Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

"Other" women in flight : sexual minority and polygynous refugee women Yorgun, Siobhan L. 2020

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2020_november_yorgun_siobhan.pdf [ 4.53MB ]
JSON: 24-1.0394731.json
JSON-LD: 24-1.0394731-ld.json
RDF/XML (Pretty): 24-1.0394731-rdf.xml
RDF/JSON: 24-1.0394731-rdf.json
Turtle: 24-1.0394731-turtle.txt
N-Triples: 24-1.0394731-rdf-ntriples.txt
Original Record: 24-1.0394731-source.json
Full Text

Full Text

  “OTHER” WOMEN IN FLIGHT:   SEXUAL MINORITY AND POLYGYNOUS REFUGEE WOMEN by  Siobhan L. Yorgun  B.A. Hons, LLB., The University of the Witwatersrand, 2008  LLM., Leiden University, 2010  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Law)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   October 2020  © Siobhan L. Yorgun, 2020  ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled:  “Other” Women in Flight: Sexual Minority and Polygynous Refugee Women  submitted by Siobhan L. Yorgun  in partial fulfillment of the requirements for the degree of Doctor of Philosphy In The Faculty of Graduate and Postdoctorate Studies (Law)  Examining Committee: Proffessor Catherine Dauvergne, Q.C., Peter A. Allard School of Law, University of British Columbia Supervisor  Associate Professor Efrat Arbel, Peter A. Allard School of Law, University of British Columbia Supervisory Committee Member  Professor, Associate Dean, Gillian Creese, Department of Sociology, University of British Columbia Supervisory Committee Member Professor Liora Lazarus, Peter A. Allard School of Law, University of British Columbia University Examiner Associate Professor Antje Ellermann, Department of Political Science, University of British Columbia Associate Professor University Examiner     iii  Abstract  Refugee law scholarship is largely focused on the application of refugee law and the refugee status determination processes of Western host countries. This geographical focus of the majority of scholarly work stands in stark contrast to the global distribution of refugees, the majority of whom are hosted by countries of the Global South. This dissertation exposes this imbalanced knowledge production in refugee law, specifically the way in which the unacknowledged bias in the study of refugee law as applied only in select Western countries, limits and indeed distorts our understanding of international refugee law.   This distortion has particular consequences for some of the most vulnerable and least mobile refugees residing in host countries of the Global South. This dissertation is based on a qualitative data analysis of interviews I conducted with refugee service providers and refugees in South Africa, supported by a study of the existing literature and legislation. This dissertation presents an alternatively located, feminist study of the understanding of especially ostracized women; polygynous and sexual minority refugee women, within the South African asylum system.   I conclude that a variety of factors contribute to the invisibility of these women to service providers, factors which are influenced by and familiar to Western refugee law studies;  stereotypical expectations of the refugee figure and of polygynous families and of queer identity, compounded by the socio-economic isolation of these women refugees. I identify particularly relevant distortions in the current study of refugee law, including the neglect of the nexus between economics and experiences of persecution, especially of continuing persecution within iv  host states. I also problematize the reliance on an identity-politics framing of persecuted identities in the study and application of refugee law, which especially disadvantages refugees with intersectional vulnerabilities. The refugees most impacted by these systemic problems in our understanding of refugee law; the economically disadvantaged and those with intersectional vulnerabilities (including, and amplified by, economic disadvantage), are similarly those whose very disadvantage and vulnerability limit mobility. They are thus those most likely to be underrepresented in the distant and expensive states whose asylum systems form the predominant focus of refugee law studies.              v  Lay Summary Refugee law studies tend to focus on the asylum cases determined by Western host states. Yet the majority of refugees seek refuge in the Global South. This dissertation unmasks this bias in the way refugee law is studied and illustrates the impact of this bias in creating a distorted knowledge of refugee law. This distortion has particular consequences for some of the most vulnerable and least mobile refugees in host countries of the Global South who are both underrepresented in the Western host state studies and who are distinctly disadvantaged by the ways of thinking such studies have produced.   This dissertation is based on an analysis of interviews with refugee service providers and refugees in South Africa, with a focus on sexual minority and polygynous refugee women. I argue a holistic understanding of refugee law requires acknowledging the current bias and studying refugee law in a variety of host states.   vi  Preface I identified and designed this research project myself and I conducted all primary interviews, research, and data analysis and reviewed the literature myself. Interviews were transcribed by an external transcriber. This work is unpublished. This research required approval of UBC’s Behavioral Research Ethics Boards, Certificate Number of Approval, Full Board obtained: H16-00625.   vii  Table of Contents Abstract .......................................................................................................................................... iii Lay Summary .................................................................................................................................. v Preface............................................................................................................................................ vi Table of Contents .......................................................................................................................... vii List of Figures ................................................................................................................................. x List of Abbreviations ..................................................................................................................... xi List of Symbols ............................................................................................................................. xii Acknowledgments........................................................................................................................ xiii Dedication .................................................................................................................................... xiv Chapter 1: Introduction ............................................................................................................... 1 1.1 Purpose of Study ........................................................................................................... 1 1.2 Overview ...................................................................................................................... 2 1.3 Methodology ................................................................................................................. 4 1.3.1 Theoretical approaches as methodological tools ................................................... 5 1.3.2 Location of study ................................................................................................... 9 1.3.3 Population being studied ..................................................................................... 12 1.3.4 Participants .......................................................................................................... 19 1.3.5 Matrix for analysis .............................................................................................. 25 1.4 Principal Conclusions ................................................................................................. 31 Chapter 2: Literature Review .................................................................................................... 34 viii  2.1 The Western Bias in Refugee Law Scholarship ......................................................... 35 2.2 Useful Insights: Lost in Translation ........................................................................... 43 2.3 Refugee Law Voices from the South .......................................................................... 54 2.4 New Approaches Emerging in Refugee Law – Economics ....................................... 65 2.5 Beyond Refugee Law ................................................................................................. 67 2.5.1 Queer and feminist theorizing on gender, family and marriage .......................... 69 2.5.2 False universalism and Third World counters..................................................... 75 Chapter 3: The Lay of the Law ................................................................................................. 82 3.1 Alphabet Soup: PSGs, LGBTs - Add Gender and Stir ............................................... 82 3.2 Over the Rainbow: The State of Refugee Law in South Africa ................................. 90 3.2.1 Painting a picture ................................................................................................. 91 3.2.2 Colour me pink .................................................................................................. 111 3.2.3 Edge of frame: how the asylum system applies to queer and polygynous lives 127 Chapter 4: Western Hosts and Southern Ghosts ..................................................................... 139 4.1 The West-as-host ...................................................................................................... 143 4.2 The West, the Rest and the Dislocated Self .............................................................. 149 4.3 We Three Kings of Orient Are ................................................................................. 154 4.4 Unsettling Others ...................................................................................................... 164 Chapter 5: Not for Love or Money ......................................................................................... 181 5.1 Inhospitable Hosts .................................................................................................... 183 5.1.1 Unwelcome........................................................................................................ 184 5.1.2 Unrecognized .................................................................................................... 186 5.2 Options for unwanted guests .................................................................................... 192 ix  5.3 Locating the Money-less Refugee ............................................................................ 202 5.3.1 Economics in international refugee law ............................................................ 204 5.3.2 Buying a safe space ........................................................................................... 206 5.3.3 Intersecting silences .......................................................................................... 211 5.3.4 The economic refugee ....................................................................................... 217 5.4 Abbreviating Minorities ........................................................................................... 221 5.4.1 The always male, queer refugee ........................................................................ 223 5.4.2 The always queer refugee .................................................................................. 238 5.4.3 The unrelated, queer refugee ............................................................................. 254 Chapter 6: Come Out, Come Out, Wherever You Are! .......................................................... 274 6.1 Women with Water in Their Mouths ........................................................................ 274 6.1.1 Theories on silence ............................................................................................ 276 6.1.2 A woman’s place ............................................................................................... 282 6.2 There’s Nowt so Queer as Folk: Bringing Intersectionality into Refugee Law ....... 290 6.2.1 Moving Out-Side ............................................................................................... 292 6.2.2 Thinking outside the box ................................................................................... 296 Bibliography ............................................................................................................................... 302 Appendices .................................................................................................................................. 320 Appendix A......................................................................................................................... 320 Appendix A.1: BREB approved sample Letter of Initial Contact – Refugees............... 320 Appendix A.2: BREB approved sample Letter of Initial Contact – Professionals........  323 Appendix B: Nationality days for Asylum Seekers at Refugee Reception Centres............ 326 Appendix C: Draft Standard Operating Procedure: Refugee Family Reunification........... 327 x  List of Figures Figure 1: Flow of LGBTI People through the International Protection System …………218   xi  List of Abbreviations  AU  African Union, previously the Organisation for African Unity CAL Coalition of African Lesbians CEDAW Convention on the Elimination of all Forms of Discrimination Against Women DHA Department of Home Affairs JRS Jesuit Refugee Services LGBTI Lesbian, Gay, Bisexual, Transgender, Intersex (discussed in detail in Chapter 5.4) LHR  Lawyers for Human Rights NGO  Non-governmental Organisation  OECD  Organisation for Economic Co-operation and Development ORAM Organization for Refuge, Asylum and Migration PASSOP People Against Suffering, Oppression and Poverty PSG  Particular social group RAB  Refugee Appeal Board RRO  Refugee Reception Office RSDO  Refugee Status Determination Officer SADC  Southern African Development Community TWAIL Third World Approaches to International Law UN  United Nations UNHCR United Nations High Commission for Refugees   xii  List of Symbols ∪ Mathematical logic set theory uses the symbol ‘∪’ to denote the union of sets, including those elements that are common to both.1 In attempting to discuss women who are queer, women who may not identify as queer as well as men who are queer, acknowledging also the complexity involved in gender being neither binary nor static, the use of ‘and’ between the groups ‘women’ and ‘sexual minority’ aggravates a problem at the heart of this dissertation: that queer persons and the refugee figure generally are predominantly imagined as male, and the inhibiting practice of a disaggregated listing of identities. Similarly the use of ‘and/or’ fails to draw attention to the overlap of intersecting identities. The symbol ‘∪’ can therefore serve as a useful means of listing overlapping and interconnected identity markers and is used in this work for this reason.                                                                1 S Shen et al, Basic Set Theory (American Mathematical Soc., 2002), p.2. xiii  Acknowledgments It is customary to acknowledge that the university is located on the traditional, ancestral, and unceded territory of the Musqueam people, before convening events at the University of British Columbia. In my case I both studied and I and my family lived on this land for the majority of my doctorate. It has been suggested that in academia, we go beyond a rote recitation of this land acknowledgment by purposively linking the work we create with the histories and experiences of this land’s indigenous peoples. In this vein, I would like to acknowledge that whilst refugee law sits uneasily with the experiences of indigenous persons forced off their ancestral land yet whose intra-national displacement has no international border crossing, the knowledge biases disadvantaging both polygynous and sexual minority women, discussed in this work, have distressing parallels for the indigenous peoples of Canada.  I would like to thank my wonderful supervisor Catherine Dauvergne – the confidence in me you always showed and the balance of your guidance meant I felt invigorated every time I left our meetings. Thanks also to my supportive and always helpful supervisory committee, Gillian Creese and Efrat Arbel, to my sister, Caitlyn for all the transcriptions, my parents, Sean and Lynn, for steering me through a lifetime of education that led me to this opportunity as well as for their and my sister’s help editing this volume of work. Thanks to my friends and co-students Anna and Meredith, for their constant friendship, energy, faith and support. To my scattered besties: Hadar, Elmarie, Cathy and Jo who were there for me with consolation and laughs in the midnight hours of caring for babies, many thanks. My greatest dept of thanks goes to my husband, Cemil Yorgun, without whose unfaltering faith in me, loving parenting and great sacrifice, I could not have written a single word. xiv  Dedication I dedicate this dissertation to my husband and children, who buttressed my life outside of this work and from whose lives I had to be absent to complete it, and to the strong souls of the refugees I met and those who work to help them.                       1  Chapter 1: Introduction This dissertation incorporates concerns with gender, sexuality, family and refugee identities, with a feminist and Third World Approaches to International Law agenda. It is located within the discipline of legal studies, more specifically, within refugee law studies. In introducing my dissertation, I set out the purpose of the study, and signpost the structure, arguments and conclusions to follow. I also provide discussion on the theoretical approaches, practicalities and politics which have guided the choices made in how I conducted my research and how the work of this dissertation is presented. This uniquely oriented project argues for a holistic approach to key constructions in refugee law of ‘host country’ and of ‘refugee’. This project both supports a call for differently located refugee law studies, through its distinctive findings, and presents as an example of a Southern-host focused study marking, and demanding the marking, of its positionality.   1.1 Purpose of Study The purpose of my research is to encourage reflective thought on the part of those engaged with refugee systems in both Western and Southern host states on what is missing, who is being left unheard and unvoiced in the study of refugee law and what implications this may have for a working understanding of the application of refugee law in the world.  This dissertation reveals a bias in the study of refugee law evident in a preoccupation with the study of the asylum systems of Western states. The effects of a disproportionate focus on (minority) Western host states in refugee law studies are pernicious, particularly for the most vulnerable and least mobile. In this regard, women ∪ sexual minorities have particular and 2  compounded vulnerabilities as refugees and are especially likely to remain within the Global South, thus being markedly neglected by refugee law scholarship focusing only on Western hosts. I demonstrate this with reference to the differences and similarities to Western-focused refugee law study themes by centering the leading destination country of new asylum-seekers from 2006 to 2012, and my home country, South Africa, in an investigation of the interaction with this asylum system of sexual minority women refugees and of women refugees in polygynous relationships, as understood by refugee service providers.   1.2 Overview  This dissertation teases out both the familiar and the novel in a focused study on the position of ostracized groups of women in the South African asylum system. This work highlights, more generally, the importance of expanding refugee law studies to cover a greater variety of host countries, wherein both new and familiar themes can progress a more holistic and authentic understanding of refugee law in the world.  It is thus necessary to identify what existing refugee law analysis, focused on Western host countries, is supported through this study’s findings, contrasted with what different perspectives and new challenges these findings unveil. In order to do this, the first chapter of this dissertation, post introduction, provides a literature review illustrating the shortcomings as well as useful insights amongst traditional refugee law studies as well as from newer approaches in refugee law and multidisciplinary perspectives, against which this study can be set.  Subsequently, Chapter 3 discusses the relevant contextual legal background, at both the international and domestic level, which informs this refugee law study, including an analysis of 3  South African refugee processes, laws and their application and failings. The South African asylum system is under such pressures that any research and representation of this system can only show it as deeply flawed and in crisis. My purpose in this work is not to vilify the South African asylum system, neither is it to romanticise its innovations or Southern-ness. The argument raised is more simply that it is a significant host country whose application of refugee law requires attention and the study of which can further our understanding of refugee law more generally. The geopolitical gap in much refugee law scholarship, identified in the literature review in Chapter 2, is theorised and expanded upon in the first half of Chapter 4. In this chapter I identify and dissect a number of unsubstantiated binaries and biases. I demonstrate how these biases create conceptual stumbling blocks that implicate the choice and supposed representivity of studies of only select host countries’ application of refugee law. The analysis of research data that follows in this Chapter builds upon and reflects the consequences of these identified biases.  The analysis of the qualitative data, presented in this dissertation, is divided into an analysis of the data collected in relation to polygynous refugees (in Chapter 4) and in relation to queer refugees (in Chapter 5). The different chapter lengths in this dissertation are thus a reflection of the differences in data volumes gathered on these two subject groups which are ostensibly opposite ends of a patriarchal spectrum of marriage or relationships, with very different sites of acceptance and disapprobation, different claims of culture and different orientations to religion; the only apparent unifying factor being that neither comply with a ‘one man and one woman’ formulation of adult, sexual relationships. This is a loose division of the data which was gathered in interviews which were not segregated by population focus. However, the subjects of study are two distinctive groups and those service providers whose services specifically focused on queer 4  clientele, generally believed they had nothing to comment on polygynous clients (though elements of a happenstance overlap are discussed in Chapter 6). Yet much of the thematic analysis garnered from these Chapters presents parallel findings on the effects and influence of both familiar and unexamined stereotypes embedded in refugee law studies and applications as regards the refugee figure, the ‘host’ label and persecuted identities. I present a discussion on the ‘missing voices’ of the women whose experiences with the asylum system this dissertation sought to investigate and the broader implications of this study’s findings in the concluding chapter, Chapter 6.  1.3 Methodology The principal focus of the analysis presented in this dissertation is the qualitative data obtained through primary interviews I conducted myself in South Africa over a two month period in 2016 with refugee service providers and with refugees. This work makes reference to limited official government data on asylum applications in South Africa and to various domestic, South African and international legal instruments and legislation.2 The discussion of both data and legislation includes the presentation and analysis of interviewee opinion concerning the relevant government data and legislation. The primary focus of the interviews was not the legislation or data per se but more specifically targeted interviewees’ perceptions of the interaction between                                                  2 The background theoretical and doctrinal research was conducted through library research using word search databases and following up on readings referenced within works already read, and looking for additional work by relevant authors. Research on the South African legislative background made use of legislative databases: The Southern African Legal Information Institute (SAFLII) available at:; the online collection of South African Government Gazettes (the Green Gazette), available at:; South African Parliamentary publications, available at:; and the reports of the South African Law Reform Commision (SALRC) available at: 5  queer and polygynous refugees with the asylum system. This subsection details how this research was conducted and I expound on the choices I made in both conducting and presenting this research.  1.3.1 Theoretical approaches as methodological tools It is my intention to turn the critical gaze not on the othered populations featured in the title of this dissertation – sexual minority and polygynous refugee women – but on the representation and understanding of these ‘others’ within the bureaucratic administration and legal and advocacy organisations. Furthermore, I will reflect upon these issues from within a firmly, purposively Southern perspective, wherein I speak of Southern-positionality not as other but as self, highlighting the falsity of the unexamined West-as-self and West-as-host narratives in comparison.3   Third World Approaches to International Law (TWAIL) and Feminist Legal theory  have both been described as methodology, as much as theories; encapsulating a political commitment underlying both the ‘what’ and the ‘how’ of research, thus providing guidance as to the way in which research is conducted, as well as the major ethical concerns.4 TWAIL require sensitivity to power relations between states and the politics of knowledge production. TWAIL scholars are consequently critical of the interests served by international law. Furthermore, TWAIL stress that                                                  3 Aileen Moreton-Robinson, Talkin’ up to the white woman: Indigenous women and white feminism (St Lucia, Qld: University of Queensland Press, 2000).at (xvi) warns that this requires conscious effort where the actual ‘self’ is so automatically othered in academia such that it becomes all too easy to fall into the trap of using othering-language such as ‘they’/’them’ rather than ‘we’/’us’. 4 Obiora Chinedu Okafor, “Critical Third World Approaches to International Law (TWAIL): Theory, Methodology, or Both?” (2008) 10:4 International Community Law Review 371–378.; Hilary Charlesworth, Christine Chinkin & Shelley Wright, “Feminist Approaches to International Law” (1991) 85:4 The American Journal of International Law 613–645 at 615-618. 6  the lived experience of Third World peoples be centred in research so as to transform law from a tool of oppression to one of emancipation.5  Feminist Legal theory, in turn, demands gender sensitive methodologies, aimed at producing research with the ethic of contributing towards gender equality.6 Thus both feminist legal theory and TWAIL emphasise the importance of missing voices othered by a hegemonic male, First World construction of international law as universal. Another common element amongst these two approaches to legal research is that, whilst maintaining a critical stance to the existing law, many scholars demonstrate an underlying faith in the possibility of law serving as a mechanism for positive change; arguing for different approaches to law-making rather than its abandonment.7 Together, these approaches will provide a lens through which to critique the prevailing Western-influenced approach to gender, family and sexuality, thus plumping out an emaciated conception of this aspect of international refugee law.  As regards the knowledge production within refugee law, Feminist and TWAIL perspectives used together reveal the way in which refugee law and the predominant study thereof is both gendered and particular to Western paradigms, yet presented as general. An awareness of the politics of knowledge production requires both an unveiling of impartial knowledge, and its makers, disguised as neutral or the ‘every-man’, and self-reflection on one’s role therein.8 Thus,                                                  5 Okafor, supra note 4 at 375-376. 6 See the debate between Halley and Romero in Janet Halley, “Queer Theory by Men”, (15 April 2016), online: Feminist and Queer Legal Theory <>; Adam P Romero, “Methodological Descriptions: ‘Feminist’ and ‘Queer’ Legal Theories”, (15 April 2016), online: Feminist and Queer Legal Theory <>; Martha Albertson Fineman et al, Feminist and Queer Legal Theory : Intimate Encounters, Uncomfortable Conversations (Routledge, 2016). 7 Antony Anghie & B S Chimni, “Third World Approaches to International Law and Individual Repsonsibility in Internal Conflicts Agora: Third World Approaches in International Law” (2003) 2:1 Chinese J Int’l L 77–104.78; Charlesworth, Chinkin & Wright, supra note 4. 8 Chandra Talpade Mohanty, Feminism without Borders: Decolonizing Theory, Practicing Solidarity (Duke University Press, 2003). at 17-19. 7  as these perspectives feed into my dissertation, it is the unacknowledged West-as-subject which compounds the gendered effect of the knowledge produced by such scholars, given the male preponderance of refugees in the West.9  The distinctive experiences of sexual minority refugees in the West as opposed to those in hosts of the Global South, similarly marks this disparity and points towards the incompleteness and disempowering effects of the Western-biased knowledge being produced.10  The intended audience of this dissertation are those working within the refugee law paradigm in both developed and developing country hosts, conscious of the significance of ‘speaking to’ in a project of greater inclusivity.11 In advocating for greater research on a greater variety of refugee host countries and their refugee law systems and application of international refugee law, this research is presented both as an example of different work focused on a host not typically studied in a style akin to the in-depth thematic studies of Western host states and as an attempt at broader audience consciousness.  As methodologies, TWAIL and feminist legal theory provide the formulation of a particular set of concerns and the analytic tools with which to explore refugee law. An approach inspired by a                                                  9 Gayatri Chakravorty Spivak, “Can the Subaltern Speak?” in Rosalind Morris, ed, Can the Subaltern Speak?: Reflections on the History of an Idea (New York, UNITED STATES: Columbia University Press, 2010) 21. The United Nations High Commissioner for Refugees UNHCR, “UNHCR Global Trends 2018”, online: UNHCR <>. published 7 June 2019, states at 61-2: “Based on the available data, the proportion of women and girls in the refugee population was 48 per cent in 2018, similar to the past few years. […]At the country level, there was wide variation in the sex and age breakdown of hosted refugees. […]These differences are also seen at a regional level [Figure 22]. The lowest proportion of both children and women was seen in the refugee population in Europe where only 44 per cent of the refugee population was female and 41 per cent was under the age of 18 in 2018 (although the data coverage was also very poor in this region so the estimates are indicative only). In contrast, the highest proportion of both women and children was in sub-Saharan Africa with 52 per cent and 57 per cent respectively.” 10 See further the discussion on host-nation continuing experiences of persecution and closetness, experienced with greater prevalence by sexual and gender identity minorities in the Global South, in Chapter 5. 11 See Spivak’s conclusion on the importance of speaking to, rather than for, the subaltern Spivak, supra note 9 at 91. 8  conjoined TWAIL and feminist project also allows for the revelation and analysis of weaknesses in each method: TWAIL has a sister in the critical race theory response to feminist theory and demands the exposition of racialized understandings at the global level, whereas a feminist approach rebukes attempts to sideline gender and sexuality concerns in a post-colonial project.12 In making the controversial juxtaposition of topics of sexual minorities and polygyny in this study, I simultaneously challenge both liberal Western and African and Middle Eastern cultural perceptions of acceptable and unacceptable women's sexualities. As Moreton-Robinson states: “Knowledge is never innocent or neutral. It is a key to power and meaning. It is used to dominate and control.”13 This idea is at once the rationale behind a criticism of the hegemonic Western-host focus in refugee law scholarship, a motivator to engage in alternative knowledge-production, and a warning of the responsibility and culpability involved in research. I argue that scholars focused on Western refugee law systems should firstly be aware of the limitations and specificity of their work, and secondly be aware of the generality of their audience, to be purposeful in ‘speaking to’ Southern scholars, advocates and bureaucrats who are indeed listening. In the angst of the limitations and dire problems of the current international refugee system, this project offers not just criticism but points to an exciting opportunity to learn more and think differently.                                                   12 Both Merry and Mullaly refer to the difficulties faced by women’s rights campaigners situated within a post-colonial political nation-building project in which a sense of the ‘fragility of nationality’ compounded with an association of women as the keepers of national identity, marks as threatening and so pushes back challenges to gender boundaries. See Sally Engle Merry, Gender Violence: A Cultural Perspective (Chicester, UNITED KINGDOM: John Wiley & Sons, Incorporated, 2008); Siobhán Mullally, Gender, culture and human rights: reclaiming universalism (Portland, Or;Oxford; Hart Pub, 2006). 13 Moreton-Robinson, supra note 3 at 93.  9  1.3.2 Location of study I conducted field research for this project between mid October and mid December 2016 in Johannesburg, Pretoria and Cape Town, South Africa. I conducted an additional interview at the end of December 2016 via a Skype phone call with an interviewee based in Pretoria. South Africa, a host country within the Global South, has laws recognising same-sex marriage, unique to Africa, as well as laws legalising polygyny, not found in Western host countries. According to the annual United Nations High Commission for Refugees world report, South Africa received the greatest number of individual asylum seekers globally, for 7 years continuously, from 2006 to 2012.14 As a South African, having worked with the refugee system as a lawyer, the South African refugee regime is already centred for me, having defined my experience of refugee concerns. The South African approach forms the basis of my assumptions of the ‘normal’ approach, contrary to the majority of refugee law scholars discussed in the Literature Review below. I aspire to meet that definition of ‘African’ as used by Tamale, to refer to those who believe: “that serious global knowledge creation requires that the lives, experiences, ideas and imagination of people throughout the continent be considered critically important”15. However, as a project in generating new perspectives, I should also “own [my] whiteness and ... social authority”, along with the fact that I am undertaking my doctorate from within ‘the West’, based as I am at the University of British Columbia in Vancouver, Canada.16 For this reason I attempt to step out of a neutral third person researcher position and make use of first person acknowledgments of position where fitting. I have not uniformly used first person, however, as                                                  14 United Nations High Commissioner for Refugees UNHCR, “UNHCR Global Trends 2012”, online: UNHCR <> at 25.  15 Sylvia Tamale, African Sexualities: A Reader (Fahamu/Pambazuka, 2011) at 1.  16 Moreton-Robinson, supra note 3 at 118.; see also Spivak, supra note 9. at 87 who insists on the marking of the positionality, as investigating subjects, of the Western theorists involved. 10  to do so could detract from the points being made where the constant disruption of academic expectations of tone and position could become distracting. I chose South Africa as the study location of my research based on my experience and knowledge of the system and the players, and this choice is supported both by the numerical significance of its asylum system as well as the uniqueness of its domestic laws in relation to the study groups. I did not make this choice because South Africa is either typical or representative of host states of the Global South. From a colony to apartheid, styled as a white, (selectively) developed country, post-apartheid South Africa continues to display a duality in its socio-economic positionality; teetering between a developed and developing status. The terms West and South are used here as politically-charged terms indicative of an imagined, rather than real Geography. An imagined geography, which, it must be acknowledged, seldom takes account of the indigenous peoples within the ‘West’.  ‘Third’ and ‘First World’ are terms that, post Cold War, are considered outdated whilst having had much political and academic work put into them. Developed and developing countries on the other hand, are terms that I also use but that I find less conducive to discussions for their attempted neutrality. These terms are approved by the United Nations (UN) and other international institutions. Whereas the term ‘Least Developed Countries’ is an empirical label applied by the UN, based upon qualifying criteria and comparative assessment, ‘developed’ and ‘developing’ are terms of choice. Within the UN system, each country may elect whether to class itself as developed or developing, with consequences for various international obligations, including environmental obligations. In contrast, the geo-spatial narrative structures under discussion here are politically and culturally imbued, with normative assumptions of permanence. The other term I seek to use is Majority World; whilst lesser known, I believe it is 11  worth attempting to gain traction.17 The term ‘Majority World’ has the potential to disrupt the dominant narrative centrally problematized in this work: that the author-academic-perspectival ‘I’/’we’/‘us’ is presumptively the West, and the constitutive West-as-host (my own term, discussed in Chapter 4). What this project illustrates is that in spite of South Africa being a regional hegemon, the economic powerhouse of Africa, a newly industrialized state and one of the ‘BRICS’18; in spite of having a European-influenced legal system and all legislative and parliamentary information available in the working language of English, making it one of the most accessible countries in Africa for research; in spite of having (currently) a comparable integrationist rather than refugee-camp policy and one of the busiest asylum systems in the world for almost a decade, it remains severely under-researched. My research excavates the presence of previously invisible women within one of the largest refugee systems and so re-frames their experiences and special vulnerabilities within the context of the most conspicuous movements of people in flight.  This research demonstrates, however, that within South Africa, sexual minority and polygynous refugee women have scarce and apparently fleeting presence. There are a variety of factors which play into their occlusion in terms of the gay community and ‘LGBTI’ spaces and their access, and conceptions of what                                                  17 A term attributed to writer and photographer Shahidul Alam. Along with Colin Hastings, Shahidul Alam co-founded a photography agency in 2004 named Majority World with the ambition of creating global market access for and highlighting the talents of photographers “from the majority world - Africa, Asia, Latin America and the Middle East”, see: “About Us – Majority World”, online: <>.   18 From the BRICS Information Centre: “Since 2008, the leaders of Brazil, Russia, India and China — the BRIC countries — have met annually to discuss issues of global significance. At their third summit in China in 2011, the leaders invited South Africa to join, thus becoming the BRICS. […]While the concept “BRICS” was first created by Jim O’Neill of Goldman Sachs to refer to the investment opportunities of the rising emerging economies, the leaders’ meetings transcend the financial context to embrace a wide range of summit-level issues relating to global governance, such as development, peace and security, energy and climate change, and social issues.” “About the BRICS and BRICS Information Centre”, online: <>. 12  marriage, gender, family and socio-economic contexts means with reference to the lives of refugees and asylum seekers. South Africa, like host countries of the West, has a disproportionately large number of young, male asylum seekers.19 Though we are told women make up half of those displaced across borders, these women remain ‘lost’ from studies focused on host states further away from conflict zones and so geographically and economically harder to reach for some of the most vulnerable and least mobile.20 Women refugees appear to be pooled in countries immediately neighboring conflict zones. As South Africa lies several countries to the South of such conflict zones, it serves rather as a bridging country to show how specific groups of women remain invisible though present in the system. South Africa’s male-prevalent asylum system, like those of most Western hosts, thus highlights the exaggerated need to engage in fine-grained legal analysis of refugee law elsewhere in the Global South, where it is applied to most of the world’s refugees, and specifically in the areas where women are most prevalent.  1.3.3 Population being studied  The title of my dissertation describes the populations which are centred as the focal point for my thesis. The populations are circumscribed as: sexual minority and polygynous refugee women. However, each aspect of these descriptors is contentious and requires further explanation and purposive use.                                                    19 Chief Directorate Asylum Seeker Management, Immigration Services, 2017 Asylum Trends Report (Department of Home Affairs, 2017) at 22.  20 UNHCR, supra note 9 at 61–62.  13 Refugees The population under study is described as refugees. Based on the UN Refugee Convention21 in which persons meeting the definition therein of a ‘refugee’ are, ab initio, refugees, and following on the work of Hathaway in this regard, the term ‘refugee’ is used here to describe persons who may fit the UN or AU Refugee Conventions’22 definition of a refugee rather than as a reference to the official recognition of their status.23   That is, a refugee is someone who:  owing to wellfounded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country; or who, not having a nationality and being outside the country of his former habitual residence as a result of such events, is unable or, owing to such fear, is unwilling to return to it.24 In the case of the AU Refugee Convention: The term “refugee” shall also apply to every person who, owing to external aggression, occupation, foreign domination or events seriously disturbing public order in either part or the whole of his country of origin or nationality, is compelled to leave his place of habitual residence in order to seek refuge in another place outside his country of origin or nationality.25                                                  21 UNHCR United Nations, “Convention Relating to the Status of Refugees”, online: Refworld <>. United Nations Treaty Series vol.189 (“UN Refugee Convention”), at 137. 22 OAU, “OAU Convention Governing the Specific Aspects of Refugee Problems in Africa, adopted by the Assembly of Heads of State and Government at its Sixth Ordinary Session, Addis-Ababa, 10 September 1969”, online: UNHCR <>. (the “AU Refugee Convention”). 23 James C Hathaway & Michelle Foster, The Law of Refugee Status, 2d ed (Cambridge: Cambridge University Press, 2014) at 1.  24 UN Refugee Convention Article 1(A)(2) 25 AU Refugee Convention Article 1(2) 14  Thus an ‘unrecognised’ refugee or a refugee without government or UNHCR26 status, is still a refugee for this purpose. Although the word is never used in either the UN or AU Refugee Conventions, I take the notion of fleeing as central to a practical and humanitarian understanding of the term ‘refugee’. The word ‘fleeing’ captures the sense of fear which compels movement, without requiring elaboration of the complex reasons therefor, and includes those “outside of their country of origin” (the more neutral term used by the Conventions) and those in the troublesome in-between-ness of borders, passport control, bridges, boats and planes. Women The second common denominator of the groups focused upon in this dissertation is that they are women. As this study will focus upon the understanding of these women amongst refugee administrators, lawyers and advocacy groups, their designation as such by these actors will be taken as the starting point. I must acknowledge that this approach may not reflect the gender as self-identified by the persons involved and so this approach may replicate the violence of such mischaracterisation. Nevertheless, the service providers’ own understanding of gender is an aspect of this study’s enquiry. A focus on women is undertaken as a feminist project in opposition to a male-centric tendency evident in refugee law and refugee law studies, in LGBT studies and in approaches to polygynous families. A destabilization of presumptive gender therefore requires an inclusive understanding of ‘women’ which canvasses all those understood                                                  26 The United Nations High Commission for Refugees, (“UNHCR”) administers or plays an advisory role in refugee status determination procedures in many countries. United Nations High Commissioner for UNHCR, “Refugee Status Determination”, online: UNHCR <>.: “In any given year, UNHCR conducts RSD under its mandate in between 50-60 countries, depending on where the applications are received. […] In approximately 20 countries UNHCR conducts RSD jointly with the government pending the state assuming full responsibility for RSD, while in many more countries UNHCR conducts a range of capacity development activities.” 15  by the author, by the service providers, by the officials and by individual refugees themselves as identifying or identified as women. The remaining adjectives of the groups under study refer to women who, in their relationships, or identification of desire, defy the monogamous, hetero-normative family most frequently understood as the default family unit. This presumptive family forms the locus around which notions of refugee women are constructed.27 Yet these descriptors: ‘sexual minority’ and ‘polygynous’, too have their limitations, discussed below. Sexual minority and queer terminology In this dissertation, refugees for whom sexual orientation-based persecution is a real fear, are most frequently termed sexual minority refugees and queer refugees. The term ‘sexual minority’ was chosen as both demonstrative of the minority status-challenges peculiar to this group of refugees, and as a more fluid term than labels such as ‘lesbian’ and ‘bisexual’. Sexual minority terminology is arguably less culturally specific whereas the term ‘lesbian’ is criticised, along with the conglomeration LGBT, as being a Western-originated term, which fails to capture the self-identity of those who may be included in this term, by Western researchers.28 However, these terms are also gaining acceptance and being taken up from the ground by developing world                                                  27 Thomas Spijkerboer, Gender and Refugee Status (Routledge, 2017) at 103.: “In the asylum procedure, women are seen as related, as mothers and as wives. Women are not so much relegated to the family in the normative sense that "women should be good mothers and wives". Women are, rather, considered inherently part of the family; when they behave contrary to the expectations of decision makers, the conclusion is not that the applicant is not a good mother and wife; instead, what the applicant says cannot be true.” 28 Oliver Phillips, “Constituting the Global Gay: Issues of individual subjectivity and sexuality in southern Africa” in Didi Herman & Carl F Stychin, eds, Sexuality in the legal arena (London: Athlone, 2000) at 18.  16  queer persons, especially by those advocating for greater rights.29 Furthermore ‘sexual minority’ is subject to the same charge of a preoccupation with the sexualisation of identity as many of the alternatives.  Whilst sexual minority was initially conceptualised in this project as a term easily demonstrative of the refugee law sexual orientation-based persecution link and served to parse this population from a trans refugee population with an anticipated gender identity-based persecution link, the resulting study called into question this distinction. As discussed further in Chapter 5, sexual-orientation based persecution appears to be both salient in transgender refugees’ experience of persecution, for reasons of their imputed sexuality, and aligns with the grounds for refugee status that may be used in these persons’ asylum applications, in South African.30 A more complete definition of the group being sought out in the data is therefore: ‘women’, identified as such by the system, the persons administering it or the persons providing legal assistance, for whom same-sex sexuality or perceived same-sex sexuality is understood as an aspect of their flight. The interviews conducted with refugees, which form part of the data analysed in this dissertation, were with six31 male-identified same-sex oriented persons and a transgender woman. Thus the broader term ‘queer’ is predominantly used to capture a gender non-binary populace. The inclusive and simple term ‘queer’, is useful for this project in its being politically infused as a                                                  29 See the chapters by both Phillips, supra note 28; Jennifer Spruill, “Post-with/out a Past? Sexual orientation and the post-colonial ‘moment’ in South Africa” in Didi Herman & Carl F Stychin, eds, Sexuality in the legal arena (London: Athlone, 2000) at 5.  30 B Camminga, Transgender Refugees and the Imagined South Africa: Bodies Over Borders and Borders Over Bodies, Global Queer Politics (Cham: Springer International Publishing, 2019) at 109.  31 Although I include in this number “Noah” whose identification is complicated by a quiet remark in passing that he was “also a woman”; see the discussion in Chapter 5.4.1. Noah, Interview with Anonymous Refugee: pseudonym “Noah” (2016), Cape Town. The interviewees’ self-identification, including choice of sexual orientation terminology is thoroughly discussed in Chapter 5.4. 17  challenge to normative productions of sexuality and identity. It is for this reason used regularly herein. Polygyny I have chosen to use the term ‘polygyny’: one man with more than one wife – over the more common ‘polygamy’: more than one spouse. In spite of the fact that polygyny is the lesser known term and is thus less accessible, and contrary to most literature which refers to polygamy even where only polygyny is actually discussed, I consistently use the term polygyny. The concern in this study, is specifically with the plural wives of polygynous marriages and the gender neutral approach which refers to ‘spouses’ or even ‘polygamy’ fails to sufficiently emphasize the gendered nature of the practice and of its prohibition. Although polygyny is a lesser known term than polygamy, in this regard the difference in terminology is essential for the advocacy position of this research. I could also easily explain my use of the term by raising the comparative definitions of the terms polygamy and polygyny to interviewees.32  The more complicated aspect of the term ‘polygynous women’ relates to the underlying relationship and its legal status. The understanding of polygny, as is used in this thesis, includes those women in both de facto and de lege marriages (in terms of the laws of their country of origin, or the country in which the marriage was celebrated, where applicable). In a discussion on polygynous refugees, and most especially where children are involved, whether the underlying                                                  32 Many service providers interviewed were indeed unfamiliar with the term ‘polygyny’ and my use of this term in place of ‘polygamy’ prompted a useful discussion on this gendered aspect of the practice with at least one such interviewee, others continued to use the term ‘polygamy’ in our interview whereas some took up the use of ‘polygyny’ themselves in their response; David Cote, Interview with David Cote, Strategic Litigation Programme Coordinator, LHR (2016), Pretoria.  18  marital relationship is ‘common law’, de facto or fully legally compliant and registered, or a traditional or formal marriage is not really relevant to their protection and unification needs and would seem an improper distinction to make in data collection. An expanded notion of marriage and a looser legal understanding of ‘spouse’ seems more in keeping with an equality-minded trajectory.  Ultimately, my choice to use the term polygyny is influenced by the same lines of thought as my choice to use the terms sexual minority and queer. I am not using the most common or accessible terms but rather the most academic word choice, for reasons of the politics and advocacy role behind the choice. These terminology choices reflect an attempt to use appropriately gendered language.  In the case of both sexual minority and polygynous women, their experiences are foreshadowed by some of the contexts and conceptual stumbling blocks identified in this dissertation but their voices are not represented here. Discussed further below under 1.3.5, this dissertation uses silence around these women, rather than representations of these women’s voices, to indicate the spaces where these women should be but from which they appear absent. The populations under study in the South African asylum system will be circumscribed to those entering South Africa, or with a partner in South Africa, subsequent to the commencement of the Refugees Act in South Africa on April 1, 2000, which states in the definitions section thereof that: “‘social group’ includes, among others, a group of persons of particular gender, sexual orientation, disability, class or caste”.33 Section 3(c) of the Refugees Act includes in the definition of a person qualifying for refugee status, the dependant of a refugee which is defined                                                  33 Government Gazette, “Refugees Act 1998”, online: <>. Act 130 of 1998 (hereafter “the Refugees Act”). 19  in the original Refugees Act as “the spouse, any unmarried dependent child or any destitute, aged or infirm member of the family of such asylum seeker or refugee”.34 An amendment to the Refugees Act, which was assented to in 2008 but only came into force in 2020, creates a new definition for both “spouse” and “marriage” that is inclusive of both same sex and plural (polygynous) partnerships.35  1.3.4 Participants There are in fact two different ‘populations’ studied; the refugee women around whom I attempt to centre my thesis, and also those who interact with them as they navigate the asylum system. In determining the understanding and approach of the latter towards the former, the participants in my study include those who provide services to refugees in South Africa; the lawyers who specialise in refugee assistance, including the Law Clinics of various South African universities; and some of the key refugee non-governmental organisations (NGOs), researchers and refugee-community advocates who are broadly referred to as ‘refugee service providers’ in this work.36                                                   34 The Refugees Act, section 1. 35 Department of Home Affairs, Refugees Amendment Act, 2003, 33 of 2008 (Government Gazette, Vol .512 No.31643, 2008). (hereafter “the Amendment Act”).  36 I had initially intended to include interviews from the South African government officials working with the refugee system: the Refugee Status Determination Officers (RSDOs), where possible, and the Refugee Appeal Board adjudicators. I wasn’t able to include any of these officials, they simply don’t seem to have a response rate which would have fitted in my timeline and I suspect require connections I did not have available. There are insufficient RSDOs to handle the workload and appalling backlog in the South African system and, as discussed further in this thesis, the Refugee Appeal Board had too few members to be officially constituted, contributing to the impracticality of interviewing these players in the course of my fieldwork. 20 Refugee service provider participants This study was conducted in accordance with UBC ethics protocols, as noted in the Preface. Initial contact was made via email, with an attached Initial Letter of Contact and a Participant Consent Form.37 All refugee service providers were given the option of remaining anonymous in this work or of using their names. Refugee participants however, discussed further below, were all anonymous bar one. Service providers were interviewed based on their own experiences and personal views and were not asked to represent the views of the organisations for which they work. Kizitos Okisai, a UNHCR resettlement officer, agreed to the use of his name specifically on the understanding that his views did not represent those of the UNHCR. The contact information for relevant individuals at the Law Clinics, Law firms, refugee advocacy groups and research institutes is publicly available online. I drew up a list of relevant refugee service providers in South Africa, focusing on the hubs of Johannesburg, Pretoria and Cape Town, based on my previous experience with the South African asylum system and the contact list available at the South African Pro Bono Directory of the International Refugee Rights Initiative, Rights in Exile Programme.38 With regard to primary data collection, I began my fieldwork by engaging with ‘key informants’, those working at my South African alma mater university’s Wits Law Clinic and Migration departments, to attain updated information on the South African asylum system, common concerns, and contacts for the persons engaged therein. Simultaneous data collection and superficial thematic analysis as well as onward referrals allowed for progressive growth in the project, informing further interviews organically as I proceeded.                                                  37 See BREB approved sample letters in Appendix A. 38 Available at:  21  Significant resource groups of whom I was already aware and contacted prior to my arrival for field research in South Africa, included the Refugee Unit of the Wits Law Clinic39, the project examining gender and asylum run by Ingrid Palmary in the African Centre for Migration & Society through the Forced Migration Studies Programme (FMSP),40 the multiple offices of Lawyers for Human Rights (LHR) and a high profile, local advocacy group for refugee rights PASSOP (People Against Suffering, Oppression and Poverty: “a not-for-profit human rights organisation devoted to fighting for the rights of asylum-seekers, refugees and immigrants in South Africa”) 41; particularly the LGBTI, Gender and Refugee Reception Centre Monitoring projects thereof, as well as the UNHCR offices in South Africa.42 Further research and referrals by interviewees grew this list to include Amnesty International’s Johannesburg office, the Jesuit Refugee Services (JRS), Future Families, the Coalition for African Lesbians (CAL), and Holy Trinity, a Catholic Church providing support for refugees in Johannesburg. The research was conducted in English. All professional interviewees were expected to understand English well, being the primary working language of South Africa, and those service providers who referred refugees for interviews also relied on English in their communications and all refugee participants were able to understand and communicate effectively in English.  I designed my own interview script, based upon some broad theme direction and open-ended discussion points. My project required and received Behavioural Research Ethics Full Board approval; a Full Board was required given the vulnerability of especially the refugee                                                  39 “Refugee Unit - Wits University”, online: <>. 40 “African Centre for Migration and Society - Wits University”, online: <>. 41 “PASSOP”, online: <>. 42 United Nations High Commissioner for Refugees UNHCR, “South Africa”, online: UNHCR <>. 22  participants.43 The length of interviews varied depending on the comfort and interest of interviewees, generally around half an hour to 45 minutes though some service provider interviews were longer and some refugee interviews were much shorter. I interviewed service providers: lawyers and community advocates, to gain insight into their awareness of the women refugees under study, their experiences in dealing with their cases, or lack thereof, their understanding of the legal concerns specific to these women and related to family (derivative) refugee status generally and their awareness or otherwise of legislation on the books which could affect these women, should it be brought into force. It should be noted that at least three of the service providers interviewed could themselves be classified as refugees, as that term is primarily used in this paper, to refer to de facto refugees. Refugee participants I sought out refugee participants indirectly through referral by their service providers, mindful of the ongoing homophobic danger faced by queer refugees and the social ostracization I expected polygynous refugee women may face. I initially expected that referrals would come from lawyers aware, through their interactions with their clients or the nature of the clients’ asylum cases, of their clients’ polygynous family form or of their sexuality or transgender identity. However, all referrals came through queer support group service providers. Due to privacy and security concerns for the refugees, the offices of their service providers upon which they would normally attend, was the preferred and suggested location for interviews with refugees. At the                                                  43 The University of British Columbia has a Full Board review process and a simplified Minimal Risk process of the Behavioural Research Ethics Board, see “BREB FAQs | Office of Research Ethics”, online: <>.  23  refugee participants’ discretion, I indeed conducted all interviews at PASSOP’s office with one phone interview facilitated through Dumisane Dube, the coordinator of an LGBT programme of Holy Trinity. As discussed in the data analysis of this dissertation, many service providers had little knowledge of or interaction with polygynous clients or queer, especially queer women, clients. The lawyer with polygynous refugee clients and an advocate who had worked with and previously interviewed a queer woman refugee both offered to try and get me access to these clients’ redacted files, however, the necessary consents were not forthcoming. In this regard, LGBT support groups were perhaps the standout service providers who had continued and regular contact with their clients, necessary for obtaining consents or interview opportunities, whereas lawyers and other service providers had ad hoc interactions and imperfect communication with their clients. Services directed to the particular needs of queer refugees are few and there is no such provision of services targeting a polygynous refugee community. I therefore sought to identify the interactions of these groups of refugees within both targeted service providers and the major providers of services to refugees. Tellingly the multiple service providers who discussed a particular lesbian refugee couple, who appeared to have sought out an unusual variety of supports, had lost contact with the couple who had fled the country. The scant interview opportunities I had with these groups of refugees are a reflection of their absence (and fleeting presence) in these spaces. It is therefore no coincidence that the majority of refugee interviews were facilitated through the LGBT Coordinator of PASSOP, the most prominent service provider for queer refugees and, as revealed by a UNHCR resettlement officer, the key strategic partner in generating better access to the queer refugee population, for the UNHCR. 24  I asked the service providers to pass on the letter of initial contact and consent letters, specific to refugee participants, together, via their normal means of communication with their clients ahead of the interview. I again provided these letters in person and discussed them at the commencement of our interviews, with all refugee participants bar the phone call interview, in which confidentiality, consent to participate and the interviewee’s comfort and safety in continuing the interview were discussed verbally only.  I informed all refugees that they would be anonymous in the reproduction of their interviews and in the collection and storage of the interview data. I have given pseudonyms to all but one refugee. In the process of interviewing one refugee, both in the discussion of the unique and high profile details of her story and her request to pass on her thanks to organisations she listed, it became apparent that either her story would have to be substantially edited or her anonymity lifted. Subsequent to the interview, this refugee, Tiwonge Chimbalanga, confirmed that she wished to be named in this research.  The pseudonyms I chose for the other anonymous queer refugees who interviewed with me – six in total, all men – are based on the boys names listed in the ‘top US baby names’ for 2016, found online at Baby Center.44 Many of those I spoke with, if they gave me their names, did actually have, or at least were using, anglicized names. Yet my choice to use this American baby name list purposefully marks the disjuncture in time and place of my representivity of their voices. These pseudonyms were ‘born’ with my brief encounter with these persons, in 2016, and the                                                  44 “Most popular baby names of 2016 | BabyCenter”, online: <>.. When discussing refugee stories, the most common approaches to naming the anonymous are to either revert to a ‘Mr X, Mrs Y’ trope – which to me comes across as a law school-style depersonalization of personal stories – or to give ‘common’ (stereotypical) country-of-origin based names. I liked Spijkerboer’s decision to use familiar names for foreigners with the intention of interrupting the reader and pointing to the alienating effect of the process in which “the applicant we see in the asylum procedure is a [in his case] Dutch product”. Spijkerboer, supra note 27 at 45.  25  popularized, American nature of these names marks my position both as white and studying in a North American university. Though I am also South African, many of those I interviewed assumed I was foreign as they knew I was coming from a Canadian university, was in South Africa for the purposes of my research and my relatively neutral accent combined with my whiteness served to mask my own ‘country of origin’. When writing my thesis and incorporating quotes from my interviews with these refugees, the forced nature of these ‘unfitting’ names served to remind me that I was not giving these people voice, but rather appropriating their voice. Nevertheless, I also struggled when footnoting the quotes with the pseudonyms I had allocated, as the names made these refugees’ voices less distinct in my head and it was harder with the pseudonyms, writing several years after our meeting, to remember their faces and to know, without referencing my notes, which stories belonged to whom.  1.3.5 Matrix for analysis Using both quantitative and qualitative aspects for analysis and cross-reference, I interpreted my data through an iterative, inductive approach where concepts and analytical themes emerge from the data and are then mapped across interview transcripts, policy and legal documents and the academic literature.45 The literature, reviewed in Chapter 2 below, was gathered through electronic university library and scholarly search engine key word searches, through consultation with my supervisory committee and other faculty members, as well as a snowballing of cross-referenced works within the literature.  An obvious shortcoming of this work is that I reviewed                                                  45Leila Ullrich, “‘But what about men?’ Gender disquiet in international criminal justice:” (2019) Theoretical Criminology, online: <>. 26  literature in English only; and focused on the library collections of a North American university (including material that is available through interlibrary loan). The brief quantitative aspect of my study explored existing official data to determine whether my populations under study officially exist on paper and contextualising their possible statistical presence. South African official data on asylum applications was aggregated only with respect to acceptance and rejection numbers, by country and region of origin and by city of application. Gender and country of origin statistics and an age breakdown were also available for some years.46 This aspect of the project was undertaken not as an attempt to create objective or reliable data but rather expecting absences and unreliability (discussed further in Chapter 3.2). The purpose of the quantitative element of my study is to make use of the official statistics as an indicator of the population groups’ (un)importance and treatment in the system. As identified by Merry, official records are themselves “cultural documents”, which indicate the (in)significance of issues to those with power.47 In the qualitative aspect of my study, I created data from interview analysis to compare and triangulate with the above quantitative initiative. I therefore inquired about service providers’ knowledge of sex orientation-based persecution as a grounds for refugee status and possible links to gender-based persecution and also questioned service providers as to the possibility/usage of same-sex derivative status (refugee status by reason of being a dependent, same-sex spouse of a refugee) and of plural spouse derivative status (refugee status by reason of being one of a refugee’s dependant spouses). Finally, refugee participants were sought out amongst interviewed                                                  46 See Chief Directorate Asylum Seeker Management, Asylum Statistics: Year 2013 (2014); Home Affairs, 2015 Asylum Statistics: Analysis and Trends for the Period January to December, Presentation to the Portfolio Committee of Home Affairs (2016); Asylum Seeker Management, Immigration Services, supra note 19. 47 Merry, supra note 12 at 107.  27  service providers’ clientele who could speak to sexual orientation-based persecution in relation to their flight or polygynous family structures. These refugees expounded on their interactions with the asylum system and provided narratives relating to the role of their gender, sexuality and non-normative family structures in their search for asylum and in their interactions with officials, lawyers and advocates. The thematic analysis of the interviews was undertaken partly through themes expected and constructed prior to the field research being conducted, which were integral to the framing of this thesis: discussions on women refugees and the relativity of their presence, opinions on statistical data on asylum seekers and refugees and on the changing and uncertain legal landscape for refugees in South Africa. This data was thus prompted by the structuring of the questions asked. In creating an interview script, prior to conducting fieldwork, the main topics on which participant’s views and experiences were sought included:   asylum claims by women, (estimate of proportions) and of sexual minority and polygynous women in particular;  asylum claims grounded on sexual-orientation based persecution or gender persecution;  major issues and challenges in women’s asylum claims, and of sexual minority and polygynous women in particular, identification of cases of significance and developments;  institutional barriers to claims and to family reunification of sexual minority and polygynous women; 28   the role of legislation in the above and awareness of derivative status (refugee status of dependants) in South Africa and presumptions of gender and sexuality;   the understanding and/or use of a broadened definition of “spouse” by lawyers, advocates and bureaucrats and awareness of and attitudes towards the Refugees Amendment Acts of 2008 and 2011. Some of the anticipated themes fell away as a focus in the process of conducting interviews. So, for instance, a focus on family reunification and on ‘derivative status’ – the use of section 3(c) of the Refugees Act to grant refugee status to the dependants of refugees – proved not to be fruitful given the level of disarray of the data and systems of the Department of Home Affairs (DHA) and particularly given the problems with file joinder and the apparent minority of queer refugees with partnerships both predating and extant subsequent to their arrival in South Africa. Other themes emerged organically, and even unexpectedly, as I conducted more interviews and in subsequent analysis, through thematic pattern recognition. This included a reconsideration of the conceptions of ‘family’ used, in place of a ‘family reunification’ focus, and an apparent ‘Somali theme’ emerging relative to polygyny, all of which are analysed in the discussions to follow. The predominance of male refugees in the South African asylum system ground the expectation of smaller numbers of polygynous women and queer women refugees. However, whilst many service providers seemed to consider polygyny as likely absent within the refugee population (discussed further in Chapter 4), most felt queer women should be present. A discussion on the theories that service providers suggested for their absence is raised in Chapter 6.  This doctoral research produced a very small number of descriptions of encounters with polygynous refugee women and with sexual minority refugee women by service providers 29  interviewed. There were also no refugee interviews conducted with polygynous refugees, only with queer refugees, all of whom, bar one, were male.  Where gaps and missing voices become apparent, it can be difficult to discern whether such gaps are the result of limitations within the research or an observation of the research. I argue that these groups of women who are only scarcely alluded to by refugee service providers and whose absence is largely replicated in this research, which purposely set out to explore an understanding of their interaction with the asylum system, is itself an important finding, signifying exclusionary spaces and conceptions. Relevant parallel discussions on the attempt to study what is not there, can be found in discourse studies. Zerubavel notes the reinforcing nature of phenomena of absence and their study: “what we ignore or avoid socially is often also ignored or avoided academically, and conspiracies of silence are therefore still a somewhat undertheorized as well as understudied phenomenon. Furthermore, they typically consist of nonoccurrences, which, by definition, are rather difficult to observe.”48 Yet Schröter and Taylor urge: “However, this should not lead us to erase absence from any research agenda when we can still use what is present to look at what constitutes, produces or indicates absence”.49 In a volume that seeks to encourage silence and absence as a line of enquiry in discourse studies, the editors, Schröter and Taylor, suggest approaches to the empirical analysis of absences with the intention of creating a methodological toolkit.50 The authors assert: “we only need to be                                                  48 Eviatar Zerubavel, The Elephant in the Room: Silence and Denial in Everyday Life (Cary, UNITED STATES: Oxford University Press USA - OSO, 2006) at 13. 49 Melani Schröter & Charlotte Taylor, Exploring Silence and Absence in Discourse: Empirical Approaches, Postdisciplinary Studies in Discourse (Cham: Springer International Publishing, 2018) at 14. 50 Ibid at 2–5. 30  concerned with meaningful absences and that for absences to be meaningful, they require an arguable alternative of presence.”51  The minimal presence of the women in this study, discernible within the asylum system and amongst those who provide services to refugees, indicates not just a failure to provide protection to these particular groups but presents a lens through which to better perceive significant challenges embedded in our understanding of international refugee law. Schröter and Taylor, amongst their methodological approaches suggested for studying meaningful silence and absence in discourse, include ‘comparison’ as a means to identify absence. “If we accept that meaningful absences require a possible presence against which the absence can be identified, then comparison seems to be a good way to locate absences.”52 This dissertation exposes the meaningful silences around polygynous refugee women and sexual minority refugee women through political-geographical comparison; in which Western-focused traditions of refugee law studies and their emergent themes can be compared with the findings of my differently located research. A gendered comparison is also used; considering male presences (including polygynous men and queer men) and extrapolating female absences. This is by no means a comprehensive account of the presence of these groups of women in South Africa and provides very little on the interaction of these women with the asylum system. Yet the work that remains to be done supports my primary call for greater, high detail refugee law research in South Africa and other under-studied host countries. If further research on this topic were to be conducted, approaching women’s violence centres, rather than queer or refugee                                                  51 Ibid at 6.  52 Ibid at 14. 31  focused groups, may prove more fruitful in accessing some missing voices (as theorized in Chapter 6.2).   1.4 Principal Conclusions This dissertation, which sought to forefront a refugee host country deserving of more attention in refugee law studies as well as to centre particularly vulnerable refugee women largely sidelined in the study of the workings of refugee law and asylum systems of the world, produced minimal data on women polygynous or queer refugee clients of service providers. This research includes no polygynous or sexual minority women refugee interviews, and one (trans)woman refugee interview. While I did not find these women, I did find some reasons for their invisibility. Within South Africa, systemic incompetence and the purposeful blocking of access to the asylum system negatively impacts all refugees, largely denuding the refugee protections guaranteed by international obligations and domestic legislation. This programmatic anti-refugee stance evident throughout the asylum system has particular and sometimes exacerbated repercussions for polygynous and queer women with compounded access and resource limitations and gendered security concerns. Identifying these women and their needs is made more complex, for both service providers and researchers, by virtue of their likely being largely outside of the asylum system. The rendering of the South African asylum system as unworkable is demonstrative of different ways of excluding compared to wealthier Western states with harder-to-access borders and greater border control. As the exclusion from refugee protection is a central topic for refugee law studies, investigating these alternative, systemic ways of excluding is necessary and thus requires an inclusive approach to studying a greater variety of host countries.  32  Polygynous and sexual minority refugee women are furthermore hidden from view as a consequence of service provider expectations echoing and influenced by Western stereotypes of the refugee figure and of polygamy and queer identities. More than just a commonality of approaches, the impact of Western tropes and knowledge production is particularly evidenced by ‘out of place’ expectations such as typifying polygyny, a domestic and regional African customary practice as a Somali, Islamic practice and a lack of regard for complex queer families whereas reports of queer persons with heterosexual marriages and children are not uncommon in Africa. Both the analysis on polygynous and queer refugee women work to demonstrate the knock-on effects for misunderstanding lived experiences in light of a refugee figure constructed, familiarly, as a static identity, characterised as single, male and poor, but whose persecution is artificially extracted from their poverty. This dissertation identifies insights applicable to the conceptualisation and study of refugee law generally, yet which can be attributed to the perspective attained from the study of a non-traditional ‘host’. This includes the urgency of a need to re-examine the role of economics in understanding experiences of and vulnerability to persecution, including continuing persecution, and the interplay of mobility challenges and identity politics limitations which disproportionately obscure some of the most vulnerable people, most disadvantaged by a narrow focus on persecuted identity, from a Western-host vantage point. The challenges polygynous and queer women face as refugees are lost in translation in the extraction and abbreviation required by the West-as-host construction, discussed in this thesis, the consequences of which are amplified through the politics of knowledge production. Having outlined the schema of this dissertation and the methodology informing the conduct, analyses and presentation of the research presented herein, the following chapter outlines the 33  varied and multidisciplinary scholarly work which has informed the ideas underlying the content and purpose of this research as well as foreshadowing its contribution.   34  Chapter 2: Literature Review  The launch pad for this work and the discussion engendered is provided by the existing literature in refugee law; the gaps identified therein have provoked the need for this research and the cumulative insights of many authors have informed my thematic analysis. This research has also been nourished by the perspectives and understandings built across other fields of study. A comprehensive review of refugee law literature, even as it pertains to particular groups of refugees is not within the scope of this dissertation, instead this serves as a selective review of the most influential works in refugee law generally as well as those specifically relevant to this dissertations’ focus and points towards cross-over opportunities for learning from other areas and disciplines. Reading exclusively in English limits my reach in this review, in particular in relation to European-based research which is only available in other languages. Nonetheless, on the basis of my discussions with my supervisory committee, my focus on influential works, which influence is a product of a substantial English bias, and my efforts to track references suggested by various texts, I am confident that this review represents the current state of the refugee law literature.  This literature review begins with a survey of a provocative gap in refugee law scholarship which is brought about by what I identify as recurring instances of an unacknowledged bias in focus. I move on to discuss the insights gleaned from within a Western-predominant, Southern occluding study of the application of refugee law which have nevertheless inspired and found reflections in this dissertation’s analysis. The existent, Southern-focused refugee law studies as well as signals of new developments within refugee law are discussed and finally, the significant and valuable contributions of a multidisciplinary approach to knowledge are canvassed.  35  2.1 The Western Bias in Refugee Law Scholarship The canonical literature in refugee law studies is represented by authors whose general and largely descriptive works on Refugee Law are used today as reference works in asylum cases across the world, such as Hathaway’s second edition, together with Foster, of The Law of Refugee Status and Goodwin-Gill and McAdam’s update of The Refugee in International Law.53 In the case of the latter, the authors’ stated aim is “to indicate with some precision the fundamental interests which must be protected as a matter of law, if the inherent worth and dignity of the individual in flight are to be upheld”, with particular focus on the UNHCR positions on the topics canvassed. This book also includes a discussion of the European Union harmonization initiatives for their potential to influence states beyond the EU and thus provide some indicators of possible future directions for international refugee law.54 Hathaway and Foster, on the other hand, make extensive use of case law in outlining both best practice and errors in the application of refugee law. This approach allows for consideration of the interesting transnational judicial dialogue, as well as the encouraging impact of academics amongst asylum case decision-makers. However, this focus also illustrates a dismissal of the host states of the Global South. South Africa, with a handful of cases referred to, represents the only developing country jurisprudence reflected in the book. Although the authors acknowledge a common law as opposed to civil law bias, they do not mention this greater geographic imbalance. Alternative, general approaches with an examination of the ethics of refugee law are proffered by authors such as Price, Haddad, Gibney and Chimni (discussed further under 2.3 below). Gibney discusses ethical political theories underlying different arguments relating to state obligations                                                  53 Hathaway & Foster, supra note 23.; Guy S Goodwin-Gill & Jane McAdam, The refugee in international law, 3rd ed ed (Oxford ; Toronto: Oxford University Press, 2007). 54 Goodwin-Gill & McAdam, supra note 53 at 60. 36  towards refugees in the context of what he conceives of as competing obligations to citizens. Gibney reveals the way in which state structures create a rational self-preservation interest in attending to the interests of insiders and that “most states are responsive to the needs of outsiders in extremis, they do not equate the needs of outsiders with the needs of citizens. When states accept refugees, they are responding to extremity, not equality”.55 Gibney thus seeks to balance ethical force with practical relevance so as to argue for politically achievable improvements to asylum processes. Gibney describes this principle as equating to a state obligation to refugees (anywhere) as long as the costs are low. His petition is directed towards Western host states and their ‘organized hypocrisy’ in increasing their efforts to prevent the entry of asylum seekers whilst continuing to proclaim the importance of the refugee regime. Price, on the other hand,  proposes that the problems of the blunt instruments increasingly used by states to deny access to would-be asylum seekers, such as unjust fast-tracked processes and increasing limitations on the rights of those that are admitted, can be resolved by ‘returning’ to a more blatantly politicized and much more restricted conception of asylum. This, he states, will more closely reflect states’ interests and so engender greater support and the provision of permanent settlement and fuller rights as the preferred solution. It is Price’s contention that there has been a ‘humanitarian turn’ in refugee law which has moved it away from its Cold War political conception as “connected to a broader political program to reform those [abusive] states”. Price bemoans the ‘unjustifiable expansion’ of protection mechanisms, decoupled from foreign policy, which do not reflect his proposed, very narrowly political understanding of ‘persecution’.56 An opposite approach is                                                  55 Matthew J Gibney, The Ethics and Politics of Asylum: Liberal Democracy and the Response to Refugees (Cambridge, UNITED KINGDOM: Cambridge University Press, 2004) at 212. 56 ‘Persecution’ on Price’s reading  (see Matthew E Price, Rethinking Asylum: History, Purpose, and Limits (Cambridge: Cambridge University Press, 2009) at 70.) occurs where persons are “targeted for harm in a manner that repudiates their claim to political membership”. His conception of refugees is constituted solely by political 37  taken by many other writers including Anker, who celebrates the significance of the justiciability and enforceability of refugee law, and its palliative rather than state-responsibility focused nature, as potentially providing focus to and an elaboration of issues at the forefront of human rights law, without facing similar challenges of state-consensus.57   Haddad argues that the refugee is inevitable and so ‘natural’ to the state system, whereas the refugee is imagined as “a moving, exceptional figure” compared to “the otherwise ‘normal’, sedentary state-citizen-territory trinity”.58 She identifies this disruption of the “national order of things” by the constructed refugee, as the reasons the refugee appears as threatening, prompting a security, rather than humanitarian based, state reaction.59 Haddad criticises the Convention grounds of persecution for focusing on problems apparently internal to states and rather points to the international context in which the refugee emerges. The real ‘problem’ therefore, on Haddad’s analysis, is the state system itself and the concept of sovereignty on which it relies. Refugees are thus “the human reminder of the failings of modern international society”.60  All of the above authors purport to engage in ‘international’ refugee law studies whilst predominantly reflecting Western host-only refugee law analysis. Chimni, on the other hand, works to expose elements of this bias in his work. Chimni is discussed further herein as a significant TWAIL, refugee law scholar. He particularly exposes and criticises the positivist tendency in refugee law studies. Chimni characterizes the depoliticized academic and advocacy focus on improving refugee determination procedures, during the Cold War, as creating a legacy                                                                                                                                                              exiles and consequently the necessary and distinctive solution required is (permanent) surrogate political membership.   57 Deborah E Anker, “Refugee Law, Gender, and the Human Rights Paradigm”, (5 July 2017), online: International Refugee Law <> at 138.  58 Emma Haddad, The Refugee in International Society: Between Sovereigns (Cambridge, UNITED KINGDOM: Cambridge University Press, 2008) at 47.  59Ibid at 90. 60 Ibid at 136. 38  of legal scholarship which disarmed itself when it came to political and ethical questioning of refugee policies. The missed role of interpretation and power dynamics, Chimni believes, led to a false assumption of the protective nature of refugee law, even in a changed historical context. Chimni describes the subsequent paradigm shift in international refugee policy as founded upon the creation of a myth of difference which itself relies upon the construction of the ‘normal’ refugee; as white, male and anti-communist, against which those fleeing from the Third World could be sharply contrasted. Like Haddad, Chimni also points to an “internalist interpretation of the root causes of refugee flows” – blaming the country of origin, without consideration for external factors and historical influences – as an element of the worldview which promotes a new, more restrictive approach.61 Comparative country jurisprudence discussed by some of the leading academics in Refugee Law, is commonly exclusively from developed countries. Hathaway and Foster’s book canvassing Western refugee case law, discussed above, is one such example as is Aleinikoff’s chapter on the meaning of ‘particular social group’ in the UN Refugee Convention. Aleinikoff, who subsequent to writing the chapter became Deputy High Commissioner of the UNHCR, usefully summarizes state jurisprudence, discussing the country case law on the protected characteristic approach compared with the social perceptions approach to defining membership of a ‘particular social group’ (PSG); Aleinikoff identifies these approaches as being focused on internal factors and anti-discrimination principles, and external factors and sociological understandings respectively (see further below on Foster under 2.2). Yet the jurisprudence he discusses is exclusively                                                  61B S Chimni, “The Geopolitics of Refugee Studies: A View from the South” (1998) 11:4 J Refug Stud 350–374 at 351. 39  Western.62 Gammeltoft-Hansen’s book, Access to Asylum: International Refugee Law and the Globalisation of Migration Control focuses on the international aspects of refugee law and forces of globalisation in a thorough excavation of modern exclusionary practices of developed (termed ‘traditional’) host states.63 It is assumed throughout that the ‘state’ being encountered is a developed, Western state, and the exclusion of asylum seekers is viewed from a Western entry vantage point only. There is therefore no discussion of the exclusionary mechanisms of Southern host states which differently, yet effectively deny entry to their asylum systems (though not their borders), as is discussed in Chapter 3 of this dissertation. Southern hosts, and the important and distinct challenges refugees may face therein, thus appear forgotten in much Western-host state focused refugee studies. In outlining the various forms of persecution, direct and indirect, faced by gay asylum seekers in their country of origin, Hathaway and Pobjoy state that it is no surprise that they have “increasingly sought the protection of Northern states that take a more sympathetic view of gay rights.”64 However, this phrasing omits the presence of LGBT asylum seekers in Southern countries who, whilst seeking protection, are simply ‘unseen’ and may well remain in the South, facing persecution in their host country. Furthermore, a “sympathetic view of gay rights”, whilst indeed a potential “pull-factor” for some host countries, does not necessarily result in a safer outcome for queer refugees in different contexts. This latter sombre point is illustrated through much of the discussion in this                                                  62 See for instanceT Alexander Aleinikoff, “Protected characteristics and social perceptions: an analysis of the meaning of ‘membership of a particular social group’” in Erika Feller, Frances Nicholson & Volker Türk, eds, Refugee Protection in International Law: UNHCR’s Global Consultations on International Protection (Cambridge: Cambridge University Press, 2003) 263. and Hathaway & Foster, supra note 23. 63 Thomas Gammeltoft-Hansen, Access to Asylum: International Refugee Law and the Globalisation of Migration Control, Cambridge Studies in International and Comparative Law (Cambridge: Cambridge University Press, 2011). See p.15: “forms of extraterritorial or externalised migration controlhave been rapidly expanding across traditional asylum countries, in particular Australia, the United States and European countries.” 64 James Hathaway & Jason Pobjoy, “Queer Cases Make Bad Law” (2012) Articles, online: <> at 318.  40  dissertation. Thus, although many LGBT refugees are believed to be attracted to South Africa because of the country’s legal protections for sexual minorities, a UNHCR resettlement officer described the entire LGBT refugee/asylum seeker population in South Africa as ‘of concern’.65 Whereas the study of Western asylum systems is regularly extracted from its specificity, the passing, though often positive, discussion of instruments like the AU Refugee Convention is discussed as distinctly located exceptions. Some writers call for the innovation of Global South refugee definitions to be ‘internationalized’ that is, incorporated into Western asylum systems. Goodwin-Gill and McAdam are critical of the fact that whilst states evidently recognize the gaps in the UN Refugee Convention for serving broader needs and matching realities, there remains a lack of willingness to create or systematize state obligations, leaving a weak discretionary gap. The authors’ position is that the AU Refugee Convention realistically extends the definition of ‘refugee’.66 Discussing Canada’s guidelines on gender-related persecution, Macklin describes the way in which millions of women are left out of the Refugee Convention protection realm, “not because the impetus for their flight was gender-specific, but because starvation, war, and environmental disaster ‘don't count’ for purposes of the legal definition”.67 In this regard she criticises the failure of countries such as Canada to take up a supplementary definition to that created by a Eurocentric liberal rights paradigm, as reflected in the African and Latin American regional agreements.                                                  65 The views expressed were the personal views of the interviewee and not attributable to the UNHCR, Kizitos Okisai, Interview with Kizitos Okisai, UNHCR Resettlement Officer, United Nations High Commission for Refugees (UNHCR) (2016), Pretoria.; See further the discussion in Chapter 5.2 66 Goodwin-Gill & McAdam, supra note 53 at 292–294.  67 Audrey Macklin, “Refugee Women and the Imperative of Categories” (1995) 17:2 Human Rights Quarterly 213–277 at 218.  41  In her seminal piece, “Can the subaltern speak?”, Spivak seeks to uncover how the disavowal of a subject position and location reinscribe the “unacknowledged Subject of the West” and the only-imagined, muted Other.68  Spivak criticises how the neutral and disembodied authorship by white, Western men investigating others as objects of study contributes towards cementing a very specific perspective of subject-object study which both masks its specificity and location and produces a knowledge of an always silent object, spoken about but not spoken with. In refugee law, the unacknowledged subject is of the Western host, disguised as a typical host and presented as representative of the application of refugee law, understood as exclusively located in Western asylum systems. The Third World ‘Other’, in the majority of Western refugee scholarship, is imagined only as the source of refugees, rather than also the primary host of the world’s refugees. This produces several gaps in analysis, missed opportunities for advocacy and indeed the positing of solutions which would be disastrous if applied in the unimagined Southern host. It is both the exposition of this bias and the analysis of its consequences that this dissertation sets out to achieve.  Many of the proposals suggested by authors whose work relies upon a limited yet hegemonic framing of Refugee Law issues, may thus be awkwardly applicable or completely inappropriate for the majority, neighbouring host states of the Global South.69 Revisiting some of the authors whose work has been discussed above, reveals such problematic proposals. Gibney refers to the national examples of Germany, the UK, the US and Australia in drawing out his conclusions on the issue of practicality of proposals for change in the asylum system; arriving at his supposedly more principled position that state obligations are owed to refugees anywhere, but only extend to                                                  68 Spivak, supra note 9 at 87.  69 See the criticism of power-imbalanced knowledge production from the Third World feminist perspective by Mohanty, supra note 8. 42  circumstances of low-cost.70 The effects of such a principle were it to be applied in the developing countries who host 86% of the world’s refugees, and especially in the least developed countries, which play host to 25% thereof, are not considered. 71 Similarly proposing a solution to the problems of excluding and rights restrictive asylum processes, Price advocates for ‘returning’ to a more blatantly politicized and much more restricted conception of asylum.72 Should Price’s proposals for eliminating refuge for those displaced for reasons of generalized violence, for those coming from states unable to protect their citizens or from failed states, be implemented in the host states of Africa, not only would this leave stranded a vast number of those displaced within the African continent, but his advocacy for the granting of refugee status as a political tool for condemnation, and one device amongst a set of interventionist strategies, would, in the case of African neighbours, effectively amount to war-mongering. In considering the general passing over of the South as not only refugee-producer but also the largest refugee-acceptor, evident in the refugee law literature, Price’s book provides perhaps the best example of how such an approach is not only incomplete but can result in dangerously inappropriate proposals.   Identifying the dissonance in these proposed solutions to an anti-refugee backlash in the domestic application of refugee law with the circumstances of the Global South which hosts the majority of the world’s refugees requires first that these works be located as exclusively based on the study of Western asylum systems and these systems’ exceptionality called out. This is the larger work undertaken and illustrated by this dissertation.                                                   70 Gibney, supra note 55. 71 United Nations High Commissioner for Refugees UNHCR, “UNHCR Global Trends 2014”, online: UNHCR <> at 2.  72 Price, supra note 56. 43  2.2 Useful Insights: Lost in Translation  Much of the literature surveyed above purports to be international in character or application, yet as my analysis suggests, such refugee law studies reveal a preoccupation with Western host states. This literature poses challenges for Southern scholars and for the application of such findings within Southern hosts, in part, due to the invisibility of their bias. Nevertheless, the work of authors writing on Western host states provides many useful insights: both in relation to the ideas which have influence in the world beyond false imaginings of ‘international’, and in raising themes familiar to other, albeit unexamined, locations. In drawing insights from different places (differently located studies and different disciplines, discussed further in 2.5 below) it is important to remember place specificity (perhaps especially so where this is not considered by the author) and one should not assume universal applicability of experience. However, the familiar themes which arise are also illustrative of how useful insights from different places can be, with the caveat that their difference be remembered. The following briefly outlines some of the instructive works and ideas which have informed the differently placed analysis of this dissertation. This suggests the more we learn about the application of refugee law from different places, this dissertation standing as an example, the better we will be able to understand which lessons are indeed widely applicable and which are more specific to place or circumstance.  Thus whilst the canvassing of jurisprudence used to create a very useful summary of an issue in refugee law is done with a Western bias (referencing only Western host state jurisprudence), the summary itself that Foster provides in her chapter in Arbel et al’s Gender in Refugee Law book, is useful in establishing the base line so influential beyond the countries from which it is drawn. Previous, problematic and now rejected approaches towards establishing a group as a ‘PSG’ are summarized: the alleged need for voluntary association, homogeneity or internal cohesion of the 44  group, and interpreted limitations with regard to the size of the group – not ‘too small’/‘too large’. Foster then turns to outlining current approaches towards a PSG: (i) the ‘protected characteristics approach’ (with associate interpretations of immutability), inspired by anti-discrimination grounds – a link which Foster sees as the desired solution to a uniform PSG approach; and (ii) a ‘social perception approach’, which requires that a PSG be cognizable as a group in society.73  Whereas some conclusions drawn by refugee law scholars studying Western host refugee law application may not be as universal as they imagine, in other instances, problems imagined as Western-host specific may surprise as being more broadly encountered. In this regard, discussions on the reliance on stereotype and cultural misreading of asylum applicants, sometimes presumed to be a consequence of the Western adjudicator’s view of Southern originating refugees, find echoes in the Southern-located adjudication of the claims of neighbouring states’ refugees. This is perhaps especially so in the case of queer refugees, who are always a minority for the adjudicator, wherever the adjudicator is placed. Examples include a paper discussion between authors where Pobjoy and Hathaway construct an argument on queer case law which is criticised in a follow-up article by Millbank; a debate is raised as to the scope of identity and of activities or expressions as part of identity.74 Hathaway and Pobjoy importantly discredit a ‘discretion duty’ argument made by many asylum adjudicators; wherein the adjudicator reads in a sui generis imposition of a duty to internalize                                                  73 Michelle Foster, “Why we are not there yet: the particular challenge of ‘particular social group’”, (16 April 2014), online: Gender in Refugee Law <>. 74 Hathaway & Pobjoy, supra note 64.; Jenni Millbank, “The Right of Lesbians and Gay Men to Live Freely, Openly, and on Equal Terms is Not Bad Law: A Reply to Hathaway and Pobjoy Symposium: Uncovering Asylum: A Conversation on Refugee Law, Sexual Orientation, and Moving Towards a Just Jurisprudence” (2011) 44:2 NYU J Int’l L & Pol 497–528. 45  repression, or to conceal identity to avoid risk, which the authors discuss and dismiss as nullifying Convention protection. The authors also identify how the modification of behavior (and so denial of the right to a private life), or the psychological impact of such repressive modification (which may constitute cruel, inhuman or degrading treatment), based on fear of persecution upon discovery or being ‘outed’ (being forced into hiding one’s sexual orientation), may itself constitute the relevant persecutory harm, which they discuss as ‘endogenous harms’. However, much of the paper is focused on a discussion of “the scope of activities protected as inherent in sexual orientation”, referring to the guidance on scope existing in relation to religion or political opinion grounds for asylum. The authors criticise the judgments in the cases discussed for failing to properly grapple with this question of the scope of inherent/integral and thus protected activities, and taking too broad a position which may call into question the nexus element of the refugee definition. The authors feel that the judgments were at once under-inclusive – in failing to be sufficiently “attentive to the endogenous harms that follow from having continually to mask one's true identity”75 – and over-inclusive in failing to interrogate the protection interests related to sexual orientation, based on “associated activities, rather than simply as a function of identity per se”.76  The focus with regard to the latter point on activities instead of identities is difficult to understand with regard to a Refugee Convention which protects only identities and never ‘activities’, beyond their link – real or imputed – to signal identity; in which case whatever the form or nature of the actual activity in question is irrelevant. This aspect of Hathaway and Pobjoy’s paper is roundly criticised by Millbank. Millbank is critical of Hathaway and Pobjoy’s analysis in so far as it rests “upon a misleading and unsustainable act/identity distinction                                                  75 Hathaway & Pobjoy, supra note 64 at 335. 76 Ibid at 336. 46  (comprising equally unsustainable binaries of integral/peripheral and necessary/voluntary acts).”77 As well as the false distinction underlying their analysis, Millbank is concerned with both the likely misapplication of their proposed test for ‘protected activities’, described as those “reasonably required to express sexual orientation”, and the way in which this analysis serves to reintroduce a form of ‘discretion’ reasoning.78 Millbank shows how adjudicators have historically clearly considered trivial what most would consider integral forms of identity expression, in determining these activities or expressions are reasonably avoidable, and hence her discomfort with Hathaway and Pobjoy’s proposal.  Lines between what is "integral" and what is "marginal" conduct associated with sexual minorities in another culture prospectively drawn by Western decision makers have often failed to properly encompass accepted human rights standards … Put bluntly, the more marginal a group is in social and legal terms, the more likely that what is experienced as core by them is deemed marginal by adjudicators.79 Although the above discussion is conducted with reference to Australian case law, the insights regarding misreading, particularly of marginalized groups are applicable beyond a Western-host, Southern refugee reading to South-South readings. Several scholars bring to light the problematic construction of the refugee figure, and of women refugees and their experiences, in particular. Barsky, examining Canadian asylum cases, tracks the movement in the process from human being to refugee status claimant and from claimant to Other. A very narrow aspect of the claimant’s experience is deemed relevant and they are expected to fit their experience into a very particular yet generic construction.80 Dauvergne and                                                  77 Millbank, “The Right of Lesbians and Gay Men to Live Freely, Openly, and on Equal Terms is Not Bad Law”, supra note 74 at 501. 78 Ibid at 513. 79 Ibid at 516. 80 Robert Barsky Constructing a Productive Other: Discourse Theory and the Convention Refugee Hearing (Philadelphia: John Benjamins Publishing Company 1994) 47  Millbank criticize the use of stereotypes and blunt proxies in their analysis of Australian, Canadian, US and UK forced-marriage case law. The authors demonstrate the way in which women applicants especially are required to meet the script of poor, rural, young and uneducated, or have their claims rejected.81  In a book titled Fleeing Homphobia, which canvasses the position of LGBTI asylum seekers within the European asylum system, several authors discuss the impact on credibility assessments of pervasive stereotypes which negatively impact evaluations of group self-identification, in particular.82  Spijkerboer, in his analysis of Dutch asylum decisions, discusses the harmful assumptions about femininity, gender and sexuality in an arena in which the powerlessness of the applicant allows the officials to construct the female applicant as they wish. Spijkerboer argues for an increased consciousness of the constructions/representations created by the refugee system and the need to “ask whether the identities we create are actually inhabitable by the people we create them for”.83  Rehaag, in a detailed analysis of sexual orientation-based asylum cases in Canada pays particular attention to the differentiated treatment of bisexual applicants. Rehaag identifies a dominant understanding of sexuality, as an innate and immutable personal characteristic, as the cause for the difficulties bisexual claimants encounter in making a claim capable of positive reception. He argues that “adjudicators should embrace an alternative understanding of sexual orientation that can accommodate a multitude of sexual-minority life stories. This understanding views sexual                                                  81 Catherine Dauvergne & Jenni Millbank, “Forced Marriage as a Harm in Domestic and International Law” (2010) 73:1 The Modern Law Review 57–88 at 77.  82 See for instance the chapters by Janna Weβels, Louis Middelkoop, Nicole LaViolette and Thomas Spijkerboer in  Thomas Spijkerboer, ed, Fleeing Homophobia : Sexual Orientation, Gender Identity and Asylum (Routledge, 2013). 83 Spijkerboer, supra note 27 at 7. 48  orientation as flexible and fluid.” Furthermore, he advocates that decision makers should “focus not on the sexual identity of claimants but rather on evidence of their persecution on account of traditional gender roles and compulsory heterosexuality”.84 All of the above discussions of the use of stereotypes and poor stand-ins for understanding the lived experiences of asylum applicants, find echoes in non-Western host country refugee status determination procedures, discussed further in the following subsection. My work, as well as that of some of the Southern scholars discussed in the subsection below, demonstrates that the criticisms of shallow understandings of issues of gender and sexuality in refugee law, emerging from Western-host focused studies, are in themselves broadly applicable. Whilst gender and sexuality, and the construction thereof, are issues much debated in feminist literature and Butler, for instance notes that “those who fail to do their gender right are regularly punished”,85 these terms are often less carefully used or related in Refugee Law analysis and this criticism is not isolated to host states in the West.86 The introductory chapter of Arbel et al’s collection Gender in refugee law: from the margins to the centre, highlights the gender and sexuality question as being especially important and requiring further study.87 These concepts, though sometimes placed beside one another, are rarely understood as intimately related or mutually constitutive (lying within one another). Specifically, the authors call for an unpacking of homosexuality to see men and women’s different experiences of sexuality and related                                                  84 Sean Rehaag, “Patrolling the borders of sexual orientation: bisexual refugee claims in Canada” (2008) 53:1 McGill Law Journal 59 at 59. 85 Butler, Judith, “Performative acts and gender constitution: an essay in phenomenology and feminist theory” in The feminist philosophy reader (Toronto;Boston; McGraw-Hill, 2008) at 99. 86 See for instance the chapters by Ibid at 97–98.; Oyeronke Oyewumi, “Visualizing the body: Western theories and African subjects” in The feminist philosophy reader (Toronto;Boston; McGraw-Hill, 2008) at 168; Catherine A MacKinnon, “Sexuality” in The feminist philosophy reader (Toronto;Boston; McGraw-Hill, 2008) at 204–208; Ann Ferguson, “Sex war: the debate between radical and libertarian feminists” in The feminist philosophy reader (Toronto;Boston; McGraw-Hill, 2008) at 223. 87 Efrat Arbel et al, Gender in Refugee Law : From the Margins to the Centre (Routledge, 2014) at 5. 49  oppression. Buscher is amongst the exceptions in explicitly discussing and emphasising that whilst ‘gender-based violence’ is a term generally “used to refer to violence against women and girls, it also encompasses violence against women and men because of how they experience and express their gender and sexuality”.88 This is a necessary insight directly relevant to the world of refugee law and references experiences which protest an understanding of ‘gender persecution’ as either persecution of women or persecution of transgender individuals and necessarily complicates male-only perspectives of sexual-orientation based persecution (discussed further in Chapter 5). As well as often conflating the term ‘gender’ with women in Refugee Law, many writers also make use of a ‘womenandchildren’ conception of ‘women’, turning what Spijkerboer points out are more or less 50/50 divisions of men and women refugees and of adults and children, into a conglomerated majority.89 As well as being inaccurate, this focus upon a constructed majority shifts attention away from the more discriminatory gender gap with regard to the discrepancy in numbers between men and women refugees actually present in the Northern as opposed to (some) Southern host states.  Some authors, in discussion on the stereotypes used to construct a refugee figure, also usefully reveal the construction of a stereotypical ‘self’ as host which has interesting echoes amongst Southern hosts. It is Macklin’s position that a feminist analysis ‘superimposed’ on the constructed Self/Other binary of refugee discourse; that is, the good, Western, refugee-acceptor ‘Self’ (who is not a refugee-producer, and whose responsibility for refugees produced elsewhere is not acknowledged) and the abusive, refugee-producer ‘Other’ (not even imagined by Macklin                                                  88 Dale Buscher, “Unequal in Exile: Gender Equality, Sexual Identity and Refugee Status” (2011) 3:2 Amsterdam Law Forum 92 at 95. 89 Spijkerboer, supra note 27 at 17. 50  as a ‘refugee-acceptor’), requires a crossing of this boundary by challenging the self-understanding of ‘non-refugee-producers’ with regard to universal gender persecution. Failure to do so can result in cultural difference being used as a yardstick to distinguish ‘our discrimination’ from ‘their persecution’; reflected in the evidentiary barrier created with regard to claims from similar ‘safe’ countries: “one strategy for deflecting this discomforting irony”.90  Literature on gender in refugee law often focuses on domestic violence as a particularly gendered experience which throws up challenges for host state asylum adjudicators who see insufficiently ‘other’ applicants in front of them; whose persecution may seem too close to home and must therefore be trivialized, lest acknowledging such experiences as persecution seem treasonous. In particular, Arbel’s Canadian study criticises the tendency of adjudicators to recognize domestic violence claimants only where their experiences can be framed as caused by persecutory cultures, rather than recognizing the persecutory practices (the violence in question).91 This approach overlooks circumstances and contexts which render women vulnerable to violence beyond ‘culture’ and creates a necessary essentializing and totalizing cultural narrative for success; which results in the rejection of those women who cannot frame their experiences in such a narrative. Arbel also points to the defensive distancing effect of a culture-focus which allows host states to maintain a narrative of superiority over refugee countries of origin in spite of the prevalence of domestic violence across cultures.92 In their Western-focused study on forced marriage in immigration and refugee contexts, Dauvergne and Millbank call for an “understanding that forced marriage is a harm that is based upon power imbalances concerning                                                  90 Macklin, supra note 67 at 264–266. 91 Efrat Arbel, “The Culture of Rights Protection in Canadian Refugee Law: Examining the Domestic Violence Cases” (2013) 58:3 McGill Law Journal 729–771 at 729. 92 Ibid at 733. 51  gender and sexuality rather than simply being a reflection of ‘culture’”.93 The authors point to the way in which the location of forced marriage as a foreign cultural practice marks Western romantic marriage traditions as ‘culture free’ and “obscures the role that the dislocations of migration, increasingly restrictive migration policy, and associated intergenerational disconnections, have played in creating a contemporary Westernized setting for forced marriage.”94 The authors particularly emphasize, however, a lack of a shared understanding and approach in forced marriage cases, between the UK’s domestic forced marriage legislation and refugee law. The discrepancy in approach is explained by the authors as evidence that refugee law, ostensibly part of the human rights system, is instead used as an immigration law screen and border enforcing mechanism.  As reflected in the discussions on gender and sexuality in refugee law by the Southern-focused scholars outlined below, as well as the data analysis of this dissertation itself, many of the above ideas regarding the role of a constructed ‘culture’ and assumptions of host state superiority in asylum adjudication provide useful takeaways applicable in (unconsidered) Southern host refugee law assessments. Whereas the ethnocentric anxiety accompanying many such Western-located criticisms assumes that there would be no such tensions in South-South asylum adjudications this thesis demonstrates the way in which a familiar yet differently located ‘othering’ process becomes similarly overlain upon asylum seekers within South Africa. I also demonstrate the problematic degree to which South African assumptions of host-state superiority disguise and complicate circumstances of continuing persecution.                                                    93 Dauvergne & Millbank, supra note 81 at 66. 94 Ibid at 63. 52  The flipside of a simplified good-host future, persecutory past reading of asylum seekers’s lives, and a weakness of much of the above literature, is that differently located authors focusing both on ‘gender cases’ and queer refugee cases are effected by a tendency to discuss family only in the negative. I identify this presumption as limiting conceptions of family in both Western and Southern host states. This characterisation of ‘family’ leaves empty the same-sex partnership as also constituting ‘family’ and overwrites the importance of family for support and as a (primary) means to escape to inaccessible Western states, particularly for women, as discussed by Williams and Bhabha.95 A notable exception is the discussion of same-sex refugee reunification constraints in Austria by Suβner who exposes the racist underpinnings of a Western normative claim to upholding domestic same-sex equality whilst neglecting or rendering null the right to same-sex reunification of refugees. Suβner points to an intersectional approach as a necessity for a rights-focused queer, anti-racist critique.96 Williams’ sociological account of cross-border marriage provides an important counter-balance to the legal literature on family and marriage in refugee law which tends to focus on these institutions in the negative as a site of harm. Here Williams points to the importance of marriage as a means of strengthening and maintaining communities, particularly in circumstances of a ‘lost homeland’, and its value in rebuilding identities shattered by conflict and war. Marriage may thus represent a healing opportunity to normalize a life rendered chaotic as well as opening up a means of escape or protection for family members. Williams’ book is particularly focused                                                  95 Lucy Williams Dr, “Marriage within Refugee Communities” in Global marriage: cross-border marriage migration in global context Palgrave Social Sciences Collection (Basingstoke;New York; Palgrave Macmillan, 2010); Jacqueline Bhabha, “Border rights and rites: generalisations, stereotypes and gendered migration” in Sarah Katherine van Walsum & Thomas Spijkerboer, eds, Women and Immigration Law (Routledge-Cavendish, 2007). 96 Petra SUßNER, “Invisible intersections, queer interventions: same sex family reunifi cation under the rule of asylum law” in Thomas Spijkerboer, ed, Fleeing Homophobia : Sexual Orientation, Gender Identity and Asylum (Routledge, 2013). 53  on agency in immigrant and cross-border marriages and she also highlights how the refugee label can shape and limit agency.  On the naming and construction of identity generally, Crenshaw raises a criticism of identity politics, with concern for intersectional experiences, which rarely enters refugee law analysis. Although ostensibly located in ‘the West’, critical race feminisms have much to contribute to a Third World approach to reorienting international law and, I argue, with particular relevance to refugee law and important parallels to this dissertation project (see particularly the discussion in Chapters 5.4 and 6.2). Crenshaw discusses the way in which the qualitatively different experiences of women of colour of violence is not represented in either feminist or antiracist discourses, and the way in which their location at the intersection of these identity groups, marginalizes these women. She shows how raced and gendered dominant positions – male persons of colour and white women – define and confine the interests of the whole group and determine the parameters of strategies. It is therefore not just that there are differences between, for instance, white women and women of colour within the ‘women’ identity category but that the former have the power to determine whether the latter’s concerns will be incorporated at all into policy formulation. By only actually reflecting a certain sub-section’s interests and experiences, Crenshaw is thus able to point out how feminist and anti-racist discourses fail even to tackle their discrete sexism and racism projects fully.  Moreton-Robinson in her Indigenous feminist work seeks to expose “an invisible racialized subject position [white women] that is represented and deployed in power relations with Indigenous women through discourse”.97 Moreton-Robinson identifies the privilege that white women are not represented to themselves as white, but rather (only) variously classed,                                                  97 Moreton-Robinson, supra note 3. at (xxii). 54  sexualised, aged and abled. A politics of difference does not include identifying whiteness as a difference that warrants interrogation and is what Moreton-Robinson describes as power-evasive. Furthermore, in a criticism echoing that of Crenshaw, Moreton-Robinson highlights how the consciousness of white Australian feminists that they live in gendered bodies, but not in racialized bodies, has profound effects on the agenda, tools and priorities of this majority group of feminists in the country. Crenshaw argues that, as potential sources of social empowerment and reconciliation, identity groups need to be buttressed by integrating the multiple identities which intersect in real people. For Crenshaw the failings of identity politics are not due to a failure to transcend difference but a failure to recognize intra-group difference. Moreton-Robinson asserts the Indigenous expectation that when advocating for equality for all women, the needs of those who are in the most unequal position in society should be attended to first. In this dissertation’s discussion of the differentiated needs of particular vulnerable people, influenced by constructions of the refugee figure and grounds for asylum phrased as identity categories, the above insights are especially valuable.  2.3 Refugee Law Voices from the South Scholarship on refugees and refugee law centred on Third World host states is comparatively scarce but brings to light significantly different realities to those canvassed by their Western counterparts. As well as the scarcity of focus on Southern host states, the focus is seldom at the level of scrutiny and detail found in scholarship on host states of the West. My project was decidedly fine-grained and several of the professional interviewees explained that because the basic functioning of the South African asylum system was so entirely their advocacy focus, the level of detail I was looking for – on issues of sexuality, gender and alternative family 55  formations – was never engaged with and indeed considered tangential to the work that needed to be done. The positivist tendency in refugee law scholarship is, I think, aggravated in scholarship focused on Southern hosts where a sense of immediate urgency and the normative necessity of ‘practical’, read as doctrinal, studies  reigns. Scholarship on refugee law as practiced in Southern host states has thus been predominantly broad-stroke descriptive work with little theoretical critique (notable exceptions, including Chimni and Abuya are discussed below). This approach substantially denudes our ability to engage with, critique and supplement the voices of the Northern-focused scholars. In this genre, presentation of ‘the law’ applicable is the key. The ‘PSG’ and other academic discussions highlighted above are therefore left within the purview of those writing from, and for, the developed world. It must be noted that the limited number of home-grown refugee law texts mean that even a classical, descriptive approach to setting out refugee law in a Southern host is thus filling an important gap. In South Africa, Khan and Schreier’s Refugee Law in South Africa98 is valued by refugee law teachers as the first and only such textbook which is specific to South Africa.99 This book is primarily descriptive of the South African refugee legislation including a critique of its application. The authors place an interesting emphasis on the principle of non-refoulement, including highlighting how systemic barriers to asylum can be considered indirect refoulement, for example through narrow or incorrect interpretation in refugee status determination interviews. They add that “where civil, political and socio-economic rights afforded to refugees by international and domestic law are calculatingly denied to refugees in the                                                  98 Fatima Khan & Tal Schreier, Refugee law in South Africa, 1st ed. ed (Cape Town: Juta, 2014). 99 Ibid at 5.; Interview with Alicia Raymond, Head of Refugee Unit, Wits Law Clinic, Johannesburg, 11 November 2016: “besides for this book, every other book that I have is for a more international perspective. So this is the only book that is South African. This is the South African perspective. Which is why I like to use it for students and I like them to read them.” 56  country of asylum thereby forcing them to leave the country of asylum, constructive refoulement can be said to have taken place.”100 One important difference in approach to refugee law studies from within a Global South host perspective, is that the refugee law-in-the-books is unlikely to be taken at face value. This is reflected in Khan and Schreier’s book as well as Abuya; in discussing the Kenyan asylum system, Abuya compares the ‘legal’ and ‘actual’ asylum systems. The eventual collapse of the Kenyan refugee protection regime (and handover to the UNHCR) is described as the result of political pressures, economic decline and rising xenophobia in combination with and in reaction to an incredible and rapid increase in refugees from multiple neighbours, sending shock waves through the asylum system.101 These are all experiences unlike those of most Western host states. Furthermore, the clear inclusion of refugee family members as refugees themselves – although initially phrased in the Kenyan legislation as the ‘wife or child’ of the male refugee and relying on a patriarchal ‘head of family’ conception – is  a point of interest more commonly found in African than Western host approaches. An aspect of family reunification remarked on by Abuya, which is not generally reflected by other authors, is that: “[a] reunited family may also be better placed to assess when conditions at home are conducive for repatriation, unlike a dispersed family.”102  Writing on the South African asylum system, Middleton raises issues which are at once unique – as the domestic refugee law has been amended so that gender (and sexual orientation) is not merely interpreted under ‘particular social group’ but explicitly listed as included therein – yet                                                  100 Khan & Schreier, supra note 98. at (xxxvii). 101 Edwin Odhiambo Abuya, “Past Reflections, Future Insights: African Asylum Law and Policy in Historical Perspective” (2007) 19:1 International Journal of Refugee Law 51–95. 102 Ibid at 79. 57  familiar to the work of differently located authors on gender in refugee law, discussed above. Middleton identifies gendered harm as being effectively understood as persecution only where (i) such harm resulted at the time of a recognized conflict (the AU Refugee Convention’s addition of conflict situations as pertinent to refugee status being key here) – being rejected en masse once a political peace is declared, irrespective of conditions on the ground; or (ii) is associated in the decision-maker’s mind with a foreign/‘backward’ culture. Although this study was conducted prior to an important legislative amendment in South Africa, all of these findings would contribute to a fuller discussion of gender under refugee law, were the concept of ‘host’, in the mind of Western writers, to be extended to the refugee hosts of the South.  As well as its unique legislation on gender and sexuality grounds in refugee law and its long-standing position as the recipient of the world’s largest number of asylum applicants, South African-focused research can particularly contribute to refugee law for, as remarked by Goodwin-Gill, South Africa has experienced “the full spectrum” of refugee movements, having been a refugee source, to now being a refugee-receiving country.103 This history carries with it the potential for thoroughly destabilizing both the geographical and supposedly fixed aspects of the benevolent Western-refugee host/rights-violating Southern refugee-producer binary evident in much Western refugee law scholarship and discussed further in Chapter 4. The above commentary on the circular nature of migration in South Africa’s history, by Goodwin-Gill, is found in a collection of writings on South Africa edited by Handmaker et al which provides a useful framework for establishing both novelty and familiarity in the South                                                  103 Guy S Goodwin-Gill, “International and National Responses to the Challenges of Mass Forced Displacement” in Jeff Handmaker, Lee Anne De la Hunt & Jonathan Klaaren, eds, Advancing Refugee Protection in South Africa (Berghahn Books, 2008) at 11. 58  African asylum system, when compared to the writers presenting only on Western host states.104 As regards the demographics of asylum seekers in South Africa, Landau reveals they are typically considerably better educated than the South Africans amongst whom they live – “confounding popular images of asylum-seekers and refugees as desperate for aid”.105 A chapter included in the collection titled “Protecting the Invisible: the Status of Women Refugees in South Africa” reveals that, like Western host countries, there are far more men than women in the South African asylum system (though women are increasing); women regularly represent only 20% of asylum applicants and 5% of those formally granted status. Like differently located authors writing on gender in refugee law (including Spijkerboer, discussed above), the South African authors, Valji, De la Hunt and Moffet, look to the 1951 UN Refugee Convention and its construction of the typical refugee as male and the private/public misinterpretation of women’s political involvement by way of partial explanation for the disparity. Furthermore, the ‘missing women’ in South Africa – like in the West – are in refugee camps elsewhere on the continent for lack of mobility, independence and resources, compared to their male counterparts: “the very low number of women seeking asylum (in South Africa) is overwhelmingly disproportionate to the actual number of women refugees in the African subcontinent”.106 Similar to the Western-focused authors discussed above, the authors argue that it is not enough to simply add gender                                                  104 Jeff Handmaker, Lee Anne De la Hunt & Jonathan Klaaren, Advancing Refugee Protection in South Africa (Berghahn Books, 2008). 105 Loren Landau, “Regional Integration, Protection and Migration Policy Challenges in Southern Africa” in Jeff Handmaker, Lee Anne De la Hunt & Jonathan Klaaren, eds, Advancing Refugee Protection in South Africa (Berghahn Books, 2008) at 35. 106 Helen Moffet, Lee Anne De la Hunt & Nahla Valji, “Protecting the Invisible: The Status of Women Refugees in Southern Africa” in Jeff Handmaker, Lee Anne De la Hunt & Jonathan Klaaren, eds, Advancing Refugee Protection in South Africa (Berghahn Books, 2008) at 216. 59  grounds to the listed grounds of persecution (discussed in Chapter 3.1), although from a position where this has actually been done, not imagined.107  Field-based research from refugee camps within host countries of the Global South, has revealed some important insights in relation to the lived experiences and practical difficulties facing refugees in relation to gender and sexuality concerns from within their host countries. The primary gendered concerns highlighted by Martin, for instance, in a regional, refugee camp setting, are framed as physical security and social and economic rights gaps, adding details often left out by those focusing on the asylum application/refugee status recognition moment. Furthermore she contributes information relating to Western hosts’ ‘missing women’: identifying Central Africa and the Great Lakes region as having the highest proportion of forcibly displaced women and Europe as having the lowest. She points to gendered-resource allocation as one of the reasons behind the discrepancy. Together with the identified socio-economic, health and gendered-structural difficulties particular to women, this also raises interesting questions (briefly alluded to by Martin) regarding the gendered effects of Western resettlement policies with reference especially to gendered differences in access to language skills and camp leadership and employment positions.108 Furthermore the problem of persecution from within the refugee’s host country is more fully elucidated by Southern-host focused authors. Buscher discusses gay Iraqi men fleeing sexual orientation persecution and having to articulate an asylum claim in countries such as Lebanon, Jordan and Syria where there are repressive host county laws toward LGBT persons. He also elucidates the problems of local community and refugee community harassment, limited access                                                  107 Ibid at 215. 108 Susan Martin, “Refugee and Displaced Women: 60 Years of Progress and Setbacks” (2011) 3:2 Amsterdam Law Forum 72. 60  to social support, employment, medical care or police assistance within host countries. The “caustic mix of marginalization in key areas of life” for LGBTI asylum seekers and refugees is further elaborated upon by the interview-based work of the Organization for Refuge, Asylum and Migration (ORAM) within Turkey.109 Also produced by ORAM, Grungras et al have collated a truly global survey on NGO attitudes towards LGBTI asylum seekers and refugees.110 ORAM’s report is unique not only as a systemic attitudinal survey of NGOs towards LGBTI asylum seekers and refugees but as the first such survey on any topic to focus on the NGOs which play vital roles in the refugee protection paradigm; ‘the refugee guardians of the civil sector’ and often the ‘gatekeepers’ to official protection channels.  The ORAM report particularly highlights problems of invisibility and silence around sexual orientation and gender identity realities, and the significant lack of awareness amongst NGOs that LGBTI refugees exist within their served populations. The report also critiques the common “blind” approach employed by some NGOs who consider impartiality to the ‘non-issue’ of gender identity and sexual orientation as sufficiently addressing non-discrimination/equality concerns, rather than understanding how these issues are germane to their clients’ protection needs. Sadly, the report also canvasses “a sizable minority” of NGOs with negative views on same-sex conduct and/or expression of transgender identity; unsurprisingly proportionally more significant amongst NGOs within African, Asian, South American and MENA regions than North America or Europe.  However, the reports’ corrective potential for a more diverse approach to the host/origin country relationship is limited by the common reference to the                                                  109 Unsafe Haven: The Security Challenges Facing Lesbian, Gay, Bisexual and Transgender Asylum Seekers and Refugees in Turkey (Updated edition), by Irem Arf et al, (A joint publication of Helsinki Citizens’ Assembly—Turkey and ORAM—Organization for Refugee, Asylum and Migration., 2012). 110 Opening Doors: A Global Survey of NGO Attitudes Towards LGBTI Refugees & Asylum Seekers, by Neil Grungras & Cara Hughes, (Helsinki Citizens’ Assembly (Turkey), Refugee Advocacy and Support Program, (Organization for Refuge, Asylum & Migration), 2011). 61  (Global South) hosts in question being described as countries of ‘transit’ or of ‘first asylum’, despite the very limited number of LGBTI refugees who are believed to openly make sexual orientation claims and become resettled.  Whereas Western-focused writers will often write on their selective focus in refugee law as if it were of universal application, Southern-focused writers will more usually write for a narrower audience or assume more location-specificity to their subject. A significant exception to this and to a positivist trend in Southern-host focused refugee law studies, however, is the work of Chimni, whose work has been discussed above. Chimni particularly works to expose and destabilize a “myth of difference” utilized in a post Cold War exclusionary turn in the application of refugee law. Chimni demonstrates that this myth of difference requires a highly selective view of history, an exaggerated sense of difference and also the representation of certain spaces as ‘filled’ and others as ‘empty’.111 Chimni attempts to rebut the assumptions underlying the myth of difference with a longer perspective on history and reference to the burden of refugee hosting which lies so much more substantially on the South, as well as outlining the role of capitalism and imperialism in causing displacement.112    Although making more of the differences between African and European refugee flows than Chimni does, Abuya similarly calls out the colonial history of the causes of displacement in Africa and the colonial underpinnings of refugee law. Abuya is especially critical of the Liberal assumption of sovereign border control which works to advantage the already-powerful Western states and has weak application to the realities of cross-border movement within Africa. In its                                                  111 See for instance Gibney, supra note 55. who directly supports the (especially Australian) state view and the construction of ‘full’ and ‘empty’ places criticised by Chimni, “The Geopolitics of Refugee Studies”, supra note 61. 112 Chimni, “The Geopolitics of Refugee Studies”, supra note 61. 62  stead, Abuya calls for locally engineered ideas and perspectives, highlighting the centrality of the AU Refugee Convention for refugee law in Africa.113 A new book, published in 2019, specific to queer refugees and written from legal and political disciplinary perspectives, in many chapters bucks a trend towards positivism, and offers both Western focused work and Southern research side by side.114 This book, an edited collection by Güler et al, also demonstrates something new: it is partitioned (without commentary) into apparently doctrinal, case law studies with a strong bias for Western hosts, sandwiched between more political and theoretical, differently located studies. Temporal divisions mark the partitioning of the book’s chapters: before status, the granting of status, and after status. The traditional core of refugee law studies (asylum case studies), the moment at which refugee law appears to be in action, and the middle section of the book is markedly the most doctrinal, the least theoretical and the most Western biased (revolving around the analysis of Western host case law). My own work is more aligned with the ‘peripheral’ work presented in Part I and Part III of this book, which parts are interestingly predominantly focused on Southern host scenarios.  The first content chapter of the book, by Fox, “employs Becker’s theory of moral entrepreneurship to examine Western influence over gender ideologies and the treatment of sexual and gender minorities across the globe, and seeks to explain why non-Western societies that once accepted sexual and gender minorities are now resistant to Western-led LGBTQ rights movements”.115 Part I also includes a chapter advocating a mixed migration approach to                                                  113 Edwin Odhiambo-Abuya, “Revisiting Liberalism and Post-Colonial Theory in the Context of Asylum Applications” (2006) 24:2 Netherlands Quarterly of Human Rights 193–227. 114 Arzu Güler, Maryna Shevtsova & Denise Venturi, eds, LGBTI Asylum Seekers and Refugees from a Legal and Political Perspective: Persecution, Asylum and Integration (Cham: Springer International Publishing, 2019). 115 Katherine Fox, “Implementing Hostility and Acceptance: LGBTQ Persecution, Rights, and Mobility in the Context of Western Moral Entrepreneurship” in Arzu Güler, Maryna Shevtsova & Denise Venturi, eds, LGBTI Asylum Seekers 63  understanding the decision-making of queer Syrians, arguing that stigma produces different reasons given for flight which reasons are also experienced as changeable. The author, Odlum, argues that a mixed migration approach, evolved beyond the traditional refugee differentiation between forced versus voluntary, migrant versus refugee, is key to actually understanding humanitarian needs.116 A chapter by Winton, discussing research located in Northern Central America and Southern Mexico, presents personal accounts of queer mobility in the region and argues for a nuanced concept of mobility based on intersectional complexities of vulnerability and the need to consider other scales of displacement, beyond traditional conceptions of migration and asylum, which are central to queer survival: Queer mobility is seen here in terms of the quest for placement, rather than as movement per se. Continuous negotiations to stay put, to make a place for oneself, were based in disadvantage which often resulted in complex displacements. Displacement in these terms is an intrinsic part of marginal queer experience.117 Part II, which reflects the largely case-law based studies of asylum decisions begins with a chapter discussing ‘randomly selected’ case studies of 40 cases from 10 host countries, wherein two South African cases represent the only sources drawn from a Southern host country in the                                                                                                                                                              and Refugees from a Legal and Political Perspective: Persecution, Asylum and Integration (Cham: Springer International Publishing, 2019) 11 at 11. 116 Alex Odlum, “To Stay or to Go? Decision-Making of LGBTQI Syrians in Mixed Migration Flows” in Arzu Güler, Maryna Shevtsova & Denise Venturi, eds, LGBTI Asylum Seekers and Refugees from a Legal and Political Perspective: Persecution, Asylum and Integration (Cham: Springer International Publishing, 2019) 71 at 71. 117 Ailsa Winton, “‘I’ve got to go somewhere’: Queer Displacement in Northern Central America and Southern Mexico” in Arzu Güler, Maryna Shevtsova & Denise Venturi, eds, LGBTI Asylum Seekers and Refugees from a Legal and Political Perspective: Persecution, Asylum and Integration (Cham: Springer International Publishing, 2019) 95. At 95: “While the ruptures associated with these displacements can cause damage, so they can disrupt established oppressions and allow room to re-accommodate one’s personal social location. Yet, since this re-accommodation is the result of complex constellations of marginalised existence, it is fragile, and hard-won gains can be short-lived. In particular, the intersection between gender and sexual transgression, economic and social marginalization, and rampant organized and targeted hate violence all translate into pervasive precarity. The gravity and complexity of the experiences shared here highlight the need to ensure that the growing body of work on queer migration and asylum does not overshadow other spatial and temporal scales of displacement which are a crucial dynamic of the relationship between queer mobility and survival.” 64  study.118 Most of the chapters in this section focus on EU, US and Canadian refugee case law. The only non-Western host focused chapter in amongst this seven-chapter Part, with a focus on Mexico, is interestingly not a case law study but rather focused on lived experiences, discussing participants living their gender as self-emancipation and whether this can be considered a protection mechanism.119 Part III, an ‘after status’ section on accommodation and integration, is in turn again a Southern-host dominated section and largely focuses on an assessment of the challenges for and different roles of the UNHCR. Particularly relevant for the discussion engaged with in this dissertation is a Kenyan chapter in which the author, Moore, disaggregates experiences of cisgender lesbian, bisexual, and queer women, registered as persons of concern with UNHCR Nairobi, from a homogenizing ‘LBGTIQ’ approach.120 Moore makes the important finding, with particular resonance in this dissertation, that “during interviews LBQ refugees said that they struggle to feel heard and understood by service providers, and that their own understandings of community and self-expression differ to other groups within the LGBTIQ acronym.”121  The construction of Güler et al’s book unintentionally summarises the priorities, limitations, as well as the opportunities currently in play for refugee law scholarship as identified in this                                                  118 Arzu Güler, “Refugee Status Determination Process for LGBTI Asylum Seekers: (In)Consistencies of States’ Implementations with UNHCR’s Authoritative Guidance” in Arzu Güler, Maryna Shevtsova & Denise Venturi, eds, LGBTI Asylum Seekers and Refugees from a Legal and Political Perspective: Persecution, Asylum and Integration (Cham: Springer International Publishing, 2019) 117. 119 María Paula Castañeda Romero & Sofía Cardona Huerta, “Seeking Protection as a Transgender Refugee Woman: From Honduras and El Salvador to Mexico” in Arzu Güler, Maryna Shevtsova & Denise Venturi, eds, LGBTI Asylum Seekers and Refugees from a Legal and Political Perspective: Persecution, Asylum and Integration (Cham: Springer International Publishing, 2019) 251. 120 Hester K V Moore, “‘The Atmosphere Is Oppressive’: Investigating the Intersection of Violence with the Cisgender Lesbian, Bisexual, and Queer Women Refugee Community in Nairobi, Kenya” in Arzu Güler, Maryna Shevtsova & Denise Venturi, eds, LGBTI Asylum Seekers and Refugees from a Legal and Political Perspective: Persecution, Asylum and Integration (Cham: Springer International Publishing, 2019) 323 at 325.  121 Ibid. 65  dissertation. If refugee law, narrowly understood, ‘happens’ when an asylum case is decided upon, this book reiterates the point made in this literature review generally, that it is largely conceived of as only really happening in Western host states. On the other hand, a broader and more critical understanding of refugee law studies may be emerging with reference to the under-examined host states of the Global South, perhaps with particular influence from feminist and queer critical perspectives.  2.4 New Approaches Emerging in Refugee Law – Economics Demonstrating an awareness of the limitations of a Western-host focused study, Macklin asserts that a “variety of psychological, cultural, and financial impediments render women less able than men to undertake the hazardous, uncertain, and expensive journey to Canada.”122 Practically speaking, she therefore reveals that irrespective of the recognition of gender persecution as a basis for refugee status, a country like Canada will never be a viable option for the overwhelming majority of displaced women. The reconsideration of the role of economics in forced displacement and the relationship between refugees and economics in their experience of persecution, economic implications for vulnerability, as well as refugees’ economic life in their host state, is a relatively new focus forming in refugee law studies and migration studies generally.  In Reconsidering Migration and Class, Van Hear argues for “renewing attention on the part class plays in shaping migration – particularly who is able to move and to where.”123 Van Hear centres                                                  122 Macklin, supra note 67 at 220. 123 Nicholas Van Hear, “Reconsidering Migration and Class” (2014) 48:1_suppl International Migration Review S100–S121 at 100. 66  his argument around the seemingly self-evident point: “that patterns and outcomes of migration are shaped by the resources migrants can mobilize, and those resources are largely determined by socio-economic background ... that different people reach different destinations, shaped largely though not entirely by endowments of resources, or class for short”.124 Van Hear goes on to demonstrate that class issues have been largely abandoned in social sciences approaches to migration studies and, where broached, are restricted to studies on migration outcomes in destination countries, not choices and routes of migration.125 Betts, in turn, attempts to challenge assumptions of refugee dependency and to revitalize an interest in refugees’ economic lives in refugee studies with his work on Refugee Economies (Refugee Economies: Rethinking popular assumptions (2014); Refugee Economies: Forced displacement and development (2016)).126  Within refugee law, more specifically, Foster has pioneered a focus on economics; engaging in a discussion of the hurdles decision-makers seem to require of asylum applicants invoking persecutory violations of socio-economic rights which have no equivalent in other cases. Foster highlights the decision-makers’ disquiet that specifically seems to attend cases of economic persecution, in light of assumptions of the role and place of refugee law; criticising a hierarchical attitude towards human rights in which economic persecution is relegated least important and unenforceable, and asylum claims on this basis, hardest to make.127 Adding to this small and emerging work on the relationships between economics and persecution, this dissertation develops this theme with the benefit of perspective from a differently located study of refugee                                                  124 Ibid at 101. 125 Ibid. 126 Alexander Betts et al, Refugee Economies: Rethinking Popular Assumptions (University of Oxford, Refugee Studies Centre, 2014); Alexander Betts et al, Refugee Economies: Forced Displacement and Development (Oxford University Press, 2017). 127 Michelle Foster, International Refugee Law and Socio-Economic Rights: Refuge from Deprivation, Cambridge Studies in International and Comparative Law (Cambridge: Cambridge University Press, 2007) at 156–167. 67  law in an under-studied host country. Benefitting from an approach which recognises the correlation between resources and mobility, this work takes a related but different approach to those above by revealing the interconnectedness of economic resources and vulnerability to and the experience of persecution.  2.5 Beyond Refugee Law Another South African author whose work has usefully informed this project is B Camminga; particularly their own sociology doctoral thesis, subsequently printed as a book: Transgender Refugees and the Imagined South Africa: bodies over borders and borders over bodies. This book focuses around the journeying of the term ‘transgender’ in South Africa – the marking of transgender phenomena – and the journeying of gender refugees, wherein Camminga dissects “the variegated role of the state and its representatives, from apartheid through to constitutionalism, as the arbiter of both sex and gender and eventually refugee status, and the logic or politics of this”.128 Whereas ‘gender refugees’ in refugee law studies is most often used to reflect the gender-based persecution claims of women asylum seekers (see Arbel and Spijkerboer) it is interesting to note that Camminga uses the term ‘gender refugees’ to mean “the journeying of people fleeing persecution, violence, and discrimination on the grounds of their gender identity/expression.”129 Camminga’s book usefully even included research and interviews                                                  128 Camminga, supra note 30 at 4. 129 Ibid at 1. In a talk given at the University of British Columbia, Allard School of Law titled “What if the State No Longer Sexed Us?” (August 27, 2019) Davina Cooper discussed how the common understanding of the term ‘gender’ is changing from its use as a stand-in for ‘women’ to a stand-in for ‘trans’, though remaining an indicator of subjects not of concern to cisgendered men. Davina Cooper, What if the State No Longer Sexed Us? (University of British Columbia, Allard School of Law, 2019). 68  (with interviews conducted prior to and published after my own interview work) conducted with one of the same participants in my study, analysed from a different disciplinary perspective. As evidenced by the previous subsections of this literature review, refugee law studies regularly call for buttressing with the approaches used in other legal arena and indeed other disciplines. Note the references above to the contribution of a West-as-subject construction by Spivak, feminist contributions from Butler, insights from the sociological study of Williams, criticisms of identity politics by Crenshaw and Moreton-Robinson, fieldwork interviews and insights drawn from the work of ORAM, and migration studies and economics approaches courtesy of Van Hear and Betts respectively. Another example of this inter-disciplinary contribution relates to the emerging literature theme discussed immediately above. Whereas the relationships between economics and persecution is relatively new and limited in refugee law studies, and even in migration studies generally, there is a greater level of specificity as regards the relationship between economics and queer experiences of persecution, outside of the asylum context within queer social justice research. Hollibaugh and Weiss’s discussion on economic vulnerabilities of queer persons and its implications, focused on economically marginalised queer experiences in America, throws up many useful insights and statistics on intersectional experiences and compounded vulnerabilities:  LGBT/Q people who are low-wage workers, immigrants, and people of color are already economically vulnerable; gender and sexuality add new layers of vulnerability that target us and keep us isolated from others who face similar circumstances. ... But old-style identity politics are not adequate to understand how the complex multiplicities of gender, race, sexuality, and class interlock and play out, changing the ways that queer people can move in the world or are vulnerable to attack.130                                                  130 Amber Hollibaugh & Margot Weiss, “Queer Precarity and the Myth of Gay Affluence” (2015) 24:3 New Labor Forum 18–27 at 20. 69  Providing some useful insights supplementing the sometimes doctrinal refugee law studies, more purposively theoretical and critical studies in law with particular application to this study’s subjects can thus be found within the literature on Queer studies, Feminist legal approaches and Third World feminisms, expanded on below.   2.5.1 Queer and feminist theorizing on gender, family and marriage My choice to focus upon sexual minority women and the women in polygynous families in this thesis is itself a feminist project; this dissertation stands as a corrective to the predominant focus on gay men, when addressing issues around homosexuality, and upon the husbands, with property/victim wives, in polygynous relationships. With echoes of the intersectionality concerns raised by Crenshaw and Moreton-Robinson, discussed above, Cooper illustrates how the relationship between gender and sexuality is problematized for those who, as lesbians, do not fit into the politically constructed and divided groupings of gender - where ‘gender’ stands for heterosexual women - or into a sexuality grouping where ‘sexuality’ stands for homosexual men. 131 Both sexual minority women and polygynous wives are arguably ‘queer’ in the challenges they pose to the imagined nuclear, heteronormative family.132 For this reason, feminist legal theory, particularly that focusing upon questions of culture and sexuality, as well as the blurred/blurring                                                  131 Davina Cooper, “A retreat from feminism? British, municipal lesbian politics and cross-gender initiatives” (1994) 7:2 Canadian Journal of Women and the Law 431.  132 Margaret Denike, “What’s queer about polygamy?”, (10 June 2010), online: Queer Theory: Law, Culture, Empire <>.. 70  line of feminist and queer theory focusing upon alternative families, have influenced this project.133  Feminist critiques of the legal and institutional understanding of gender are more developed within international law than within refugee law. In criticizing the role and Resolutions of the UN Security Council with regard to their (i) selective engagement with feminist ideas, (ii) tendency to invoke ‘protective stereotypes’ of women and (iii) poor accountability mechanisms, Otto raises her perspective on the possibilities for combining feminist activism with a critical approach. Otto considers the “institutional reception and management of feminist ideas”134 as an instance of cooption, in which such ideas are divested of their emancipatory content; the use of the term ‘gender’ in the UN system is, for her, one such instance. ‘Gender’ is here read as a synonym for ‘women’s issues’, stripping the term of its focus on contesting the feminine and masculine and so its political content and thus similarly limiting the ‘gender mainstreaming’ project in international law. Furthermore, Otto refers to concerns that the mainstreaming of gender language can serve to become a goal in itself, rather than a means to promote the actual goal of gender equality. This approach provides interesting fodder for the contemplation of whether gender should be included as a listed grounds of persecution in refugee law, engaged with by various authors (discussed further in Chapter 3.1). Crompton provides a rare historical analysis of the impact on lesbian women of homophobic criminal legislation which, though not undertaken with a queer asylum case in mind, usefully                                                  133 Romero, supra note 6 at 190–196. Romero criticises the “troubling assignment of ‘proper’ objects of study (gender here, sexuality there); [which] obscures significant currents of feminist and queer thought; and falsely assumes possible an identifiable yet general distinction between feminist theory and queer theory.” Romero instead advocates for an open and undefined approach to both feminist and queer studies, not conceived of as project-specific, in the interests of preserving a democratizing role and possibilities for transformation. 134 Dianne Otto, “The exile of inclusion : reflections on gender issues in international law over the last decade” (2009) 10:1 MELBOURNE JOURNAL OF INTERNATIONAL LAW 11–26 at 13. 71  informs an understanding of state persecution directly relevant to queer women’s claims. Crompton provides a historically-based rebuttal to assumptions that male homosexual acts only were targeted. The conflation of sexual-act prohibitions with the short-hand: ‘sodomy laws’, which in turn is read as referring to male same-sex sex acts only, is effectively displaced by Crompton’s history of the laws, their origins, uses and her discussion of some of those persecuted by them. Crompton highlights how assumptions that criminal laws historically ‘ignored’ lesbian acts are based upon laws specific to England which are not historically or geographically representative. Crompton illustrates the applicability and indeed historical application of legislation (with death penalty provisions) prohibiting ‘acts contrary to nature’ to lesbian sex acts. She also canvasses the existence of laws specifically aimed at women and even of the history of references to ‘sodomy’ as including references to women ‘abusing’ other women.135 Recognising the way in which criminal sanctions can and have impacted upon queer women is an important counter-narrative in support of this dissertation’s refugee law analysis and feminist advocacy position which is doubly endangered by a male-oriented understanding of both minority sexual orientations and of persecution for reasons of sexual orientation. Other useful insights for this thesis’ subject matter can be gleaned from feminist and queer authors examining institutions of family and the legal expectations and dynamics created thereby. The power of alternative families to challenge and change fixed notions of the ‘proper’ or ‘normal’ family, positively, is raised by many authors. Kapur speaks of the disruptive sexual subaltern potentiality and Butler of the “[d]isruptive effects of kinship variability” for national projects. This challenge is particularly important in light of the transfer and reproduction of                                                  135 L Crompton, “The myth of lesbian impunity: capital laws from 1270 to 1791” (1980) 6:1–2 Journal of homosexuality 11. 72  culture which is associated with heteronormative, child-bearing couples, wherein this national-cultural project implicitly carries norms of racial purity and domination.136  These authors dissect assumptions of ‘difference as danger’ which play out in nation-building projects. These authors demonstrate how such defensive reactions to human variability feed into family laws generally. Specific to my project, the work of these authors provide analytical tools that can be extended into the realm of refugee and immigration laws. Here, nation-building policies in a different incarnation can be identified; reflected in refugee and immigration laws and implicating responses to foreign families.  Some writers go on to consider, amongst the new insiders and outsiders137 the place of polygamy in the ‘hierarchy of illegitimacy’138, and continuing stigmatization of “alternative models of intimacy”.139 Denike particularly broaches the constitution of polygamy as queer within the same-sex marriage debate by referring to the conjuring of polygamy within the anti-same-sex marriage camp as well as the reaction thereto amongst its advocates. Advocates for same-sex marriage invariably distanced their campaign from polygamy-as-a-slippery-slope and polygamy thus becomes a backdrop for staging gay relationships as proper and not queer, underwritten and substantiated through racial and colonial assumptions about the practice of polygamy.140 Racial and post-colonial assumptions about the practice of polygyny are evidenced in the preoccupation with Islamic practices, conspicuous in many writings, including the thorough comparative survey                                                  136 Judith Butler, “Is Kinship Always Already Heterosexual?” (2002) 13:1 differences 14–44 at 22; Ratna Kapur, “De-radicalising the rights claims of sexual subalterns through ‘tolerance’”, (10 June 2010), online: Queer Theory: Law, Culture, Empire <> at 44. 137 Brenda Cossman, Sexual citizens: the legal and cultural regulation of sex and belonging, Book, Whole (Stanford, Calif: Stanford University Press, 2007). 138 Butler, supra note 136 at 23. 139 Susan B Boyd, “Marriage is more than just a piece of paper’ : feminist critiques of same sex marriage” (2013) 8:2 National Taiwan University Law Review 263–298 at 274. 140 Denike, supra note 132 at 138. 73  of Bailey and Kaufman on polygamy. Bailey and Kaufman consistently reference the Muslim populations of the countries canvassed, which works to defuse the otherwise clear relationship between the ban of polygyny and colonialism, impacting African colonies as well as indigenous peoples, including amongst First Nations in Canada.141 Like other authors, Denike draws attention to the need to consider how one participates in other prohibitions/exclusions in the quest for assimilation into institutions such as marriage, although her attention on the prohibition of polygamy within this framework is rare.  The unspoken tension for feminist writers in acknowledging polygamy as queer is that, unlike the positive assumptions of queer challenges to heteronormativity, polygamy is considered an unemancipatory ‘other’. It must be noted, however, that even as relates to queer challenges to heternormativity, many writers discuss the way in which, particularly a focus on same-sex marriage has engendered co-option and claim that state recognition of same-sex relationships has had a de-radicalising effect upon the heteronormative challenge itself. Many queer scholars discuss the need to consider those excluded from the new horizons of state-legitimated ‘normal’/normalized relationships.142 Continuing a challenge to the assumption of queer progressiveness, Boyd offers a critique of the re-inscribed privatization of the costs of social (re)production, which is merely redistributed within the newly legitimated, ‘good’ families.143  The concept of choice and autonomy is a fraught one amongst feminist writers. The subject of choice in feminist scholarship is marked with disagreements over the concept of free choice,                                                  141 Martha Bailey & Amy J Kaufman, Polygamy in the monogamous world: multicultural challenges for Western law and policy (Santa Barbara, Calif: Praeger, 2010). A discussion of the colonial underpinnings of anti-polygynous initiatives can be found in Denike, supra note 132 at 138., 145-146; specifically relating to First Nations in Canada see Sarah Carter’s The Importance of Being Monogamous: Marriage and Nation Building in Western Canada to 1915 (Edmonton: University of Alberta Press, 2008). 142 Butler, supra note 136; Cossman, supra note 137. 143 Susan B Boyd, “Family, Law and Sexuality: Feminist Engagements” (1999) 8:3 Social & Legal Studies 369–390; Boyd, supra note 139. 74  allegations of ‘false consciousness’ countered by criticism of patronising readings of othered women’s choices, and threads of a postmodern disillusionment with notions of choice and agency.144 Nevertheless, polygamy – conceived of in the form of polygyny – is understood by some of those who discuss it in terms of choice, rather than assume duress, as a hard choice, curtailed by “a package of laws, policies and practices that maintain gender inequality”145 and continuing subordinate status.146 Thus irrespective of a stance on agency, polygyny is understood here, for the majority of women in these relationships, to be an effect of and have the effect of classed and gendered social subordination. One of the prominent limitations of much of the literature reviewed above, for the purposes of my study, is the lack of attention to or awareness of foreign queer families. Although Cossman makes the important statement that sexual practices are a central dimension of contemporary citizenship, her discussion revolves solely around the already-citizen.147 She, and other writers, make use of citizenship, and border crossing, as an allegory for the gay-citizen.148 However, reducing immigration and citizenship issues to illustrative terms only overlooks the intersectional experiences of non-normative immigrant families and erases such families from the reader’s                                                  144 See Saba Mahmood, “Agency, Performativity, And The Feminist Subject” in Pieties and Gender (2009) 11. Mahmood proposes a different conception of ‘agency’: "not as a synonym for resistance to relations of domination but as a capacity for action that historically specific relations of subordination enable and create". This conception allows for a contextual and contingent understanding of women’s agency, speaking to the problems of a choice/force binary. Her insight allows for the criticism of an agency-as-resistance discourse, which does not necessarily encapsulate the self-conceptions, motives or desires of the women concerned. Mahmood argues at 25 that “the capacity for agency is entailed not only in acts that resist norms but also in the multiple ways in which one inhabits norms”. 145 Bailey & Kaufman, supra note 141 at 5. 146 Penelope E Andrews, “Who’s afraid of polygamy? Exploring the boundaries of family, equality and custom in South Africa” (2009) 11:2 Journal of Law & Family Studies 303. 147 Cossman, supra note 137 at 5. 148 See also Butler, supra note 136. and Carl F Stychin, “‘A Stranger to Its Laws’: Sovereign Bodies, Global Sexualities, and Transnational Citizens” (2000) 27:4 Journal of Law and Society 601–625., who make use of the non-citizen/’illegal alien’ as illustrative of gay citizen’s disenfranchisement or identity as ‘dangerous stranger’ respectively. 75  vocabulary, resulting in the continued silencing of the ‘outsider within outsider’.149 An exception: Seuffert, unpacks the raced, gendered and class normativity inherent to the immigration reforms in relationship-recognition, revealing how they serve to reproduce rather than disrupt imperial domination and its domesticated, ‘properly sexualised’ subjects.150 My dissertation revolves around women experiencing a confluence of ostracized identities, for whom gender and sexual orientation or different marital arrangements influence their vulnerability yet to artificially separate these identities from the impacts of their foreign status would be harmfully reductive of their experiences in South Africa.  2.5.2 False universalism and Third World counters My focus in this dissertation on polygynous and sexual minority refugee women raises questions of culturally-specific approaches to normative sexuality and family groups. Norms of sexuality and of family formation can be described as culturally specific as opposed to universally shared or neutral. Such an approach where sexual prohibitions or family-type provisions are considered cultural products, often frame the resultant laws as a clash between culture and human rights.151 Nyamu, in her analysis of human rights and development discourse, criticises the way in which conventional gender equality arguments, through their reference and opposition to an impoverished and vague notion of ‘culture’, work to implicitly endorse dominant articulations of                                                  149 Lara Karaian, “The Troubled Relationship of Feminist and Queer Legal Theory to Strategic Essentialism: Theory/Praxis, Queer Porn, and Canadian Anti-discrimination Law”, (15 April 2016), online: Feminist and Queer Legal Theory <> at 389.: "The troubled relationship of feminist and queer legal theory to strategic essentialism: theory/praxis, queer porn, and Canadian anti-discrimination law". 150 Nan Seuffert, “Reproducing empire in same-sex relationship recognition and immigration law reform”, (10 June 2010), online: Queer Theory: Law, Culture, Empire <> at 173–175. 151 See the criticism of this tendency in Celestine I Nyamu, “How should human rights and development respond to cultural legitimization of gender hierarchy in developing countries?” (2000) 41:2 Harvard International Law Journal 381; Seyla Benhabib, The Claims of Culture: Equality and Diversity in the Global Era (Princeton University Press, 2018). 76  culture as being ‘The’ accurate description of social custom. “By distancing themselves from the debate on the construction of culture, proponents of gender equality allow culture to be defined exclusively by those whose view of culture disadvantages women”.152 By referencing colonial-influenced succession and property laws, Nyamu also shows how the focus on culture as the problem can miss the extent to which law can be the problem; highlighting the inter-relationship between formal law and a vague notion of culture and the active role of state apparatus in defining culture.153  Phillips, discussing Southern Africa, reveals how the construction of ‘culture’, when phrasing it as oppositional to gender rights, relies on an imagined immutable culture. This very mode of conceiving of a “congealed tradition” is itself a colonial narrative – used to refer to the barbaric ‘is’ and civilized ‘ought’.154 Furthermore, the selectivity of portrayals of ‘culture’/‘tradition’ is highly political and tends to involve a construction of consensus around the choices which, as remarked upon by Nyamu, best serve the power interests of those involved.155 A conglomeration of normative readings of fixed past versus mobile progress, overlaid on place, is a narrative with echoes in the good-host, bad country-of-origin binary discussed further in Chapter 4.  Moreton-Robinson discusses the way in which Indigenous women’s gender has been (mis)represented, reflecting the ethnocentric notions and gender politics of the writers. Moreton-Robinson critiques the prevalent white, feminist scholarship in which Indigenous women are only written about, not for, by, or with and argues the representations thus produced disempower Indigenous women because of the way in which they become ‘truth’ in feminist discourse and                                                  152 Nyamu, supra note 151 at 401. 153 Ibid. 154 Phillips, supra note 28 at 26. 155 Nyamu, supra note 151. 77  white culture. Lines are drawn between ‘traditional’ and ‘contemporary’ Indigenous women by white anthropologists for whom this line comes to distinguish authenticity or contamination of an a priori ‘race’ and ‘culture’ and against which imagined representation (from which image there cannot be resistance), Indigenous women are measured. This discussion adds another nuance to the idea of stereotypes and the role of ‘culture’ in refugee law adjudications, raised in 2.2 above. Mohanty focuses a critical gaze within feminist discourse, on ‘Western’ feminist writings and academia and their depiction of and impact upon ‘Third World women’. This focus is based on an understanding of the power of (a hegemonic) feminist discourse in producing knowledge and framing solutions which have material effects upon the women they canvass.156 This insight, extrapolated into refugee law scholarship, points to the effects of the production of Western-focused knowledge, whose results include the sidelining of the AU Refugee Convention which, though incorporated into domestic refugee law in South Africa is under-utilised, poorly understood and appears neither sufficiently researched, theorised or taught.  To counter the (Western) refrain wherein patriarchal practices are deemed as emanating from a perceived cultural milieu, such that Third World ‘culture’ is phrased as oppositional to gender rights, some authors argue for pluralism over universalism and allege the infeasibility of universal human rights. This line of argument alleges that supposedly universal norms are not compatible with or natural to all cultures and thus a pluralist approach to the ‘international’ is more tenable.157 However, other authors seek to rescue the universal normativity of human rights, often based on the understanding that the concept of universal human rights is a flawed                                                  156 Mohanty, supra note 8. 157 See generally Brian Z Tamanaha, “Understanding legal pluralism : past to present, local to global” (2008) 30:3 SYDNEY LAW REVIEW 375–411. 78  yet necessary resource in the fight for complex equality. This response consequently involves (i) a critique of the construction of culture in the pluralist argument as well as (ii) a reorientation of the critique of Eurocentric human rights as a problem of essentialism and of purported but insufficient universality, rather than a problem with the concept of universal human rights.158  Benhabib states that anxieties about a universalist approach being ethnocentric or falsely eliding differences, itself relies on falsely homogenous views of Western and non-Western cultures, of an imagined ‘us’ and ‘them’, which misses the complex history of global dialogues across cultures and civilisations and the multiplicity of any ‘identity’.159 Echoes of this international law debate can be seen in relation to the criticism raised above and further elaborated on in this dissertation with regard to a Western bias in the construction of a refugee ‘host’ identity as presumptively a Western host, which fails to take account of changing and changeable state roles and of the ethnocentric concerns which ignore the dynamics of refugee host states of the South. I do not intend to claim that this bias means that Western scholars are not critical of the refugee determination systems, and refugee jurisprudence, of Western states.  Certainly, such critiques dominate this literature. Rather, what I am terming ‘Western bias’ arises because little or no attention is paid to Southern states (especially Southern host states), or Southern perspectives, despite the persistent presence in refugee law of people from the Global South: the refugees themselves. Kapur argues that binaries such as West/Rest, colonizer/colonized, here/there are particularly acute in the arena of sexuality where pleasure, desire and agency are associated with the West and violence, victimization and impoverishment make up the Third World gendered and sexual                                                  158 Benhabib, supra note 151; Mullally, supra note 12. 159 Benhabib, supra note 151. 79  subject.160 Denike refers to “the ethnocentric notion that the (mis)treatment of women is a measure of the backwardness and incivility of other cultures”.161 This binary thinking and construct of linked connotations is repeated and exacerbated by the false host/country of origin distinction perpetuated by refugee law scholarship, criticised in Chapter 4 herein which links up the multiple problematic assumptions attached to the label ‘host’ as both beneficent and located exclusively within the West.  The international human rights law ‘abolitionist approach’ (with regard to ‘harmful traditional or cultural practices’, which phrase includes polygyny) is criticised by Nyamu as being state-centric and shaming, both of which aspects hinder the sustained engagement with communities required for effective social transformation. A holistic, context-rich understanding of practices and thus comprehensive and creative solutions are blocked. Furthermore, this approach assumes an absence of women’s rights within custom or local practice, silencing the potential for alternative readings of culture to themselves offer solutions that can be owned by the community and instead opening the way for accusations of cultural imperialism. The proposed solution is the necessity for parallel processes of internal discourse and cross-cultural dialogue to locate the legitimacy for human rights principles within all cultures. Nyamu advocates for a “critical pragmatic engagement with the politics of culture” which considers both the flexibility and variation of custom to combat ‘cultural’ justifications for gender inequalities. Positive opportunities within cultural and religious traditions exist and should be used to advance gender equality arguments rather than dismissing culture and/or religion as only ever a negative influence.                                                  160 Kapur, supra note 136. 161 Denike, supra note 132. 80  In a similar vein, Benhabib argues for the “recognition of the radical hybridity and polyvocality of all cultures”.162 An understanding of the ongoing change within and multiple and alternative readings of ‘a culture’ provides room for “insider-methodologies”163 to work towards gender-positive change as coming from within rather than imported into any culture.  Several writers who support the ideal of universal rights respond to the pluralist case by arguing that the universality of human rights does not require uniformity in approach.164 This perspective is of particular relevance to refugee law which, birthed at the level of international human rights conventions, is domestically, and differently, incorporated and implemented. International law authors such as Charlesworth promote, rather than a distinctive morality, the reference to distinctive experiences which are missing from the discipline of international law and thus prevent its universal validity.165  As an ardent believer in the importance of refugee law in the interests of protecting vulnerable people and as a tiny yet strong instance of true freedom of movement, as well as of the utility in this regard of refugee law’s claim to international normativity, my project falls very much in line with this logic for adding distinctive and missing experiences. The many applicable insights of these TWAIL-feminisms can be usefully applied to my project. However, an additional step of extrapolation is necessary to consider their application to the ‘Third World Woman’ who is othered yet remains within the Third World as a refugee.                                                   162 Benhabib, supra note 151 at 25. 163 Mullally, supra note 12. at (xlii). 164 See for instance Bailey & Kaufman, supra note 141; Nyamu, supra note 151; Mohanty, supra note 8 at 7. who advocates for a “communicative, in-process understanding of the “we” ... the result of active struggle to construct the universal on the basis of particulars”. 165 Charlesworth, Chinkin & Wright, supra note 4. 81  This research thus responds to the gaps and blind spots within refugee law scholarship which I have canvassed in this literature review, and has been fortified by multiple disciplinary perspectives and approaches which complement one another by at once filling gaps, exposing different insufficiencies of their own and providing critical insights in comparison with one another. This literature review has set out the many conceptual building blocks and ideas which have percolated this thesis. It is also an acknowledgment of, and expression of gratitude for, all that I have learnt from these very different authors and an interdisciplinary approach to research. 82  Chapter 3: The Lay of the Law  This chapter presents the background context to the research project, giving an outline of the relevant aspects of refugee law doctrine pertinent to the analysis of data findings which are presented in Chapters 4 and 5 to follow. The refugee-law frame of reference is laid out, working from the international, to the domestic South African setting. Beginning with the international refugee law sketched below, this section outlines the legal and academic refugee law issues which form the architecture for this study and against which the difference and sameness of a South African host study locale can be measured.  3.1 Alphabet Soup: PSGs, LGBTs - Add Gender and Stir It is customary in refugee law studies to begin with a legal analysis of ‘who is a refugee’. This section introduces some of the more technical academic debates which spin off this definitional concept in relation to gender and sexual orientation as impacting refugee identity and refugee claims, so as to ground the discussion in the chapters to follow.  The United Nations Convention Relating to the Status of Refugees (the UN Refugee Convention) is the international law launch pad for all modern domestic refugee legislation as well as the starting point for subsequent regional binding and non-binding refugee law initiatives. As such, the UN Refugee Convention has been very thoroughly reviewed in the academic literature.166                                                   166 See for instance the work cited above by Goodwin-Gill & McAdam, supra note 53; Hathaway & Foster, supra note 23., and the work of the UNHCR’s legal protection division itself. 83  The term “refugee” is defined in Article I of the UN Refugee Convention. Article 1A(2) (edited to exclude the original temporal limitation which restricted the notion only to World War II refugees) reads: For the purposes of the present Convention, the term “refugee” shall apply to any person who:  [...] owing to wellfounded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country; or who, not having a nationality and being outside the country of his former habitual residence as a result of such events, is unable or, owing to such fear, is unwilling to return to it. Although national refugee legislations, adjudications and many authors typically discuss the granting of refugee status, Hathaway has consistently argued that the term ‘refugee’ applies to a person who meets that definition, at the moment they meet the criteria of the definition, irrespective of the asylum system machinations.167 Refugee status is thus recognised through domestic (or UNHCR) processes, not granted by them. The analysis presented in this dissertation adheres to this understanding of a refugee as someone factually meeting international law criteria, albeit prior to, contrary to, or even subsequent to a determination of their status through an asylum system. This approach to the refugee definition in the UN Refugee Convention and, for the purposes of refugee law in South Africa, to the extended AU Refugee Convention definition, paves the way for a different kind of examination of refugee status. One that is capable of contemplating a depth of experiences which encompass both the de facto and de jure refugee as well as consideration of the limitations which create a disjuncture between factual and officially recognised refugees.                                                  167 See Hathaway & Foster, supra note 23 at 1.which restates this position and also references Hathaway’s previous work. 84  Contesting our ability as legal scholars to broaden our understanding of variety and nuance amongst lived experiences is the practical need to classify, summarise and, ultimately, stereotype. Law and legal systems are beholden to the need to classify and refugee law is typical in this respect; having taken a description of a category: ‘refugee’ and, referencing the definition found in the UN Refugee Convention – “persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion” – using this description as a closed list. This closed list has, in turn, become a tick-box of identity categories in the hands of officials and administrators, with much academic and judicial muscle having been put behind a new list of identity categories to definitively populate the daunting ‘other’ that is “members of a particular social group” (PSG).   Significant and important work has thus been geared towards populating ‘members of a PSG’ with descriptors attempting to capture particularly vulnerable, persecuted people who are thought not to necessarily have a claim based on fear of persecution for reasons of their race, religion, nationality or political opinion. Membership of a PSG is typically the dominant category applied to the asylum cases of many women and of sexual minorities, whose defiance of gendered and heteronormative social norms are seldom read as political. Thus PSG has been read to include those persecuted for reasons of their sexual orientation and those persecuted for reasons of their gender, or gender expression.168  The resulting descriptors of ‘the group’, which scholars have identified in Western host administrative or judicial decisions on asylum applications, have been criticised for their sometimes elaborate and bizarre definition, particularly as pertains to women. Rather than simply                                                  168 James C Hathaway & Michelle Foster, “Membership of a Particular Social Group: Discussion Paper No. 4 Advanced Refugee Law Workshop International Association of Refugee Law Judges Auckland, New Zealand, October 2002” (2003) 15:3 International Journal of Refugee Law 477–491 at 478–479. 85  describing the relevant social group as ‘women’ where a female refugee claimant fears persecution for reasons of their gender, Foster in a chapter on the challenges of ‘particular social group’, refers to several examples in multiple Western jurisdictions of overly narrowed definitions of the group in question. This includes one Australian judge who defined a woman refugee’s social group as:  A particularly vulnerable group of married women in Pakistan, in dispute with their husbands and their husbands’ families, unable to call on male support and subjected to, or threatened by, stove burnings at home as a means of getting rid of them yet incapable of securing effective protection from the police or agencies of the law.169 Such narrow definitions become circular and run afoul of the interpretive rule that a PSG should not be defined by the persecution, lest the description of a ‘refugee’ becomes: any person who, owing to wellfounded fear of being persecuted for reasons of being a member of a persecuted group...170 These unnecessarily detailed ‘group’ descriptors desperately attempt to restrict the group label to as few persons as possible, reflective of host state’s politically-infused xenophobia which produces an anxiety, frequently coded in the decisions as a ‘floodgates’ argument.171 Amongst those advocating for the recognition of refugees fleeing gender and sexual orientation-based persecution, there has been discussion whether the UN Refugee Convention or even domestic legislation would be improved by explicitly adding these grounds of persecution.172 This has been particularly discussed as regards women refugees fleeing gender-based persecution given the overly detailed and circular group definitions that have emerged in much                                                  169 Foster, supra note 73 at 30. citing Khawar 2002, [129]. 170 Aleinikoff, supra note 62 at 286. citing the case law established by Applicant A. and Another v. Minister for Immigration and Ethnic Affairs and Another, High Court of Australia, (1997) 190 CLR 225; 142 ALR 331, at p. 341. 171 Foster, supra note 73 at 30. 172 Anker, supra note 57 at 139. 86  of the studied asylum cases.173 Many of those advocating for women refugees and for better adjudication in their cases have argued, however, that simply adding gender to the list of grounds of persecution would not aid a properly gender-sensitive approach to decision making in asylum cases.174   To add gender to the UN Refugee Convention would elevate gender-based persecution to an equal ground to the other listed grounds such as race and political opinion and remove any debate on the subject. However, most writers do not advocate for this change, in part due to a general fear amongst refugee law advocates that any revision of the UN Refugee Convention could result in a step backwards and loss of protections; a belief that to add ‘gender’ could potentially marginalize women further as not having political opinions, being activated by religion or having ‘racial presence’; and, predominantly, a cynicism that simply adding gender would be insufficient since the majority of hurdles facing women within the legal asylum process originate with the attitudes and interpretations of decision-makers rather than with the black-letter law.175 Simply adding in tokenistic wording and a reliance on amended, yet narrowly understood lists are indeed insufficient for achieving inclusive and appropriate protection goals. Dianne Otto identified a similar problem to the limitations consequent to simply ‘adding gender’ as a listed grounds of persecution within the realm of human rights and the United Nations system wherein an increase in the use of the term ‘gender’ and specialized sub committees                                                  173 Arbel, supra note 91 at 745. 174 Like most writers, in discussing the tactical merits of augmentation (including the importance of naming) versus reinterpretation of existing categories, Macklin falls on the side of interpretation. Macklin, supra note 67 at 258. describes the former approach as insufficient to address the gender concerns raised, and having the tendency to create the assumption “that every case of gender persecution is persecution because of gender” as well as potentially marginalizing women further as not having political opinions, being activated by religion or having ‘racial presence’. Anker, supra note 57 at 139., takes the position that gender should be read into the interpretation of refugee law, rather than requiring the amendment thereof and advocates for a gender-mainstreaming approach as having “surprising normative effect”. 175 Arbel, supra note 91; Macklin, supra note 67; Buscher, supra note 88. 87  dedicated thereto have more often sidelined these ‘women’s issues’ than actually mainstreamed gender concerns.176  These arguments in refugee law scholarship are typically made in a way that is anticipatory of adding gender as a listed ground in refugee law, rather than discussions of inclusive amendments already made, such as in South Africa. A chief detraction of the literature on this issue is thus that most writers discuss ‘adding gender’ in the hypothetical. However, their arguments that this would ultimately be insufficient to cater for the concerns they raise would be buttressed if they were to refer to one of the largest asylum systems in the world, which also has in fact ‘added gender’ to the domestic refugee legislation, South Africa, yet they routinely fail to do so.177 What is thus interesting in the South African context, yet which has generally failed to catch the attention of authors focused only on Western host refugee law applications, is that PSG membership is itself detailed with explicitly listed example groups in our legislation. The South African Constitution is a relatively new democratic Constitution and has received much international acclaim for its thoughtful and progressive wording.178 The Constitutional anti-discrimination provisions expressly prohibit unfair State discrimination including on grounds of gender, sex or sexual orientation.179 This modern and express anti-discrimination position with regard to gender and sexual orientation has consequently influenced the South African Refugees Act. A PSG is further defined in the South African Refugees Act: “‘social group’ includes,                                                  176 Otto, supra note 134 at 11. 177 See Julie Middleton’s lonesome exposition on this issue in South Africa: Julie Middleton, “Barriers to protection: gender-related persecution and asylum in South Africa” in Ingrid Palmary, Professor Erica Burman & Peace Kiguwa, eds, Gender and Migration: Feminist Interventions (London, UNITED KINGDOM: Zed Books, 2010) 67. 178 See Rosalind Dixon & Theunis Roux, eds, Constitutional Triumphs, Constitutional Disappointments: A Critical Assessment of the 1996 South African Constitution’s Local and International Influence (Cambridge: Cambridge University Press, 2018) at 1. 179 Constitution of the Republic of South African, 1996 section 9(3) “The state may not unfairly discriminate directly or indirectly against anyone on one or more grounds, including race, gender, sex, pregnancy, marital status, ethnic or social origin, colour, sexual orientation, age, disability, religion, conscience, belief, culture, language and birth.” 88  among others, a group of persons of particular gender, sexual orientation, disability, class or caste”.180  Thus, at least in terms of the written legislation, if not in its actual application, refugees in South Africa having fled gender or sexual-orientation based persecution should not encounter debate as to these groups being each a PSG for the purpose of refugee status determination. In spite of this, and likely influenced by the Western-focused refugee law resources on the subject, I encountered South African-based practitioners who referred to gender or sexual orientation-based persecution grounds as ‘unspecified’ or ‘unlisted’ grounds.181   Furthermore, an appropriate understanding of ‘gender’ appears absent from the asylum process. Middleton criticises the approaches apparent in South African first instance refugee status adjudication, particularly as regards assumptions of ‘private’, family harms of domestic violence and much sexual orientation-based persecution. She points to similar dismissals of gender-based persecution amongst higher level officials, such as the previous Chairperson of the Standing Committee for Refugee Affairs who said domestic violence was difficult to fit into a definition of persecution.182 A strategic litigation specialist interviewed commented that, generally, “gender hasn’t really been well defined in the South African law.”183                                                  180 Refugees Act 130 of 1998 (hereafter “the Refugees Act”) available at 181 Cote, supra note 32.: “what we’ve tried to do is say that we use the ’51 Convention and the ‘social groups’, but the new amendments to the Refugees Act is going to allow gender to be a specific area of persecution, so there might be a place …” Alicia Raymond, Interview with Alicia Raymond, Head of Refugee Unit, Wits Law Clinic (2016), Johannesburg.: “[in cases of clients who fled sexual orientation-based persecution] we specifically make it on a specific ground – membership of a Particular Social Group, because it doesn’t have membership of a specific sexual orientation as one of the grounds in the Refugees Act, so we put it under that ground of membership of a social group.” 182 Julie Middleton, Barriers to Protection: Gender-Related Persecution and Asylum in South Africa (Master of Arts, University of the Witwatersrand, 2009) [unpublished] at 49.: “The Chairperson of the Standing Committee said that they have seen a number of domestic violence cases, however such cases are not seen as fitting under the definition of a refugee, and so are usually rejected. “Domestic abuse we have difficulty in setting aside the RSDO 89  In South Africa both the UN and AU Refugee Conventions are binding and incorporated into the domestic legislation. A common ground for refugee applications, applying to those from certain countries who may not want or feel able to disclose their sexual orientation or gendered experience of persecution is to rely essentially on country conditions; the AU Refugee Convention’s reference to: [E]very person who, owing to external aggression, occupation, foreign domination or events seriously disturbing public order in either part or the whole of his country of origin or nationality, is compelled to leave his place of habitual residence in order to seek refuge in another place outside his country of origin or nationality184 In reality, polygynous and queer refugee women in South Africa therefore have different opportunities and options for interacting with the asylum system which circumvents much of the discussion on PSGs above. South Africa, as a case study, could, for this reason, usefully supplement the debate on the necessity or utility of including gender as a listed ground in the refugee definition, as illustrative of an alternative path. The following section takes a closer look at the peculiarities of the South African asylum system, detailing the applicable legislation and moving beyond the written legislative presentation of the asylum system to a discussion of its (dys)functioning and the assessment of the system by those service providers who work with it.                                                                                                                                                               decision. We have a number of domestic violence cases, but the domestic violence is not easily covered by the definitions. So we would get a case where the woman claims her husband beats her up or the husband’s family doesn’t like her or something like that, but seeing her as a particular social group and suffering domestic violence is difficult to currently bring into the Refugees Act. Those that we have seen have been decided as manifestly unfounded.” Similar to the previous chapter, where RSDOs implied that rape outside the time of war is too normal to be persecution, here the Chairperson of the Standing Committee diminishes the harmful impact of domestic violence. He compares “her husband beats her up” to “the husband’s family doesn’t like her or something like that,” trivialising the significance of a claim based on domestic violence. Such claims are rejected, not based on the merits of the cases, but because violence in the home or family is perceived as private and accepted as too commonplace to be persecution.” 183 Cote, supra note 32. 184 AU Refugee Convention, Article 2. 90  3.2 Over the Rainbow: The State of Refugee Law in South Africa To begin, this section provides a brief account of the actual processes, as determined by South Africa’s Refugee legislation which, at least on paper, dictates what should be the typical progress of an asylum application within the Rainbow Nation.185 This section thus commences with an examination of what the legislation says about refugee law in South Africa, supplemented by what the service providers think about refugee law in South Africa and what this means for the subject groups and this study. Interspersed with a thematic collection of quotes from service providers and refugees on the pitfalls of the South African asylum system by those who experience it, this section will attempt to provide some insight on how the South African system actually works, or fails to, with reference to the thesis subjects as asylum seekers or recognised refugees.  This will also include a discussion of the debates which appear to have arisen around the dubious official data, problematic statistics and their limitations and the confusion surrounding attempts to amend South Africa’s refugee legislation. The interesting combination of international law foundations and domestic law implementation that marks refugee law means that state asylum systems are conceptually related and often draw on foreign jurisprudence for inspiration and analysis of principles yet are often unique in their processes. A background on the official processes of South Africa’s asylum system as well as the way it is experienced by those on the ground who work with it is intended to ground an understanding of what is systemically different and what i