality of men and women, of human rights and fundamental freedoms in the political, economic, social, cultural, and civil or any other field.
The Convention requires that all forms of discrimination against women that lead to an infringement of their human rights be eliminated. The words 'distinction, exclusion or restriction' are interpreted in an extensive way. 21 Such a broad interpretation of the non-discrimination principle moves beyond the mere formal equality principle, 22 indicating a human rights approach to combating discrimination. 23 In such an approach, discrimination against women is seen as an instance of their oppression. 24 The notion of oppression goes beyond mere unequal treatment, but also includes instances of marginalisation, powerlessness, cultural imperialism, and violence. 25 Byrnes, in this respect, notes that there is no reason why the Convention should not be applied to provide protection for women who are discriminated against because of their sexuality where this 'has been used to subordinate women and reinforce male superiority', for example, where 'a lesbian's right to life is violated when she is subjected to death threats for not conforming to dictated heterosexual norms'. 26 It could therefore be argued that the phrase 'all forms of discrimination against women' in the Convention's title also includes oppression that is related to women's sex, sexuality, gender roles, and/or gender identity. II.iii. The objectives of the Convention: space for the inclusion of LGBTI issues?
The interpretation that the Convention also covers oppression that is related to women's sex, sexuality, gender roles, and gender identity can further be underscored when the overall objectives of the Convention are taken into account. It is now widely acknowledged that CEDAW has three main objectives, which are shortly indicated here as: (a) to realise full legal equality between men and women; (b) to enhance de facto equality of women; and (c) to take away the cultural and social roots of gender inequality. 27 The third objective, which is in particular expressed in articles 2(f) and 5a CEDAW, implies that state parties must address gender roles and gender stereotypes that lie at the basis of women's discrimination. 28 According to the CEDAW Committee, article 5a, in conjunction with article 2(f), means that the Convention acknowledges that gender stereotypes and fixed parental gender roles 'affect women not only through individual acts by individuals but also in law, and legal and societal structures and institutions'. 29 Therefore, the Convention not only addresses personal convictions, cultural practices, and traditional values, but also addresses systemic and structural discrimination against women, 30 and in order to overcome the structural discrimination that results from that inequalitycalls for transformative equality or 'equality as transformation'. 31 The CEDAW Committee has made it clear that a correct implementation of the Convention requires 'the recognition that women can have various roles in society, not only the important role of mother and wife, exclusively responsible for children and the family, but also as an individual person and actor in her community and in the society in general'. 32 In this way, the Convention recognises that all human being are equal, have equal rights and deserve respect for their human dignity, but at the same time may have divergent ideas about the way they want to spend their life. 33 Therefore, the concepts of individual autonomy, freedom, and diversity are crucial for a correct understanding of the content and scope of article 5a and of the Convention as a whole.
Women's sexuality and their reproductive capacity are crucial for the construction of gender stereotypes and fixed parental gender roles. This means that the construction of human sexuality as (exclusively) heterosexual forms part of the construction of patriarchal gender relations. 34 The most blatant transgression of the patriarchal female gender identity and her fixed gender role is the lesbian woman who chooses to renounce a male sexual partner and thereby also rejects the protection of the male head of household, and all other forms of male supervision on and control of her life. 35 The obligation to modify gender stereotypes and fixed parental gender roles is also of great importance to men who do not want to conform to their assigned heterosexual masculine identity and male gender role. Beyond that, this obligation is equally important for different from male/female sexed persons (intersex and transgender) and persons with a different from heterosexual sexuality (lesbian, gay, and bisexual). 36 Gender stereotypes and fixed parental gender roles directly affect the lives of all persons who renounce traditional heterosexual and patriarchal feminine and masculine gender identities and gender roles. Through an expansive interpretation of article 5a, all of these situations may be brought under the scope of the Convention.
Violations of LGBTI human rights that are based on gender stereotypes and fixed parental gender roles and sexual roles, therefore, could (at least in theory) be addressed in the framework of the third objective of the Convention, more specifically on the basis of article 5a CEDAW. 37 Examples of such violations that are, in our view, covered under CEDAW and could (and should) be condemned by the Committee are practices such as forced marriages of lesbian, bisexual or transgender persons; taking away their children and placing them under the custody of male relatives or their (former) husbands; denying these persons the right to healthcare, reproductive care, housing, education and employment; and (gang) rape and other forms of violence because of these persons' sexual or gender identity. The Committee's interpretation shows that it, at least partially and cautiously, is going in this direction, at least as far as bisexual and lesbian women and transgender persons are concerned. The question remains whether the Committee does indeed use article 5a for that purpose; ie is it constructing the issue of discrimination against lesbian women (and TBI women) as an issue that falls under the obligation on state parties to change all traditions, practices, and beliefs that somehow contribute to the construction of unequal and oppressive gender relations and heteronormativity?

III. The CEDAW Committee's Position as Regards LGBTI Rights
The CEDAW Committee is the main authority that can interpret the Convention. Its views are not binding, but its monitoring has proven valuable because it allows for 34 See J Butler, Gender Trouble: Feminism and the Subversion of Identity (Routledge, 1990) 1-34 and 110-128; J Butler, 'Imitation and Gender Subordination' in D. Fuss (ed.), Inside/Out: Lesbian Theories, Gay Theories (Routledge, 1991) 13. See also A Gross, 'Sex, Love, and Marriage: Questioning Gender and Sexuality Rights in International Law' (2008) 21 Leiden J of Int L 235. At 251, Gross summarises Butler's position as follows: 'the division in two genders as part of the institution of compulsory heterosexuality, (which) requires a binary polarised gender system since patriarchy and compulsory heterosexuality are only possible in a world built on such a hierarchised division'. Real liberation or emancipation of women and gay and lesbian people, according to this author, requires 'undoing gender', instead of accepting the thus pre-fixed gender categories and identities. Another way of expressing this principle is saying that a transformation of gender and sexuality needs to take place. 35 Lesbian women being gang-raped in order 'to cure them' from their outrageous 'abnormal' sexual preference, is an example of this kind of 'correction'. See for example, Report of the UN Special Rapporteur on the question of torture and other cruel, inhuman or degrading treatment or punishment, Commission on Human Rights March 2006, UN Doc E/CN.4/2006/6/Add.1, para 180 and 183. 36 That is, different from the heterosexual norm and other than the binary male-female scheme. 37 It is therefore in our view not necessary to 'unisex' CEDAW, as is argued by Rosenblum (n 19). Rosenblum does mention article 5 of the CEDAW, but does not appear to see the full potential of this provision in relation to enhancing LGBTI rights. See also section III of this article. dialogue and a critical analysis of the anti-discrimination policies in the various state parties. This section discusses whether the Committee uses the possibility offered by article 5a to interpret the Convention in a way which allows various forms of discrimination against LGBTI persons (either male or female) to fall under the scope of the Convention. Do the Committee's Concluding (Country) Observations and General Recommendations, 38 its Decisions and Views on Individual Communications submitted under the Optional Protocol, or its Report of investigations under the Protocol's inquiry procedure 39 contain any references to the position of LGBTI-persons and the obligations of state parties of CEDAW with respect to their rights? If so, which rights are being given to LGBTI persons on the basis of the Convention? Or, to put it the other way around: what obligations do state parties to the Convention have to respect, protect, and fulfil the human rights of LGBTI persons, specifically connected to their sexual orientation and gender identity?

III.i. Quantitative analysis
For the purpose of this contribution, we carried out a systematic analysis of the Concluding Observations adopted by the CEDAW Committee from 1 January 2010 until 31 December 2014. A total of 110 Concluding Observations were considered in this period. We undertook a full-text search of this material, using the search terms gay, lesbian, transgender, transsexual, LGBT, same-sex, and sexual orientation, in order to determine whether the Committee mentioned (discrimination on the grounds of) sexual orientation or gender identity. In 37 cases, ie in roughly one third of all Concluding Observations included in the analysis, a reference was made to one or more of these categories. Noteworthy examples discussed below include the Committee's Concluding Observations on reports of countries from all continents, for example those of Argentina, Finland, Georgia, Guyana, India, the Netherlands, New Zealand, Norway, Panama, Paraguay, the Russian Federation, Singapore, South Africa, Uganda, Venezuela, and Zimbabwe. 40 Since 1986, the Committee has also published 29 General Recommendations. 41 Sexual orientation is mentioned in only two of these documents: Under article 18, all states which ratify the Convention must report to the Committee. The Committee examines the report, also relying on information sent in by non-governmental organisations, and subsequently publishes Concluding Observations on the state party. Along with offering advice to reporting states in the Concluding Observations, the Committee has the ability to issue General Recommendations on its views on matters concerning the elimination of discrimination against women. women's autonomy in relation to marriage and family matters, recognising 'various forms of family.' 43 In the Committee's case-law regarding complaints by individuals, sexual orientation is mentioned only once. 44 In the three inquiry procedures that were concluded before June 2015, there was no mention of sexual orientation or any of the other search terms used in our analysis. 45 From a quantitative perspective, therefore, the attention the Committee devotes to LGBTI issues seems to remain limited in all categories of documents except for the Concluding Observations. As such, in the next section we turn to a detailed analysis that primarily deals with this category of Committee output.

III.ii. Qualitative analysis: Pre-2010
Our analysis did not include the period before 2010, 46 but it is important to note that the question whether the term 'sex', as employed in the Convention, covers issues concerning women's sexual orientation or sexuality has been a contentious issue for years, as is the extent to which the Convention provides protection to transgender and intersexual persons. 47 Byrnes, in the leading Commentary on the Convention, holds that 'there is, in principle, no reason why the Convention should not be applied to provide protection for women who are discriminated against because of their sexuality'. 48 However, the Committee's approach to this subject in the years up to 2010 is best characterised as vague and cautious. It has not taken a clear stance, although prior to 2010, the Committee had expressed its concern about the criminalisation of same-sex partnerships, 49 while also mentioning discrimination on the grounds of sexual orientation in the two aforementioned General Recommendations published in 2010. 50

III.iii. Qualitative analysis: Post-2010 Concluding Observations
The following are noteworthy findings resulting from our analysis of the quantitative data. First, lesbian women and transsexual, transgender, or intersex persons are most often mentioned by the Committee in relation to the issue of intersectionality, or mentioned in a list of particularly disadvantaged or vulnerable groups of women who need special protection. The categories of sexual orientation and trans-and intersex are apparently 'special' in the sense that they need to be included in such lists. An example of this approach can be found in General Recommendation 28: Intersectionality is a basic concept for understanding the scope of the general obligations of States parties contained in article 2. The discrimination of women based on sex and gender is inextricably linked with other factors that affect women, such as race, ethnicity, religion or belief, health, status, age, class, caste and sexual orientation and gender identity. Discrimination on the basis of sex or gender may affect women belonging to such groups to a different degree or in different ways to men. States parties must legally recognize such intersecting forms of discrimination and their compounded negative impact on the women concerned and prohibit them. 51 Secondly, LGBTI issues are often addressed in the section of the Concluding Observations that deals with violence against women. In some of these Concluding Observations, these categories of women are mentioned in the context of violence against women belonging to vulnerable and disadvantaged groups. The Concluding Observations on Venezuela provide a typical example: 36. The Committee regrets the lack of effective measures taken to address discrimination and violence faced by disadvantaged groups of women, such as indigenous and Afro-descendant women, migrant women, older women, women with disabilities as well as lesbian, bisexual, transgender women and intersex persons, and other women facing multiple and intersecting forms of discrimination. 37. The Committee recommends that the State party adopt appropriate measures to address the particular needs of disadvantaged groups of women. The State party should provide comprehensive information and disaggregated data in its next periodic report on the situation of these women and the measures adopted to address their specific needs. 52 In other Concluding Observations, however, the Committee specifically addresses acts of violence against women on grounds of sexual orientation or transgender identities. The Concluding Observations on Albania are an example of this: 42. While welcoming the adoption of the Law on Protection from Discrimination, which expressly prohibits discrimination on the grounds of gender identity and sexual orientation, the Committee expresses concern about discrimination and acts of violence against women on such grounds. 43. The Committee calls on the State party to implement fully the Law on Protection from Discrimination in relation to discrimination based on gender identity and sexual orientation by providing effective protection against discrimination and violence against women on such grounds. 53 Similarly, in the Observations on Georgia, the Committee calls upon the state to take measures, specifically addressing 'violence against and harassment of lesbian, bisexual and transsexual women'. 54 The examples of Albania and Georgia show that the Committee devotes particular attention to the protection of women who are discriminated against on the basis of their sexual orientation or gender identity. This notwithstanding, it should 51 CEDAW/C/GC/28, para 18. be noted that this attention is still dependent on the Committee's determination to combat violence against women per se, rather than on a broader equality-based approach.
Moreover, while mentioning discrimination or violence against LGBTI persons illustrates the concerns of the Committee on this topic, these references remain fairly gratuitous if they are not coupled with recommendations on how to actually end such discrimination. Most often, the Committee does not offer any such recommendation but merely declares that all women should be able to live without fear of violence. 55 Even considering the constraints under which the Committee needs to operate (such as the sensitivity of the issues it addresses), and while taking into account the cautious and respectful approach that it normally adopts in its recommendations to the state parties, still, the 'recommendations' seem to have the character of a formulaic recitation rather than specifically tailored to the circumstances of the particular state party. A recurring recommendation relates to the enactment of comprehensive anti-discrimination legislation that includes the prohibition of discrimination on the grounds of sexual orientation; 56 sometimes, the Committee (also) recommends the decriminalisation of same-sex relations as an additional measure. 57 Such advice is very broad and accordingly less likely to be effective than concrete, tailored and detailed recommendations. More concrete recommendations can, for example, be found in the Concluding Observations on the Russian Federation, Uganda, South Africa, and Zimbabwe. The Committee's (identical) advice in those instances includes 'launching a sensitization campaign aimed at the general public, as well as providing appropriate training to law enforcement officials'. 58 In the Concluding Observations on Panama, the Committee recommends fighting 'gender stereotypes . . . and violence on grounds such as sexual orientation and gender identity' with 'awareness-raising programmes in school curricula, the training of teachers and the sensitization of the media and the public at large, including actions specifically targeting men and boys'. 59 Other more specific advice can be found in the Concluding Observations on Singapore, where the country is invited to put in place 'a comprehensive strategy to modify or eliminate patriarchal attitudes and stereotypes that discriminate against women, including those based on sexual orientation and gender identity'; 60 or in the observations on Norway, where the state is called upon to take 'specific measures to address difficulties faced by lesbian and transgendered asylum seekers'. 61  orientation or gender identity as a kind of obiter dictum, goes some way to provide state parties with specific recommendations aiming at eliminating discrimination on grounds of sexual orientation.
As regards the question of whether the Committee uses the normative framework of article 5a to address the discrimination or violence against LGBTI persons or to clarify the obligations of state parties in this regard, our findings hold that this is rarely the case. We have found six Concluding Observations in which the role of stereotypes in relation to the situation of lesbian women, bisexual, intersex or transgender persons was mentioned; 62 in two of these a reference was made to articles 5a and 2(f). 63 An excellent example may be found in the Concluding Observations on Panama, in which the Committee writes that: 23. The Committee urges the State party to increase its efforts to design and strengthen comprehensive awareness-raising programmes to foster a better understanding of, and support for, equality between women and men at all levels of society. Such efforts should aim to modify stereotypical attitudes and cultural norms about the responsibilities and roles of women and men in the family, workplace, political life and society, as required under articles 2 (f) and 5 (a) of the Convention. The Committee also urges the State party to transform its recognition of the problem of multiple forms of discrimination into an overall strategy for eliminating gender stereotypes relating to women in general and, in particular, to discrimination against women as specified in paragraph 22. This strategy could include awarenessraising programmes in school curricula, the training of teachers and the sensitization of the media and the public at large, including actions specifically targeting men and boys.

III.iv. Preliminary conclusion
The attention that the CEDAW Committee devotes to LGBTI issues remains scarce in all categories of documents except for the Concluding Observations. Our analysis of the Committee's Concluding Observations since 2010 shows that sexual orientation and some forms of gender identity (most notably transgender) are mentioned in over one third of the documents concerning countries from all continents. The Committee, however, continues its cautious approach and has so far refused to take a clear stance on the question of whether the discrimination ground 'sex' in the Convention includes all identities captured under the LGBTI initialism. This notwithstanding, the Committee has at the same time gradually expanded the protection to lesbian, transgender, transsexual and intersexual persons by means of frequent references to discrimination or violence on these grounds. While many of these references are formulaic recitations rather than specific recommendations, this exploration does however also purport to show that some Concluding Observations do contain concrete recommendations that are specifically intended to help state parties combat discrimination on grounds of sexual orientation and gender identity. Such examples go to show that the Convention, to an increasing extent, is applied by the Committee to provide protection for women who are discriminated against because of their sexuality or gender identity. Although the Committee rarely mentions article 5a in relation to discrimination or violence against these women, it does occasionally mention that the causes of such discrimination and violence lie in gender stereotypes that assign particular roles to women or that conceive of women as subordinate or inferior to men. 64 However, the Committee could make its Concluding Observations and General Recommendations much more persuasive, in our view, by exploring more deeply this link between discrimination against LGBTI persons and gender stereotyping and by developing specific guidelines as regards the obligations that state parties have to 'eliminate' these causes of oppression of and discrimination against LGBTI persons.
The question remains how the Committee could achieve a less erratic and ad hoc approach to this topic. Ernst proposes the adoption of a General Recommendation focusing on discrimination against women based on sexual orientation and gender identity, arguing that the Convention's 'prohibition of discrimination "in any field" ought to apply to their choice of life partner as well, and that they should not be forced by traditional gender roles into having a male partner.' 65 Another suggestion would be the elaboration by the Committee of a General Recommendation on article 5a in which it pays attention to the close link between gender stereotypes and traditional gender roles and the oppression of lesbian, bi-sexual, transgender, intersex and bisexual persons. 66

IV. Conclusion: Is It Wise to Use the Convention to Enhance LGBTI Rights?
In section II of this contribution, we argued that, from a legal point of view, it is possible to interpret the Convention in such a way that the discrimination and violence against LGTBI persons is brought within its scope. The analysis of the existing 'jurisprudence' of the Committee in section III shows that this possibility has not yet been fully utilised. The choice to use CEDAW (or not) for the purpose of enhancing LGBTI rights is also of a political and ideological nature. It is therefore important to also take the theoretical and strategic perspectives into account. We therefore conclude this article with a brief discussion of the pros and cons, or possibilities and problems, of trying to enhance LGBTI rights by means of bringing these categories under the scope of CEDAW.

IV.i. Some theoretical considerations
Although at a theoretical level it is acknowledged that CEDAW covers not only unequal treatment, but all forms of oppression that women experience in life because of their sex and gender, 67 the Convention aims at the elimination of discrimination (against women). The term discrimination in CEDAW has a wider meaning than merely unequal treatment of women as compared to men. Other forms of oppression, exclusion, and violence that are related to sex (ie to being a woman) are covered as well. 68 However, equality, equal treatment and non-discrimination remain central concepts in the interpretation and application of CEDAW. 69 This means that with any attempt to bring LGBTI rights concerning sexuality and gender identity under CEDAW, implicitly, a choice is being made to conceive of these rights mainly through the prism of 'equality' and 'non-discrimination'. The demand for equal rights of the LGBTI movement once again shows that equality or the right to equal treatment continue to have a strong appeal to all organisations that strive to emancipate their constituency. 70 For example, the demands for the right to marriage, the right to custody over children or to establish family law ties with children, and the right to protection of private life of LGBTI persons, are often phrased as demands for equal rights in that regard. Is non-discrimination law indeed something that 'we cannot not want to have'? 71 Many legal theorists have profoundly doubted whether 'rights can be rearticulated outside the traditional liberal legal framework without drawing us back into that framework.' 72 When rights are constructed as equal rights, or rights based on the principle of non-discrimination, regardless of whether legal equality is seen as an anti-classification or an ant-subordination principle, it is most problematic 73 that this law is constructed around the notion of non-discrimination grounds. What then, are the classifications or groups that are thus reinstated and confirmed by this law?
Some considerations concerning 'non-discrimination grounds' Non-discrimination grounds are so-called identity factors: they refer to characteristics of (groups of) persons that, for various reasons, make these persons particularly vulnerable to oppression and exclusion. In non-discrimination law, the characteristics that are thus protected or that deserve strict (judicial) scrutiny are much debated. Various lists of criteria exist, in which most often the factors 'history of persistent past and present discrimination', 'innate or immutable aspect of personal identity ', and 'dignity' are mentioned. 74 In particular, the notion of 'innate or immutable characteristics', like sex, race, colour, sexual orientation, is most problematic. Non-discrimination law, through that criterion, instead of conceiving of differences as features arising out of the inter-group relationships, considers them as inherent characteristics of the non-dominant group. 75 In that way, nondiscrimination law is a 'regulatory power' that contributes to the creation of these (often submissive) identities and makes them even more inescapable for human beings. 76 In regard to LGBTI as a prohibited ground of non-discrimination, it is important to discuss to what extent a distinction can or should be made between sex as a non-discrimination ground and other categories or identities, such as gender, sexuality, and gender identity. 77 Is it at all possible to distinguish between these 'identity factors ', or  In relation to claiming rights in the framework of liberal Western legal systems, Brown (n 12), 231 argues thatalthough being a mixed blessing and leading to many paradoxeswe cannot not want to have rights. closely related that they are inseparable, in as far as human beings' own experience of their personal identity is concerned? Do the categories of sex, intersexuality, and transgender refer to purely biological categories of 'being' of a particular 'kind' (and 'having' particular features like hormones and chromosomessituated in the body), and does sexuality refer to 'feeling' and 'acting' or 'doing' (ie to behaviour)? Or does this distinction lead us away from the real issue, because all denominations of particular features of human existence are human 'acts' and 'performative', ie continuously contributing to their own (re)construction in a particular social, cultural, and economic context? In that regard, the question must be raised whether this division between being and behaving is not a false distinction in itself, denying that both categories are the result of and subjected to social and cultural constructions of the categories that are captured under the LGBTI initialism.
The discussion surrounding the construction of LGBTI persons' equal rights therefore requires a continuous, critical assessment of the meaning of the words ('grounds') that are central in that discussion, like 'sex', 'woman', 'man', 'sexuality', and 'gender'. How are these concepts constructed in social, cultural, legal, and political life, and in what 'power play' are feminist and LGBTI activists involved in when trying to shift the meanings and workings of these concepts? And, just as discussed above, it is not the case that human beings can never be pinned to just one clearly demarcated 'identity factor' , but that subordination, exclusion, oppression, and discrimination, are determined by multiple factors, including wider cultural, social, and economic structures that determine people's life chances? 78 Some considerations concerning 'comparability', 'disadvantage' and 'justifications' Apart from these theoretical problems concerning the choice, construction, and delineation of discrimination grounds, the application of existing equality or non-discrimination law also leads to complicated issues regarding the comparability, disadvantage, and possible justification of the unequal treatment. A whole legal doctrine has emerged over the past 50 years in which the 'equality analysis' or 'equality test' is developed in great detail. 79 The application of the equality principle, through the possibility of bringing forward justifications for the unequal treatment, opens the door for a discourse on conflicting rights and often results in the right to equality having to give way to other rights, most notably the right to religious freedom. 80 Claiming equal rights or, most importantly, the right to non-discrimination, also runs the risk of leading to assimilation to the existing dominant (in this case: heterosexual) 77 In European Union equal treatment law, the categories sex (including transsexuals) and sexual orientation are clearly distinguished and regulated in separate Directives with distinctive scopes of application and possibilities to justify inequality of treatment. Sexual harassment is regulated only in the context of European Union sex equality law, which leaves LGB persons unprotected against this form of discrimination. See R Holtmaat, 'Sexual Harassment and Harassment on the Ground of Sex in EU Law: a Conceptual Clarification' (2011) 2 Eur Gender Equality L Rev 4. norms. 81 As Rubin points out, demands for equal rights of persons with a 'different sexuality', instead of creating real sexual liberty, at best leads to giving them the right to have sex in a safe surrounding of registered partnership or marriage, leaving all sexual relations outside these legal constructions in the categories of 'deviant', 'abnormal', or even 'criminal' forms of behaviour. 82 This leads to the urgent question of whether it is necessary or desirable at all to use this existing legal equality or non-discrimination framework to enhance the protection of the (already existing!) human rights of LGBTI persons. In other words: why claim that LGBTI persons have an equal right to marry, instead of simply claiming that they have the right to marry as such? 83 Is it not true that all human beings have the right to marry, the right to family life, the right to be free from violence, the right to freedom of expressionincluding expression of their sexual and gender identity, etc? Why not continue on the road of the Yogyakarta Principles and claim that these existing human rights will finally be applied to all human beings? 84 IV.ii. Some strategic issues A last, but certainly not least important, dimension of the question of the instrumental value of CEDAW for the purposes of enhancing LGBTI rights is strategic: what are the possible positive and negative effects of such a strategy? Theoretical concepts and constructs like 'gender' or 'gender identity' are developed as tools for a critical assessment of existing unequal power relations between the sexes and of forced heteronormativity and can be used in the context of emancipatory projects. Our main concern at this juncture is that they can also be used by powerful (often religious) groups to organise resistance against any changes in the unequal relationships between the sexes and against changes regarding forced heteronormativity, and thus can become useless or even damaging for these same emancipatory projects. What repression or opposition may be expected, in particular from the religious right in the world, when any such an attempt is made? 85 Would that opposition also damage the enhancement of 'women's human rights' to such an extent that we should not go down this route? Or is it more important that we can escape from the detrimental effects of the asymmetrical approach when we leave the paradigm of 'women's rights' behind us? 86 On the other hand: would a shift to a symmetrical approach not bring us from bad to worse, as this would mean that the m/f comparator model of sex equality is thereby adopted, with all its assimilationist effects? Answering these questions is beyond the 81 A Gross (n 34), 246-247, who discusses this problem in relation to demanding same-sex marriage. 82 G Rubin, 'Thinking Sex: Notes for a Radical Theory of the Politics of Sexuality', in CS Vance (ed), Pleasure and Danger: Exploring Female Sexuality (Routledge 1984) 267. See also A Gross (n 34) 246-247, who holds that demanding the equal right to marry 'reinforces the message that marriage is the ultimate form of human relationship, whether different sex or same-sex', and that 'this message is discriminatory towards all who do not participate in it.' 83 Of course, these rights have to be expressly claimed by LGBTI persons because they have historically been, and still continue to be, denied to them in all cultures and all legal systems. There is an important role to be fulfilled by the judiciary and the treaty bodies in interpreting these (universal) rights in such a way that they are truly inclusive, ie are truly granted to all human beings. An overview of ways in which courts and treaty bodies have done so can be found in A Byrnes, 'Gender Challenges for International Human Rights', in S Shearan and N Rodley (eds), Routledge Handbook of International Human Rights Law (Routledge, 2014) 631-633. 84 The Yogyakarta Principles have been described in detail in M O'Flaherty's contribution to this issue. 85 For a discussion of this resistance in relation to the concept of gender, see J Scott, 'The Uses and Abuses of Gender' (2013) 16 Tijdschrift voor Genderstudies 63. 86 The argument was advanced by D Otto (n 10). exclusive competence of lawyers, and requires an in-depth study from political scientists and other social science disciplines as well. We hope that this article contributes to having this discussion.
We would also like to devote a few final words to the proposals that were made in order to either shift from an asymmetrical to a symmetrical approach ('unisexing CEDAW'; Rosenblum 87 ) or to explicitly extend the scope of CEDAW to cover the elimination of all forms of discrimination against LGBTI persons ('supersexing CEDAW'; Hernández-Truyol 88 ). In our view, Rosenblum's main reason to unisex CEDAW (ie in order to give rights to LGBTI persons), is not a convincing argument to change the non-discrimination ground of the Convention from 'women' to 'sex'. A prohibition of discrimination on the grounds of sex is commonly interpreted in a strictly binary or bipolar scheme in which the (presumably essential) male and female sexes are compared with one another; only when one of the two sexes suffers a certain disadvantage for which there is no objective justification, this non-discrimination norm becomes applicable. The vast (feminist) literature in the area of sex equality rights shows the strong tendency towards assimilation to the male norm that is inherent in any formal sex-equality norm. 89 Besides, a 'gender neutral' sex equality norm suffers from the dangerexemplified in eg the practice of EU sex equality lawthat very often men appeal to this norm in order to claim the advantages women might have over them (eg in the area of protection of motherhood). 90 The result of such legal actions then is to water down these entitlements, leaving men and women without these advantages. Apart from this practical problem, there is also the ideological obstacle which holds that changing the understanding of sex to include 'other sexes' and sexualities, as is suggested as the solution to this problem by Rosenblum, might appear to be as difficult and controversial as acknowledging that differences between men and women are culturally and socially constructed instead of 'natural' or 'God-given'.
In reaction to Rosenblum's proposal, Hernández-Truyol suggests the adoption of a second optional protocol to CEDAW in which principles similar to the Yogyakarta Principles are laid down in order to 'super-sex' CEDAW, thereby expressly including the rights of LGBTI persons under the Convention. 91 To our mind, this proposal is also rather unrealistic. Progress in the area of creating international instruments that guarantee the human rights of LGBTI persons is very slow. Like Garvey has demonstrated in his article on the history of the Brazilian Resolution that aimed to enhance LGBTI rights, any such attempt may invoke so much resistance that progress is blocked, instead of enhanced. 92 Garvey therefore argues in favour of enhancing these rights through the slow and piecemeal judicial process, instead of through the creation of new international law documents. Critical assessments of the very limited or adverse instrumental value of liberal and formal equality law are abundant. See eg S Fredman, Discrimination Law (OUP, 2011), in particular from 8. Increasingly, it is acknowledged that replacing the notion of formal equality with that of substantive equality will also 'not deliver the goods'. See eg R Hunter, 'Introduction: Feminism and Equality' 8; R Hunter, 'Alternatives to Equality', and K van Marle, 'Haunting (In)equalities', all in R Hunter (n 70). 90 See above in section II, where we discuss the effect of a symmetrical sex equality norm.