Forced Marriage and the Exoticization of Gendered Harms in United States Asylum Law Columbia Journal of Gender and Law FORCED MARRIAGE AND THE EXOTICIZATION OF GENDERED HARMS IN UNITED STATES ASYLUM LAW JENNI MILLBANK* AND CATHERINE DAUVERGNE** Refugee law scholars and advocates have devoted a great deal of attention to gender-related persecution since the 1980s. The Office of the United Nations High Commissioner for Refugees (UNHCR) first contended that gender was a valid basis for refugee claims in 1985 and released its original guidelines for the protection of women as refugees in 1991.1 Critical scholarship has focussed on refugee law's bias towards recognition of masculinised experiences and on how its Professor of Law, UTS, Australia. This research was supported by a grant from the Australian Research Council. Thanks to Katherine Fallah, Marianna Leishman and Anthca Vogl for their research assistance. Thanks also to the Hastings Center for Gender and Refugee Studies for generously making available copies of Immigration Judge Decisions from their collection and to Connie Oxford for her comments on an earlier draft. Professor and Canada Research Chair in Migration Law, UBC, Canada. U.N. High Comm'r for Refugees [UNHCR], Executive Committee Conclusions, Refugee Women and International Protection, § (K) (1 8 Oct. 1985), available at http://www.unhcr.org/3ac68c43a8.html [hereinafter UNHCR, Refugee Women and International Protection]; UNHCR, Guidelines on the Protection of Refugee Women, U.N. Doc ES/SCP/67 (1991), available at http:// www.unhcr.org/3d4f915c4.html. Columbia Journal of Gender an Law categorizations confine women to narrow, victimized identities. 2 After more than twenty years of concerted effort, one might expect to see an increasingly nuanced refugee jurisprudence concerning gender. With this in mind, we began a study of forced marriage as a basis for refugee claims. 3 While claims of forced marriage or pressure to marry as the, or a, main basis of persecution represent only a tiny portion of refugee claims overall, they provide an illuminating sliver reflecting the major recurring themes in gender and sexuality claims from recent decades. Forced marriage is an important case study of gender in refugee law because it involves longstanding and unambiguous human rights standards, it arises in diverse settings and the harms associated with it take many forms and impact differently depending upon the gender and sexuality of those involved. Our examination of these cases reveals the profound schism between human rights norms and refugee law's protections. While we acknowledge that there are many valid criticisms to be made of international human rights discourse, our analysis in this article reflects our belief that meaningful consent to marriage is nevertheless a gendered human rights issue of vital importance. We are also aware of concern that policymakers and others either completely conflate arranged and forced marriage or else pose (consensual) "arranged" and "forced" marriages as if they are diametric opposites; whereas consent to all kinds of marriages may take place within a 2 See, e.g., Deborah Ankcr, Women Refitgees: Forgotten no Longer?, 32 SAN DIEGO L. REV. 771 (1995); Deborah Anker, Lauren Gilbert & Nancy Kelly, Women Whose Governments are Unable or Unwilling to Provide Reasonable Protection front Domestic Violence may Qualify as Refugees under United States Asylum Law, II GEO. IMMIGR. L.J. 709 (1997); Audrey Macklin, Cross-Border Shopping for Ideas, 13 GEO. IMMIGR. L.J. 25 (1998); THOMAS SPIJKERBOER, GENDER AND REFUGEE STATUS (2000); HEAVEN CRAWLEY, REFUGEES AND GENDER: LAW AND PROCESS (2001); Karen Musalo & Stephen Knight, Steps Forward and Steps Back: Uneven Progress in the Law of Social Group and Gender-Based Claims in the United States, 13 INT'L J. OF REFUGEE L. 51(2001). 1 As this study draws on an international data set, we use the term "refugee," which is defined in international law and is used consistently throughout our case set. The term "asylum" is perhaps more common in the United States. Columbia Journal of Gender and Law continuum of pressure and coerciQn. 4 In this article'we intend "forced marriage" to include any marriage in which one or both participants have been deprived of the opportunity of free or meaningful consent through threats, including emotional and economic threats, pressure or coercion. Our research has affirmed our understanding that refusal to marry is a flashpoint for expressing non-conformity with expected gender roles for heterosexual women, lesbians and gay men. We proceed from the premise that the state has a role in protecting, and indeed a duty to protect, consent to marriage. This role extends to responding to claims for assistance from citizens and, in some circumstances, non-citizens. This paper presents results from our study of 168 refugee decisions where part of the claim for refugee protection concerned actual or threatened forced marriage. We gathered every decision available in English that meets these criteria during the past fifteen years from Australia, Canada, the United Kingdom and the United States (the "receiving countries"). 5 In the present discussion, we highlight our findings from the cases from the United States (the American Case), while detailed findings regarding the broader intemational data set are published elsewhere. 6 While there are notable differences in the cases arising from each receiving country we studied, the American cases stand out as a group distinct from the rest. We found a marked reticence on behalf of American decision makers to grapple with gendered harms in general and forced marriage in particular. Where the American cases do analyze harm as gendered, the discussions are markedly more focused on exoticized elements, such as foreign cultural practices that tend to distance and objectify women, than do decisions from other 4 See, e.g., Sundari Anitha & Aisha Gill, Coercion, Consent and the Forced Marriage Debate in the UK, 17 FEMINIST LEGAL STUD. 165 (2009); Anne Phillips & Moira Dustin, UK Initiatives on Forced Marriage: Regulation, Dialogue and Exit, 52 POL. STUD. 531 (2004). 5 We did not find any decisions fitting these criteria from New Zealand. 6 Catherine Dauvergne & Jenni Millbank, Forced Marriage as Harn in Domestic and International Law, 73(l) MOD. L. REV. 57 (2010). Columbia Journal of Gender an Law jurisdictions. The American decisions also tend to describe such practices in prurient detail. Furthermore, the American cases are notably more insular than those from other countries. Among the forty-eight American decisions in our data set, we did not find a single reference to a non-American decision or to an international human rights standard. This may be the norm in American refugee law, but it is certainly not the global norm and is one of many factors demonstrating that the United States' asylum law is alarmingly out of step with developments elsewhere. Our analysis treads a fine line between a temptation to generalize and the impossibility of doing so. In the United States and the United Kingdom in particular, we have access to a very small number of decisions compared to the total number of refugee determinations made during this time period. Our American analysis is limited especially by scarce access to Board of Immigration Appeals (BIA) decisions. The BIA, the major refugee decision making agency in the United States, benefits from considerable legislative and judicial deference. Limited access to these rulings means a serious lack of transparency in American asylum law. We have no reason to believe that the decisions we have found are atypical of American jurisprudence generally (as even in this electronic age it remains the case that important and leading decisions are reported and that the very best and very worst of decisions become well known in advocacy circles), but we cannot be certain. We cannot, of course, draw any quantitative conclusions about United States decision making. We present this analysis for what it is: a glimpse of what it is currently possible to know about American refugee decisions regarding forced marriage. The aim of this paper is to analyze the American decisions against the comparative backdrop of our international data set. We present this analysis in four steps. The first section compares recent attention to forced marriage as a domestic policy issue in European law with fledgling American developments. The second section outlines the framework of American asylum law and policy with regard to forced marriage through the development of gender analysis guidance documents incorporating international human rights standards. This section also explores the failure to integrate these standards through case studies of two high-profile cases: In re Kasinga in 1996 and Gao Columbia Journal of Gender and Law v. Gonzales in 2007. 7 Following this, we examine how the key requirements of refugee jurisprudence-persecution, particular social group, and nexus-have been approached in the American forced marriage cases. Through this examination we compare American cases with those from the international data set. Finally, we turn to how successful American claims differ (or not) from successful claims elsewhere. I. COMPARING AMERICAN AND EUROPEAN FORCED MARRIAGE DEVELOPMENTS IN DOMESTIC IMMIGRATION AND FOREIGN POLICY Public and political concern over forced marriage emerged in Europe in the early 1990s, at least a decade before any interest in this issue developed in the United States. The policy trajectory varied in different European countries, but in each case it arguably arose from an implicit understanding of vulnerable brides as "ours" (nationals or dual nationals), while imposed grooms are "theirs" (migrant spouses). This generated an intense early focus on immigration restrictions as the "answer" to the problem of forced marriage. 8 This is distinct from the contemporary American discourse where, by contrast, forced marriage concerns center almost exclusively upon child marriage, an issue which is presented as being geographically confined to "developing countries." We will consider the importance of this contrast after briefly surveying the European developments. Of all European countries, Denmark directed its reform energies concerning forced marriage most explicitly and continuously towards immigration restriction. Legal changes limiting family reunification immigration provisions began in 7 in re Kasinga, 21 1. & N. Dec. 357 (BIA June 13, 1996); Gao v. Gonzales, 440 F.3d 62 (2d Cir. 2006). 8 The use of immigration rcstrictions by European countries to address forced marriage has been strenuously criticised as anti-Muslim, intertwined with the war on terror and unduly punitive toward immigrant women. See, e.g., Sherene Razack, Imperiled Muslim Women, Dangerous Muslim Men and Civilised Europeans: Legal and Social Responses to Forced Marriages, 12 FEMINIST LEGAL STUD. 129 (2004); Amrit Wilson, The Forced Marriage Debate and the British State, 49 RACE & CLASS 25 (2007). Columbia Journal of Gender an Law Denmark in 1998 and were tightened again in 2000, 2002 and 2004. 9 The impact of such immigration law changes reached far beyond forced marriages, but were justified on the basis that the greatest vulnerability was faced by young people with little independence from their families who were being coerced into marriages with overseas-born, often older, spouses from the same ethnic background. Only after most of these restrictive regulations were in place did Denmark produce an "Action Plan on Forced, Quasi-Forced and Arrangement Marriages" with proposals for broader empowering strategies such as counselling, education for teachers and case workers, residential facilities and a research program. 10 In contrast, Norway pursued an inverse trajectory, beginning in 1998 with an "Action Plan" that did not focus on immigration restriction (indeed it suggested liberalizing immigration policies might actually reduce incentives to forced marriage)." The initial 1998 Norwegian plan focused on education and support for victims. Immigration law changes were not introduced in Norway until 2003, and were minimal in comparison with Denmark. In the same year, a specific criminal provision on forced marriage was introduced in Norway, a move, replicated by Germany in 2005 and Belgium in 2007.12 During the same time period, France made several 9 Anja Bredal, Tackling Forced Marriage in the Nordic Countries: Between Women's Rights and Immigration Control, in 'HONOUR': CRIMES, PARADIGMS AND VIOLENCE AGAINST WOMEN 342-45 (Lynn Wclchman & Sara Hossain cds., 2005). 10 THE DANISH GOVERNMENT, THE GOVERNMENT'S ACTION PLAN FOR 2003-2005 ON FORCED, QUASI-FORCED AND ARRANGED MARRIAGES (2003), available at http://www.nyidanmark.dk/NR/rdonlyrcs/ 05ED3816-8159-4899-9CBBCDD2D7BF23AE/O /forced-marriages.pdf. Bredal, supra note 9, at 33-35 12 Brigittc Clark & Claudina Richard, The Prevention and Prohibition of Forced Marriages-a Comparative Approach, 57 INT'L & COMP. L.Q. 501, 503 (2008). Columbia Journal of Gender and Law changes to procedural requirements to ensure genuine consent for marriage. 13 The United Kingdom provides an interesting example of the development of a multifaceted approach shaped by community and feminist involvement. While initial action focused on immigration, including raising the age requirements for spousal visas, it rapidly moved in a number of other directions. Rather than criminalization, the United Kingdom created a range of new civil remedies under the Forced Marriage (Civil Protection) Act, which passed in 2007 and came into effect in December 2008.14 The centrepiece of this law is the creation of a "forced marriage protection order" designed to protect a person at risk of forced marriage or who has already been forced to marry.' 5 The legislation creates a flexible tool and strenuously reinforces a proactive role for the courts in confronting and potentially averting forced marriage. A key aspect of the United Kingdom's approach was the establishment in 2005 of the Forced Marriage Unit (FMU), a joint initiative of the Foreign and Commonwealth Office and the Home Office. The FMU agenda reflects a feminist and community-informed understanding that forced marriage is a harm based upon power imbalances concerning gender and sexuality. The Forced Marriage Unit information brochure for the lesbian and gay community states: A forced marriage is conducted without the consent of one or both people, and pressure or abuse is used. This could include both physical pressure (when someone threatens to or actually does hurt you) or emotional pressure (for example, when someone tries to make you 13Id. 14 Forced Marriage (Civil Protection) Act, 2007, c. 20 (U.K.). Scotland recently passed similar legislation: see Forced Marriage etc. (Protection and Jurisdiction) (Scotland) Act 2011. See also the recent decision of the Supreme Court of the UK striking down the age limit of 21 for marriage visas: R and R v. SSHD [2011] UKSC 45. 15 Forced Marriage (Civil Protection) Act, supra note 14, at pt. I. Columbia Journal of Gender an Law feel that your sexuality brings shame on your family) to get married. 16 Policy initiatives include roles for schools and teachers, health care professionals, social workers, police, community organizations and individuals in being alert to and responding to situations of forced marriage. These initiatives articulate a "protective" role of the state that extends to proactive service .provisions. A statutory guidance document accompanying the new legislation states that in the first nine months of 2008, 1,300 "instances of suspected forced marriage" were reported to the FMU. 17 In terms of ongoing casework, the FMU reports that it currently deals with around 1700 cases annually. 18 The FMU has also developed a unique capacity to act overseas to assist Britons and dual citizens facing forced marriage. The FMU coordinates with consular staff abroad to intervene directly when the unit or consular staff are notified that someone is at risk of forced marriage, or has been forced to marry overseas. By 2008 the unit had reportedly assisted with 180 such cases overseas. 19 What the European initiatives have in common, whether anchored in immigration, criminal or family law, is a central concern about forced marriages taking place within Europe. The United Kingdom has gone further than other states by extending this protective concern beyond its citizens to 16 FORCED MARRIAGE UNIT (U.K.), GUIDE TO FORCED MARRIAGE FOR LGBT PEOPLE 2 (2007), available at http://www.fco.gov.uk/rcsourccs/en/pdf/ foced-marriage-lgbt. 17 HM GOVERNMENT, THE RIGHT TO CHOOSE: MULTI-AGENCY STATUTORY GUIDANCE FOR DEALING WITH FORCED MARRIAGE 5 (2008), available at http://www.fco.gov.uk/resources/en/pdt73849543/forced-marriagc- right-to-choose. '9 Foreign and Commonwealth Office, Forced Marriage Unit, http:// www.fco.gov.uk/cn/travcl-and-living-abroad/when-things-go-wrong/forced- marriage/ (last visited Nov. 21 2011). 19 Owen Bowcott & Jenny Percival, Bangladeshi 'forced marriage" GP due back in Britain tomorrow, THE GUARDIAN, Dec. 15, 2008. Columbia Journal of Gender and Law include even individuals who are not citizens and not being forcibly married within the United Kingdom, but who are connected to it only by a residency right.20 Contemporary American concern about forced marriage starts from precisely the opposite position. Current initiatives in the United States are focused on non-citizens located only in so-called "developing" countries and affected by a "harmful traditional practice. '21 Considering the shape this issue has taken in the domestic policy of the United States illuminates how the understanding of forced marriage has been limited, and arguably misunderstood, in American asylum law. There do not appear to be any domestic non- government organizations (NGOs) staging campaigns about preventing forced marriages within the United States, nor is there an academic literature discussing social, political and legal aspects of forced marriage. In searching for a domestic discourse about forced marriage within the United States, we find a domestic politics of concern about child marriage in foreign countries, as well as a strand of anti-trafficking politics which considers the linkages between human trafficking and forced marriage, but only a few very small signs that a broader European-style engagement may be on the horizon. Concern in the United States about child marriage has crystallized in proposed legislation under the title International Protecting Girls by Preventing Child Marriage Act. Parallel bills were introduced into the House of Representatives and Senate in the spring of 2009.22 The House passed its version on 10 June 20 1d. 21 International Protecting Girls by Preventing Child Marriage Act, H.R. 2103, 111 th Cong. (2009) §2 (introduced Apr. 27, 2009); International Protecting Girls by Preventing Child Marriage Act, S. 987, 111 th Cong. (2009) §2 (introduced May 6, 2009). 22 H.R. 2103, supra note 21; S. 987, supra note 21. The bills are identical in substance but present material in a differing order with the result that section numbers are not identical. Columbia Journal of Gender an Law 2009.23 The original House and Senate bills state in their respective "Findings" sections that "child marriage, also known as 'forced marriage' or 'early marriage' is a harmful traditional practice that deprives girls of their dignity and human rights," 24 and is framed with statistical information regarding child marriage worldwide. 25 While girls are named in the bill's title, child marriage is defined as "the marriage of a girl or boy, not yet the minimum age for marriage stipulated in law in the country in which the girl or boy is a resident. '26 The bill authorizes expenditures for a variety of assistance programs aimed at reducing and eliminating child marriage and gives priority to areas with a high occurrence of child marriage, activities that have proven successful, and pilot projects that agree to share their evaluations. This assistance is to be coordinated with existing foreign aid initiatives. 27 The bill would also require the Secretary of State for Foreign Affairs to develop a multi-year strategy for confronting child marriage, to develop research capacity surrounding the issue and to include information about child marriage in the annual State Department Human Rights Reports. 28 23 This was passed as part of the Foreign Relations Authorization Act for Fiscal Years 2010 and 2011, H.R. 2410, 111th Cong. (2009), §1111. In this version of the bill, the Findings section is eliminated as are some generalized provisions regarding assistance. All specific requirements were carried forward. 24 H.R. 2103, supra note 21, at §2(1); S. 987, supra note 21, at §2(l). 25 Throughout §2, the global prevalence of. child marriage is discussed, eleven countries in Africa and South Asia are named as particular problem areas, and protection against forced marriage in the Universal Declaration of Human Rights is cited. H.R. 2103, supra note 21, at §2; S. 987, supra note 2 1, at §2. 26 H.R. 2103, supra note 21, at §8; S. 987, supra note 21, at §3. 27 H.R. 2103, supra note 21, at §4; S. 987, supra note 21, at §5. 11 H.R. 2103, supra note 21, at §§5-7; S. 987, supra note 21, at §5-7. Columbia Journal of Gender and Law It is unknown at the time of writing whether the bill will become law in the future. 29 The tenor of the legislation is strikingly at odds with our analysis of American asylum decisions in that it is overtly linked to a concern over international human rights standards. In part, this disjuncture may be because it focuses on a distinct subset of the problem of forced marriage. Addressing only child marriage, only developing countries, only "traditional practices," and focusing almost exclusively on girls obviates the possibility of the kind of robust and wide-ranging discussion taking place in the United Kingdom and to a lesser extent elsewhere in Europe. Framed only as an issue of early marriage and the impossibility of consent, it excludes discussion of how forced marriage may be used to control non-conforming sexuality across a broad range of experiences, including adults in their twenties, gay men and lesbians and young women choosing interracial or religious partners. It also excludes discussion of forced marriage in contexts where the practice is sharply modernized (for example involving international travel or the theft of the victim's passport and cell phone). 30 The legislation takes the unnecessary step of pronouncing that "child marriage" is also known as "forced marriage," thus defining away all other aspects of forced marriage. 3' The 2009 legislation has its antecedent in a bill co- sponsored by then-Senator Hillary Clinton in 2006 under the 29 On December I, 2010, Senate passed an amended version of the bill, but upon return to the House it was defeated on December 16, 2010.: see International Protecting Girls by Preventing Child Marriage Act of 2009, S.987, I 1Ith Cong. (2009), available at http://www.govtrack.us/congrcss/ legislation.xpd. 3 0 See, e.g., FORCED MARRIAGE UNIT (U.K.), supra note 16. 31 Although some commentators have made an effort to suggest that the practice is not solely "foreign" or "other," it is the "child" element that is seen to occur within the United States, not the "forced" aspect. That is, while there is some limited recognition that a significant number of teenagers do marry in the United States, there is not yet any concern about forced or coerced marriages occurring "at home." See, e.g., Bojana Stoparie, Anti-Poverty Efforts Face Child Marriage Hurdle, WOMEN'S ENEWS, Aug. 22, 2006, http:// www.womensenews.org/article.cfm/ dyn/aid/283 1. Columbia Journal of Gender an Law title "International Child Marriage Prevention and Assistance Act of 2006.'32 This earlier legislation was similar in its focus on foreign aid initiatives for developing countries, backed up by strategy development and reporting requirements. The most significant difference was its focus on the health risks to girls becoming pregnant and giving birth before adulthood. 33 The bill was introduced by Senator Durbin with a speech on maternal mortality rates.3 4 Given this concern, an exclusive focus on girls was logical. The current version of the legislation focuses more directly on marriage itself, but this antecedent does help contextualize the emphasis on "child" rather than "forced" marriage. The second area where concern about forced marriage arises in American discourse is in the domain of human trafficking. The United States has staked out a leadership role in the increasingly globalized effort to criminalize and eradicate trafficking in persons, and since 2001, the State Department has issued an annual Trafficking in Persons Report assessing trafficking prevalence and prevention efforts in countries around the globe. 35 Since its earliest edition, this report has included some references to women and girls who are trafficked within or 32 S. 3651, 109th Cong. (2006). This bill was referred to the Committee on Foreign Relations and then stalled. The other sponsors were Senators Dick Durbin and Chuck Hagel. See id. 33 A specific aim of the bill was to reduce the global incidence of obstetric fistula. Id. at §6. 34 Senator Dick Durbin, Remarks introducing the International Child Marriage Prevention and Assistance Act of 2006 (Dec. 8, 2006), available at http://thomas.loc.gov/ cgi-bin/query/R?r109:FLD0OI:S61684 (follow "MATERNAL MORTALITY--(Senate - Dec. 8,2006)"). 35 For an overview of this effort, see CATHERINE DAUVERGNE, MAKING PEOPLE ILLEGAL: WHAT GLOBALIZATION MEANS FOR MIGRATION AND LAW 69-92 (2008); James Hathaway, The Human Rights Quagmire of "Human Trafficking" 49 VA. J. INT'L L. 1 (2008). Columbia Journal of Gender and Law across borders for the purpose of forcible marriage. 36 The intensity of the Report's focus on forced marriage has increased somewhat since 2001. Given the very high proportion of Chinese asylum claims in the American data set, it is apposite to point out that the inaugural 2001 report raised a concern about women trafficked into China and through China for the purpose of "arranged marriages. '37 In contrast to the new and fledgling public discourse about child marriage, public and political attention to the issue of human trafficking is well established and sustained in the United States.38 Forced marriage does not have a central place in this discourse but it is recognized as a related issue of concern. Finally, there are also small snippets of evidence of some American policies beginning to reflect awareness of how forced marriage may affect American citizens. The 2005 version of the State Department Foreign Affairs Manual contains a seven-page chapter addressing forced marriage. 39 This text is distinct from the domestic discourses of child marriage and trafficking and has similarities with the activities of the British Forced Marriage Unit. Although the chapter is titled "Forced Marriage of Minors," it opens by observing that "[t]he issue of 36 See generally U.S. DEPARTMENT OF STATE, TRAFFICKING IN PERSONS REPORT (2010), available at http://www.state.gov/g/tip/rls/tiprpt/. 37 U.S. DEPARTMENT OF STATE, TRAFFICKING IN PERSONS REPORT 37 (2001). 31 See Hathaway, supra note 35. See also Joan Fitzpatrick, Trafficking as a Human Rights Violation: the Complex Intersection of Legal Frameworks for Conceptualizing and Combating Trafficking, 24 MICH. J. OF INT'L L. 1143 (2003); Janie Chuang, The United States as Global Sheriff: Using Unilateral Sanctions to Combat Human Trafficking, 27 MICH. J. INT'L L. 437 (2006); Anne Gallagher, Human Rights and the New UN Protocols on Trafficking and Migrant Smuggling: A Preliminary Analysis, 23 HUM. RTS. Q. 975 (2001); Kara Abramson, Beyond Consent, Toward Safeguarding Human Rights: Implementing the United Nations Trafficking Protocol, 44 HARV. INT'L L.J. 473 (2003). 39 U.S. DEP'T OF STATE, FOREIGN AFFAIRS MANUAL VOLUME 7 - CONSULAR AFFAIRS (2005), available at http://www.state.gov/documents/ organization/86822.pdf. Columbia Journal of Gender an Law forced marriages involves more than just child victims," 40 and also states that fifteen percent of victims are male. The first paragraph concludes with strong advice to American diplomats: Cases involving US citizen/national children that come to your attention cannot be disregarded, or simply referred back to their parents. You must take all possible steps to protect the US citizen/national child in these cases. 4 1 The chapter then sets out the legal authorities both for understanding forced marriage as a human rights infringement and for supporting consular action; distinguishes forced marriage from arranged marriage (a distinction often disregarded in asylum cases); 42 and outlines specific actions to be taken. American diplomats are not empowered to confront forced marriage of their citizens as assertively as the British, in part because of the authority granted by the United Kingdom's 2008 Forced Marriage (Civil Protection) Act, but they have nonetheless been given unambiguous guidance about the harm of forced marriage. 40 Id. at 1. 41 Id. 42 The chapter states: Arranged marriages have been a long-standing tradition in many cultures and countries. The Department respects this tradition, and makes a very clear distinction between a forced marriage and an arranged marriage. In arranged marriages the families of both spouses take a leading role in arranging the marriage but the choice whether to accept the arrangement remains with the individuals. Id. at 3-4. Contra Petition for a Writ of Certiorari, Keisler v. Gao, 552 U.S. 801, (2007) (No. 06-1264) (see infra pp. 23-26, where forced marriage is repeatedly characterised as "arranged."). 19.4 Columbia Journal of Gender and Law This advice appears to have been taken to heart by only one American embassy in Dhaka, Bangladesh. The embassy website has a forced marriage page stating: The U.S. Embassy in Dhaka is willing to assist victims or potential victims of forced marriage. If you are an American citizen in Bangladesh, or know an American citizen in Bangladesh, who has been, is being, or fears being forced into marriage against your/their will, please contact the U.S. Embassy in Dhaka 43 The 'FAQ' page gives advice to those citizens already in Bangladesh, to American citizens yet to travel there and to citizens who have already been forcibly married. The page states that some individuals may be eligible for loans from the United States government to help them return to the United States and also advises, "If possible take a cell phone with you, and have the contact number of the U.S. Embassy in Bangladesh stored in it." 44 The State Department also includes a warning about forced marriage as part of its travel advisory for Bangladesh. 45 While Bangladesh is defined by the State Department as a developing country, the concern about forced marriage is addressed to American citizens with twenty-first century sensibilities. In sum, the contrast between American and European approaches to forced marriage reveals a sustained, broader and more detailed engagement with this issue in Europe. Within the United States, with the exception of some isolated references to the possibility of forced marriage of American citizens abroad in 43 Embassy of the United States: Dhaka, Bangladesh, Forced Marriage Homepage, http://dhaka.usembassy.gov/forccd-marriagehome.html (last visited Nov. 21, 201 1). 44 Embassy of the United States: Dhaka, Bangladesh, Forced Marriage FAQ page, http://dhaka.usembassy.gov/forcedmarriagefaq.html 45 U.S. Department of State: Bureau of Consular Affairs, Bangladesh Country Specific Information, http://travel.state.gov/travel/cispatw/cis/ cis_1011 .html (last visited Nov. 21, 2011). Columbia Journal of Gender an Law diplomatic and consular materials, the issue of forced marriage has been subsumed into concerns either about child marriage taking place in foreign countries, where age is used as a blunt proxy for consent (and where child pregnancy has been an overwhelming concern), or about human trafficking with its own strong politic. In this context, forced marriage is overwhelmed and fails to emerge as a distinct concern worthy of separate analysis and action. In the next section, we outline how gender guidance within refugee policy in the United States has developed to acknowledge forced marriage as a gendered harm. Yet, through an examination of key cases, we explain how this policy guidance has in fact been honored far more in the breach than the observance. We explore how female genital mutilation ("FGM") has come to dominate all discussion of gendered persecution in the American refugee context to the exclusion of other, less exoticized, forms of gendered harms. 1I. FORCED MARRIAGE AS A GENDERED HARM IN AMERICAN ASYLUM LAW AND POLICY Nation states implement their obligations under the United Nations Convention Relating to the Status of Refugees (hereinafter "Refugee Convention") differently. At international law, a refugee is someone who: [O]wing to a well-founded 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 a 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 Columbia Journal of Gender and Law unable or, owing to such fear, is unwilling to return to it. 46 On the basis of this definition, international refugee law provides "surrogate" protection for individuals whose country of nationality cannot or will not protect them from certain types of harm. It is clear in the jurisprudence that states are not required to protect their citizens from every breach of an international human rights standard: some breaches constitute persecution and others do not. While the United States is similar to the other countries we discuss in that it has an onshore adjudication system with limited avenues of judicial review from initial administrative decisions, there are several unique features. First, the American system is bifurcated with separate institutions, processes and evidentiary standards for "affirmative" claims (those made proactively by the applicant) and "defensive" claims of asylum, which are made to withhold deportation proceedings. 47 Second, since 1996 there has been a strict statutory requirement of timeliness, with a one-year period for claims to be made and limited exceptions.48 This is in contrast to many other countries 46 United Nations Convention Relating to the Status of Refugees, July 28, 1951, 189 U.N.T.S. 150, art. I(a)(2), amended by Protocol Relating to the Status of Refugees, Jan. 31, 1967, 606 U.N.T.S. 267. 47 In an affirmative claim for asylum the applicant must demonstrate the Convention standard of a "well founded fear" which must be more than a mere possibility but does not need to meet the balance of probabilities. In a defensive or "withholding of removal" claim the applicant must show that it is "more likely than not" they will be subject to persecution. See DEBORAH ANKER, THE LAW OF ASYLUM IN THE UNITED STATES 15, 19, 77 (3 rd ed., 1999); THOMAS ALEXANDER ALEINIKOFF, DAVID MARTIN & HIROSHI MOTOMURA, IMMIGRATION AND CITIZENSHIP: PROCESS AND POLICY (5t , ed., 2003). See also Paul O'Dwyer, A Well-Founded Fear of Having my Sexual Orientation Asylum Claim Heard in the Wrong Court, 52 N.Y.L. SCH. L. REV. 185 (2008). 48 See Karen Musalo & Marcelle Rice, Center for Gender & Refugee Studies: The Implementation of the One-Year Bar to Asylum, 31 HASTINGS INT'L & COMP. L. REV. 693 (2008). In our study, see for example, Matter of A-D-A (Bos., MA Immigration Court, Sept. 19, 2005) (on file with The Hastings Center for Gender & Refugee Studies); Xuan Li Zheng v. Ashcrofl, No. 02-73656, 2002 U.S. App. LEXIS 10776 (9th Cir. June 3, 2002). Columbia Journal of Gender an Law where a time delay may be taken into account as adverse to credibility, but does not prevent the claim being heard on the merits. 49 Third, the judicial review structure in the United States, which has eleven separate numbered federal courts of appeal not bound by each other's rulings, has generated a distinctly chaotic approach to questions of legal interpretation regarding the refugee definition, in particular on how to approach gender and the definition of "particular social group". 50 Further, while the interpretation and application of refugee law is highly politicized in all of the countries under discussion, 51 the United States is unique in its heightened deference to the role of the Executive in decision-making. So, for example, immigration judges who find that an applicant has satisfied the legal standard in an affirmative claim for asylum nevertheless have discretion to deny it. 52 In addition, the Attorney General has the power to issue instructions on legal interpretation of the relevant provisions and to directly intervene to vacate decisions of the Board of Immigration Appeals (BIA). On the whole, we contend that administrative and statutory responses to gendered refugee issues in the United 9 In addition, since 2005, the REAL ID Act includes a statutory power to make negative credibility determinations without regard to whether any inconsistency or inaccuracy is in fact central to the claim. REAL ID Act, 8 U.S.C. § 11 58(b)(I)(B)(iii) (2009). " See O'Dwyer, supra note 47; Fatma Marouf, The Emerging Importance of "Social Visibility" in Defining a "Particular Social Group" and its Potential Impact on Asylum Clains Related to Sexual Orientation and Gender, 27 YALE L. & POL'Y REv. 47 (2008). 51 For example, in recent years, as part of a global trend of refugee receiving nations narrowing eligibility in on-shore claims, Australia amended its legislation to narrow the definition of persecution, while the United Kingdom and Canada included mandatory consideration of certain negative credibility factors in their legislation. See Migration Act 1958 (Austl.) section 91 R; Asylum and Immigration (Treatment of Claimants, etc.) Act, 2004 (UK) section 8; Immigration and Refugee Protection Act, 2001 S.C., ch. 27 § 106 (Can.). 52 For one example of the effect of this policy, see the comments in Manani v. Filip, No. 08-1530, 2009 U.S. App. LEXIS 1980, at *3 (8th Cir. Jan. 28, 2009). Columbia Journal of Gender and Law States have been marred both by delay and a lack of coherence. 53 In 2001, then Attorney General Janet Reno intervened to overturn a 1999 Board of Immigration Appeals decision that women facing domestic violence could not form a "particular social group" and proposed new regulations for gender-related claims under which the case should be decided. When, after six years, these regulations had still not been finalized, the Department of Homeland Security submitted a brief to the Attorney General stating that "married women in Guatemala who are unable to leave the relationship" should be recognized as a social group and directed that the BIA should decide the pending original case on this basis. The regulations still did not eventuate and in 2008 the Attorney General took the step of lifting the original stay so that the BIA could itself resolve the issue. 54 A 2009 brief by the Department of Homeland Security in another domestic violence refugee claim before the BIA, acknowledged the delay of over nine years in producing regulations on gender but insisted that the Department had not "abandoned" the effort, and stated that its new leadership (installed as a result of the Obama Administration) was "considering the best way forward. 55 53 For background, see Karen Musalo & Stephen Knight, Steps Forward and Steps Back: Uneven Progress in the Law of Social Group and Gender-Based Claims in the United States, 13 INT'LJ. REFUGEE L. 51 (2001). 54 Matter of RA, 24 I. & N. Dec. 629 (BIA 2008); see also Gao v. Gonzales, 440 F.3d 62 (2d Cir. 2006) (discussing earlier developments). 55 Supplemental Brief of Department of Homeland Security at 4, n.5, In re L.R. (Apr. 13, 2009), available at http://graphics8.nytimcs.com/packages/ pdf/us/20090716-asylum-brief.pdf. While the brief has been lauded as "open[ing] the way for foreign women who are victims of severe domestic beatings and sexual abuse to receive asylum in the United States," the brief in fact represents a fairly restrictive approach to domestic violence and particular social group. The brief maintains the Administration's refusal to accept gender more broadly as the basis for a refugee claim. Julia Preston, New Policy Permits Asylum for Battered Women, N.Y. TIMES, July 15, 2009, at Al. (LR's claim ultimately succeeded in 2010: see Julia Preston, Asylum Granted to Mexican Woman in Case Setting Standard on Domestic Abuse, N.Y. TIMES, Aug. 12, 2010). See further discussion under "Particular Social Group," infra note 133. Columbia Journal of Gender an Law In contrast to this extraordinary period of delay and ambivalence over domestic violence specifically and gender more broadly, the statutory definition of refugee was amended in 1996 as a result of advocacy by conservative Christian groups to specifically deem forced abortion and sterilization persecution on the basis of political opinion. 56 The statute thereby prioritizes these harms over other forms of gendered persecution,"7 and the United States receives very high numbers of claims from China as a consequence. 58 Our research demonstrates that forced marriage is still not widely accepted in American asylum law as a persecutory harm which may give rise to refugee status. This is particularly shocking given that forced marriage was explicitly addressed in the 1995 Immigration and Naturalization Service ("INS") "Considerations for Asylum Officers Adjudicating Asylum Claims from Women" (hereafter "the INS Gender Guidelines").5 9 In an introductory overview, the Guidelines note that women claimants may face particular forms of harm for "breaching social mores" such as "marrying outside of an arranged marriage," 60 and later in the section on persecution state that forms of harm "that are unique to or more commonly 56 8 U.S.C. § I 101(a)(42) (2009). See also infra note 101. 57 See Matter of Y-T-L, 23 1. & N. Dec. 601 (BIA 2003), discussed in Matter of A-T, 24 I. & N. Dec. 617 (Attorney Gcn. 2008). 58 In 2008 China was the leading country of origin for those receiving grants of on-shore asylum (encompassing both affirmative claims and withholding of removal) in the United States, comprising a staggering twenty- four percent of protection grants. See Daniel Martin & Martin Hoeffer, Dep't Homeland Sec. & Office Immigr. Stat. Pol'y Directorate, Refugees and Asylees: 2008, ANN. FLOW REP., June 2009, at 5, http://www.dhs.gov/xlibrary /assets/ statistics/publications/ois rfa fr 2008.pdf (last visited Feb. 20, 2010). 59 Memorandum from Phyllis Coven, Office Int'l Aff. on Considerations for Asylum Officers Adjudicating Asylum Claims from Women to All INS Asylum Office/rs & HQASM Coordinators (May 26, 1995), (reprinted in 7(4) INT'L J. REFUGEE L. 700 (1995)) [hereinafter INS Gender Guidelines]. 60 Id. at4. Columbia Journal of Gender and Law befall women" include "sexual abuse, rape, infanticide, genital mutilation, forced marriage, slavery, domestic violence and forced abortion." 61 Additionally, the INS Gender Guidelines state that "the evaluation of gender-based claims must be viewed within the framework provided by existing international human rights instruments and the interpretation of these instruments by international organizations. 6 2 The international instruments referenced in the INS Gender Guidelines include the Universal Declaration of Human Rights ("UDHR") and the Convention on the Elimination of All Forms of Discrimination Against Women ("CEDAW"), while the international organizations and interpretations listed include the original UNHCR Gender Guidelines (since substantially revised) and the Canadian Gender Guidelines (which are notably described as "a model for gender-based asylum adjudications"). 63 Current versions of these latter two documents expressly characterize forced marriage as a form of gender- based persecution. 64 The references to more general international law instruments, the UDHR and CEDAW, are also significant, as these characterize the choice of whether and whom to marry as a fundamental human right. The requirement that marriage be undertaken only with the "free and full consent" of both parties first appeared in Article 16(2) of the 1948 UDHR and was later 61 Id. at 9. 62 1d. at2. 63 Id. at 2." 64 IMMIGRATION AND REFUGEE BOARD OF CANADA, GUIDELINE 4, WOMEN REFUGEE CLAIMANTS FEARING GENDER-RELATED PERSECUTION §§ A.I. 4, B (1996). U.N. High Comm'r for Refugees Guidelines on International Protection No. 1: Gender-Related Persecution within the Context of Article IA(2) of the 1951 Convention and/or its 1967 Protocol relating to the Status of Refugees, I 36(vii), U.N. Doc. HCR/GIP/02/01 (May 7, 2002) [hereinafter UNHCR, Gender Guidelines] ("Female claimants may also fail to relate questions that are about 'torture' to the types of harm which they fear (such as rape, sexual abuse, female genital mutilation, 'honour killings', forced marriage, etc.).") See also UNHCR, Guidance Note on Refugee Claims Relating to Sexual Orientation and Gender Identity, 14, 27, 28, (Nov. 21, 2008). Columbia Journal of Gender an Law incorporated in various other U.N. human rights treaties. 65 In 1979, CEDAW Article 16(l)(b) expanded the language of consent to include "[tlhe same right freely to choose a spouse and to enter into marriage only with their free and full consent. 66 The 1995 INS Gender Guidelines formed the basis for the 2001 INS "Gender Guidelines for Overseas Refugee Processing." This later document restates that forced marriage is a gender-based form of harm which may be persecution in refugee law. 67 In addition, the issue of forced marriage is dealt with in detail in training materials for immigration officers produced by the INS: the "Gender-Related Claims Training Workbook" ("Workbook"). The 2002 version of the Workbook mentions forced marriage on a number of occasions as a form of gender-based harm. 68 The Workbook also reiterates that national Gender Guidelines (specifically those of Canada, the United 65 The right was reiterated in Article 23(3) of the International Covenant on Civil and Political Rights (ICCPR) in 1966. International Covenant on Civil and Political Rights, art. 23(3), Dec. 16, 1966, 999 U.N.T.S. 171. The International Covenant on Economic, Social and Cultural Rights (ICESCR) in 1966 used the more limited language of "free consent." International Covenant on Economic, Social and Cultural Rights, art. 10(I), Dec. 16, 1966, 993 U.N.T.S. 3. 66 Convention on the Elimination of All Forms of Discrimination Against Women, art. 16(l)(b), Dec. 18, 1979, 1249 U.N.T.S. 13. 67 IMMIGRATION AND NATURALIZATION SERVICES [INS], GENDER GUIDELINES FOR OVERSEAS REFUGEE PROCESSING 1, 5 (2001), available at http://cgrs.uchastings.edu/ documents/legal/genderguidelines/ USDOSOverseasGender_Guidclines.pdf. 68 IMMIGRATION OFFICER ACADEMY, ASYLUM OFFICE BASIC TRAINING COURSE: FEMALE ASYLUM APPLICANTS AND GENDER-RELATED CLAIMS, PARTICIPANT WORKBOOK 5, 9, 24 (2002) [hereinafter GENDER- RELATED CLAIMS TRAINING WORKBOOK]. Columbia Journal of Gender and Law States, the United Kingdom and Australia),69 U.N. Gender Guidelines and the international human rights instruments mentioned above are relevant evaluative tools in assessing whether harms faced are contrary to international human rights norms. The Workbook also expressly references the 1964 U.N. Convention on Consent to Marriage, which provides that marriage should be entered with the full and free consent of the parties. 70 The 2006 version of the Workbook included for the first time detailed discussion of forced marriage as one of the enumerated examples of harms against women. 71 The 2006 and 2009 versions of the Workbook state, Forced marriage violates numerous human rights. It provides an arena in which sexual abuse, sexual exploitation, domestic violence, forced labor, and slavery often go unnoticed. Women in forced marriages may have fewer educational and work opportunities and their freedom of movement may be restricted. Also, in some cultures, women and girls may be subjected to female genital mutilation prior to the forced marriage. Additionally, a woman's attempt to refuse the forced marriage may result in abusive and/or harmful treatment. 69 IMMIGRATION APPELLATE AUTHORITY (UK), GENDER ASYLUM GUIDELINES §§ 1.13, 2.A.24-25 (2000). These guidelines operated at a tribunal level, and drew heavily upon a model developed in 1998 by the Refugee Women's Legal Group. REFUGEE WOMEN'S LEGAL GROUP, GENDER GUIDELINES .FOR THE DETERMINATION OF ASYLUM CLAIMS IN THE U.K. (1998). See also DEPARTMENT OF IMMIGRATION AND MULTICULTURAL AFFAIRS (AUSTL.), GUIDELINES ON GENDER ISSUES FOR DECISION MAKERS (1996). 70 GENDER-RELATED CLAIMS TRAINING WORKBOOK, supra note 68, at 8, art. 1(1). 71 IMMIGRATION OFFICER ACADEMY, ASYLUM OFFICE BASIC TRAINING COURSE: FEMALE ASYLUM APPLICANTS AND GENDER-RELATED CLAIMS, PARTICIPANT WORKBOOK (2006). The Gao v. Gonzales decision, which had not yet been overturned, is listed as required reading. Id. at I. Columbia Journal of Gender an Law Forced marriages have been asserted, and may under some circumstances qualify, as a form of persecution. . . . The key question in determining whether a forced marriage might constitute persecution is whether the victim experienced or would experience the marriage, or events surrounding the marriage, as serious harm. 72 We endorse this definition of forced marriage and its relationship to the violation of women's human and civil rights. Such an approach is entirely in keeping with developing international understandings of forced marriage as a gendered harm and given that it is being promulgated in the training to every asylum office in the United States, it raises the expectation that American asylum cases would be increasingly in comity with it. However, we found that there was rarely, if ever, any judicial analysis even approaching this level of understanding in the available American cases. To the contrary, we found a widespread and continuing reluctance to accept forced marriage as the basis for asylum in the United States. A. From Kasinga to Gao The celebrated case of Kasinga represents the kernel of much that has happened in American asylum law relating to forced marriage. Fatin (1993) and Kasinga (1996) were two groundbreaking early US cases raising gender issues, and it is notable that both of them remain required reading in the current version of the Workbook. While Fatin concerned a feminist woman from Iran who claimed that she would not comply with 72 Id. at 14-15; IMMIGRATION OFFICER ACADEMY, ASYLUM OFFICE BASIC TRAINING COURSE: FEMALE ASYLUM APPLICANTS AND GENDER-RELATED CLAIMS, PARTICIPANT WORKBOOK 15-16 (2009), available at http:// www.uscis.gov/filcs/ articlc/AOBTC%2OLcsson%2026%2OFcmalc%2OAsylum %20Applications%20and%2OGender-Rclated %20Claims.pdf. Columbia Journal of Gender and Law religious observance and dress requirements, 73 Kasinga involved a claim made by a young woman from Togo, Fauziya Kassindja, that she had been forced to marry at the age of seventeen and would be subjected to female genital mutilation ("FGM") prior to the consummation of the marriage. 74 Forced marriage was a critical aspect of Kassindja's flight from Togo; it formed both an independent claim of harm and an integral aspect of the FGM claim from the start. As part of her BIA case, Kassindja filed a 10-page affidavit. 75 In the first page of this affidavit, she states twice that she did not want to marry but that her aunt forced her. On page two, Kassindja notes under "Family History" that her father did not support coercion in marriage and that all of her four older sisters chose their own husbands. On pages four to six under "Marriage," Kassindja relates on five more occasions that she did not want to marry the man her aunt had dictated and repeatedly told this to her aunt. She notes that her aunt accepted a bride price and arranged a day for the wedding that was kept secret from Kassindja. On the day of the wedding, Kassindja refused to sign the marriage papers. Marriage and FGM are linked throughout her narrative; FGM will happen because it is expected by both the aunt and the husband as a requirement of marriage, but also because as a married woman Kassindja will not be able to disobey or leave her husband and is far less likely to be able to avail herself of state protection. 76 In the "Conclusion" section she states: 13 Fatin v. INS, 12 F.3d 1233 (3d Cir. 1993). In that case the court accepted that women could form a particular social group but held that the applicant had not proven a likelihood of persecution on this basis. Id. at 1240. 14 In re Kasinga, 21 1. & N. Dec. 357 (BIA 1996). Kassindja's name was misspelt by the original immigration officer; thus her case name and actual name do not match. 75 Brief for the Respondent at Exhibit A (Affidavit of Fauziya Kasinga), In re Kasinga, 21 1. & N. Dec. 357 (BIA Dec. 4, 1995), available at http://www.justice.gov/ eoir/efoia/kasinga.htm. 76 1d. at 5. Columbia Journal of Gender an Law Now that I am married, my husband has the right to demand that I return to him and that I be circumcised according to tradition. The rest of the community will not protect me since a husband has a right to say what will happen to his wife. No one will do anything now that I am married .. .As a married woman in Togo, the only legal place for me is with my husband. If I were to try and go somewhere else, the police would come and find me.... I would be forced to go to a husband I did not want and risk my life being circumcised in order to be in a marriage that my Aunt made me enter into against my wishes. 77 Forced marriage in this narrative is an integral aspect of the FGM claim because it necessitated, as well as guaranteed, imminent FGM. It also contributed to a failure of state protection and meant that internal relocation was not possible. In addition, forced marriage was clearly articulated by Kassindja as a distinct and separate harm to FGM: while the forced marriage would result in FGM, it was not the sole harn, nor was it the endpoint of the harm. Yet Kassindja's claim of forced marriage was not addressed at all in the immigration judge's decision (although he did refer to her in the passive object form when he stated that she was "committed to marry before being circumcised." 78). In the BIA decision, the majority notes in the opening section, "The Applicant's Testimony" that "her aunt forced her into a polygamous marriage in October 1994, when she was 17."79 In a section headed "Background Information: The Asylum Application," the decision also notes that a translated copy of the 77 Id. at 10. 78 In re Kasinga at * II, A 73 476 694 (Aug. 15, 1995). 79 In re Kasinga, 21 1. & N. Dec. at 358. Columbia Journal of Gender and Law applicant's marriage certificate, signed by her husband but not by Kassindja herself, was attached to the asylum application. 80 Yet, the issue of forced marriage appears only under a heading of "Ancillary Matters" as an "alternate claim" that was unnecessary for the BIA to address. Thus, in Kasinga, forced marriage disappeared almost entirely from the judicial record as well as from the extensive public discussion and academic commentary on the case, all of which centred exclusively upon FGM. 81 This disappearance presaged much of what was to come in American asylum law as a multitude of gender-based issues, including forced marriage, have been marginalized in place of a major-and we suggest, excessive-focus on FGM. Notably, the BIA does not even index the terms "gender," "women," "domestic violence" or "forced marriage" in its Headnote Charts, although FGM is included. 82 Likewise, on the INS website under "Asylum 80 1d. at 360. 81 See, e.g., Linda Malone, Beyond Bosnia and In re Kasinga: A Feminist Perspective on Recent Developments in Protecting Women from Sexual iolence, 14 B.U. INT'L L.J. 319, 329-37 (1996); Connie Ericson, In re Kasinga: An Expansion of the Grounds for Asylum for Womnen, 20 HOUS. J. INT'L L. 671 (1998); Mary Sheridan, In re Fauziya Kasinga: The United States has Opened its Doors to Victims of Female Genital Mutilation, 71 ST. JOHN'S L. REV. 433 (1997). The case received extensive coverage in the press, commencing with a front page article in the New York Times prior to the BIA determination. See Celia Dugger, Woman's Plea for Asylum Puts Tribal Ritual on Trial, N.Y. TIMES, Apr. 15, 1996; Celia Dugger, U.S. Frees African Fleeing Ritual Mutilation, N.Y. TIMES, Apr. 25, 1996; Op.-Ed., Refugees Fron Mutilation, N.Y. TIMES, May 2, ,1996; Celia Dugger, April 21-27; Seeking Asylum From Genital Mutilation, N.Y.TIMEs, Apr. 28, 1996; Celia Dugger, Board Hears Asylum Appeal in Genital-Mutilation Case, N.Y. TIMES, May 3, 1996; Celia Dugger, U.S. Grants Asylum to Woman Fleeing Genital Mutilation Rite, N.Y. TIMES, June 14, 1996; Celia Dugger, The Asylum System Needs Work, N.Y. TIMES, June 22, 1996. The INS took the extraordinary step of writing a letter to the editor in response. See David Martin, Letter to the Editor, U.S. Backs Asylum for Mutilation Cases, N.Y. TIMES, May 9, 1996. 82 See Index to Precedent Volumes 16-25. Likewise in the BIA Headnote Chart, available at http://www.usdoj.gov/eoir/vll/intdec/ headnote_chart.htm (updated as at August 2011) there is no listing for "gender" or any gender related term under the any of the Asylum categories, although FGM has its own topic under Persecution. Columbia Journal of Gender an Law Resources," there is no document addressing gender-based claims under the "Alert Series" and "Question and Answer Series," although there has been a specific document on FGM since 1994.83 Connie Oxford, in her sociological study of gender- based asylum claims in the United States found the erasure of complex and multiple dimensions of women's experiences of persecution extended beyond the realm of the formal legal judgments we discuss here. Oxford undertook fieldwork in the early 2000s, comprising observations and interviews with a range of service providers and agents in the asylum process, such as doctors, psychologists and lawyers. She found that a broad range of agents in the asylum process actively encouraged applicants to pursue FGM grounds of claim and subordinated, ignored or failed to inquire about other forms of gendered harm such as forced marriage and domestic violence. 84 The other significant aspect of the Kasinga decision is that it reveals an attempt by the INS to frame issues concerning gender in sharp contradistinction to the then-recently released INS Gender Guidelines. The INS attempted to put forward a broad "framework of analysis" for FGM claims in which it conceded that the risk of involuntary FGM in the future could be a form of persecution, but attempted to exclude those who had experienced FGM in the past from eligibility for asylum. While the majority of the BIA ignored these arguments, focusing instead on the elements of the case at hand, the INS "framework" argument was briefly touched upon by the concurring opinions. Four aspects of the INS claims which emerge from these fleeting references merit discussion. First, the INS arguments were informed by a combination of "floodgates fear" and cultural relativism that we 83 U.S. Citizenship and Immigration Services, Asylum Resources, w ww. u s c i s. go v/ port al/s i t c/u s c i s/ menu item. c b I d 4 c 2 a 3 e 5b 9 ac 89243 c 6 a 7543 f6 d I a/ ? vgnextoid=d2d I e89390b5d010VgnVCM 10000048f3d6a I RCRD&vgnextchanne I=d2dl e89390b5d010VgnVCM 10000048f3d6a I RCRD/ (last visited Nov. 21, 2011); INS, ALERT SERIES: WOMEN: FEMALE GENITAL MUTILATION (1994), available at www.uscis.gov/files/nativedocuments/alnga94-001 (fgm).pdf. 84 Connie Oxford, Protectors and Victims in the Gender Regime of Asylum, NWSA J., Fall 2005, at 18, 29-30. Columbia Journal of Gender and Law see repeated again and again in later gender-based claims. Speaking of FGM in particular, the BIA notes that, "The [INS] points out that it is 'estimated that over eighty million females have been subjected to FGM.' It further notes that there is 'no indication' that 'Congress considered application of [the asylum laws] to broad cultural practices of the type involved here. " 85 The INS Gender Guidelines note on a number of occasions in discussing both the issue of persecution and relevant Convention grounds, that women may face harm on account of breaching gender-related social mores in their country of origin. They quote also with approval the then-current UNHCR conclusions that women "who face harsh or inhumane treatment due to their having transgressed the social mores of the society in which they live may be considered a particular social group." 86 It is clear, therefore, that the INS's own Gender Guidelines considered that 'broad cultural practices' directed towards the oppression of women were highly relevant to, indeed paradigmatic examples of, gendered harm analysis.87 Second, as part of an overt policy argument that the BIA should use the case as precedent to restrict rather than enlarge eligibility, the INS contended that human rights norms were not relevant to the analysis of persecution of women: The Service further argues that "the Board's interpretation in this case must assure protection for those most at risk of the harms covered by the statute, but it cannot simply grant asylum to all who might be subjected to a practice deemed 15 In re Kasinga, 21 1. & N. Dec. at 370 (Filppu, J., concurring) (citations omitted, emphasis added). 86 See INS Gender Guidelines, supra note 59, at 3, 4,114. 87 See also GENDER-RELATED CLAIMS TRAINING WORKBOOK, supra note 68, at 21 ("The fact that a practice is widespread, (eg: domestic violence, FGM, rape as part of an occupation during war) is not relevant to determining whether the alleged acts constitute persecution."). Columbia Journal of Gender an Law objectionable or a violation of a person s human rights."88 This approach directly contradicts the INS Gender Guidelines instruction that "[t]he evaluation of gender-based claims must be viewed within the framework provided by existing international human rights instruments and the interpretation of those instruments by international organizations." 89 Third, in the words of the BIA, the INS argued that the test for persecution should 'exclude past victims of FGM from asylum eligibility if "they consented" to it or "at least acquiesced", as in the case of a woman who experienced FGM as "a small child." ' 90 (Binary distinctions between past and future FGM plagued American cases until very recently settled by an order of the Attorney General.)91 For the purposes of our discussion it is extremely troubling that the INS should characterize small girls who lack the ability to meaningfully consent or power to resist as "acquiescing" to a practice that the INS itself concedes is a human rights abuse. In the context of forced marriage claims this would mean that children previously subject to marriage before the legal age of consent could likewise be characterized as "acquiescing" to it (rather than as a priori forced to marry because they lacked the ability to consent). Finally, American law conclusively bars those who themselves have been persecutors from claiming asylum. The 88 In re Kasinga, 21 1. & N. Dec. at 371 (citations omitted, cmphasis addcd). 89 See INS Gcndcr Guidelincs, supra note 59, at 2. 9°In re Kasinga, 21 1. & N. Dec. at 371. 91 Mattcr of A-T, 24 I. & N. Dcc. 617 (Attorney Gen., 2008), The case overruled In re A-T, in which the applicant had undergone FGM in thc past but feared future marriage-the BIA ignored the marriage and had found that persecution was past only as FGM could not be repeated. 24 I. & N. Dec. 296 (BIA 2007). Columbia Journal of Gender and Law INS argued in Kasinga that it would be "[Alnomalous if persons facing death in their homelands because of religious or political persecution were denied protection ... simply by virtue of being parents of FGM victims and having followed tribal custom."9 2 In making this argument the INS implicitly posited a hierarchy of refugee protection in which the real grounds of claim (religious or political) and real forms of persecution (death) were being mistakenly transplanted by considering a "custom" or "broad cultural practice" enforced by non-state actors to be persecution. In this discursive twist we must worry about the consequences of legal developments in gender-related asylum law for family members who are persecutors, as it is they who are actually the rightful victims. At the time Kasinga was decided, the INS Gender Guidelines had been publicly available for one year, were required reading for all interviewing and supervising asylum officers and had been included in training materials. In one of the two concurring opinions in Kasinga, Judge Rosenberg noted with considerable understatement the "curious" fact that the INS made no reference to its own published gender guidance in its arguments. It is even more disturbing that, eleven years after Kasinga was decided and twelve years after the promulgation of the Gender Guidelines, government lawyers were still making many of the arguments outlined above in their Supreme Court petition for certiorari to overturn the strongest judicial statement on asylum and forced marriage to date: the Second Circuit Court of Appeals decision in Gao. In Gao, a young woman from China, Hong Yin Gao, had been promised in marriage in exchange for a bride price. The immigration judge in 2003 characterized this as a "family dispute" (because Gao's mother "violated the oral contract" with the groom) and held that there was no particular social group.93 This finding was summarily affirmed by the BIA in 2004.94 In 92 In re Kasinga, 21 1. & N. Dec. at 373, n.2. 93 Petition for a Writ of Certiorari at *25, Kcisler v. Gao, 552 U.S. 801 (2007) (No. 06-1264), 2007 WL 835007. (2d Cir. March 16, 2007). Columbia Journal of Gender an Law 2006 the Court of Appeals for the Second Circuit held that the relevant social group was "women who have been sold into marriage (whether or not that marriage has yet taken place) and who live in a part of China where forced marriages are considered valid and enforceable. '95 The court also held that "lifelong involuntary marriage" was a form of persecution. 96 In 2007 the Attorney General's petition to the Supreme Court commenced by characterising the Second Circuit decision as, [E]stablishing a novel and potentially sweeping interpretation of the INA that could have far-reaching implications for the Executive Branch's enforcement of immigration law in the highly sensitive context of culturally diverse approaches to marriage. 97 The Attorney General's petition went on to reiterate various permutations of floodgates and cultural relativism arguments, for example that "60% of all marriages worldwide and 96% of marriages in India, are arranged on terms that are often similar" to those in Gao,98 and that they reflect "broad 95 Gao v. Gonzales, 440 F.3d at 70. 96 Id. 97 Petition for a Writ of Certiorari at *3, Keisler, 552 U.S. 801(No. 06-1264). For repeated references to the "sweeping" and "novel" approach of the court, see generally Reply Brief on Petition for a Writ of Certiorari, Keisler, 552 U.S. 801 (No. 01-1264). 11 Petition for a Writ of Certiorari, Keisler, 552 U.S. 801 at *3, *19 (No. 06-1264). Columbia Journal of Gender and Law cultural and religious acceptance" in the countries of origin. 99 Like the eighty million women potentially subject to FGM, the sixty percent of women in arranged marriages evoke a veritable tidal wave of claimants, which must be held at bay by stringent immigration control.100 Yet on-shore claims by women have always represented a minority of asylum claims in the United States' 0' (as elsewhere) and many of these claims will not, of course, involve gender-related persecution. 102 As in Kasinga, the government's position on "consent" is extremely problematic. In Gao, the fact that arranged marriage involving the payment of a bride price was a common 99 Id. at *21 (emphasis added). See also the references to "consistent with cultural tradition" at *17, sensitive "cultural questions" of "marriage traditions and practices ... worldwide" at * 19, "long-standing tradition in many cultures and countries" at *20, and the "deep roots of such practices in the cultures and religions of a number of foreign nations" at *22. This may be contrasted with the position stated on the website of the American Embassy in Dhaka, Bangladesh, which reads: "Forced marriages are not the same as arranged marriages. Arranged marriages arc a part of many cultural traditions and involve the free and full consent of both parties. Some people, however, find themselves compelled to marry against their will, either in the United States or overseas. This is called a forced marriage, and it is a human rights concern, as human rights principles seek to advance the freedom and inherent dignity of each individual." Embassy of the United States: Dhaka, Bangladesh, supra note 43. 10o Connie Oxford finds in her study that domestic violence asylum claims frequently invoked floodgates discourse from Immigration judges, despite their numerical infrequency. Oxford, supra note 84, at 23. 101 'Connie Oxford cites unpublished INS figures on affirmative claims by sex for the years 1998-2002 in which women make up approximately one third of claims. Connie Oxford, Gender-based Persecution in Asylum Law and Policy in the United States 20 (2006) (unpublished Ph.D. dissertation, Univ. of Pittsburgh), (on file with authors). 102 Id. at 46-48 (regarding the difficulty in the American context of obtaining precise data regarding the number of gender-related persecution claims). Although this was expressly acknowledged in a recent brief by the Department of Homeland Security in a refugee case concerning domestic violence. See Supplemental Brief of Department of Homeland Security, supra note 55, at 13-14 n. 10 (demonstrating that the Department still defined the gendered group as narrowly as possible). Columbia Journal of Gender an Law practice to which the applicant did not object in principle (rather, she did not want to marry the chosen groom in particular) led the Attorney General to characterize the case as a broken engagement, involving merely a "private dispute" between two families 10 3 and, breathtakingly, as a contract dispute. 104 In other contexts, the United States government would characterize the payment of money to others for possession of women as slavery, sexual slavery and/or trafficking-all of which are both domestic and international crimes. 105 Lastly, the Attorney General's position in Gao continues to re-order persecutors as victims, restating the argument from Kasinga that a finding of persecution based on this "cultural practice" would exclude those who participated in it from obtaining asylum under American law, "thereby 103 Petition for a Writ of Certiorari, Kcisler v. Gao, 552 U.S. 801 at 21 (2007) (No. 06-1264). 1I4 Id. at *15, *17, *22. 105 See, e.g., Protocol to Prevent, Suppress and Punish Trafficking in Persons, especially Women and Children, supplementing the United Nations Convention against Transnational Organized Crime, G.A. Res. 25, annex II, U.N. GAOR, 55th Sess., 62d plen. mtg., U.N. Doc. A/45/49 (2001), (ratified by the United States on Nov. 3, 2005); Supplementary Convention on the Abolition of Slavery, the Slave Trade and Institutions and Practices Similar to Slavery, Apr. 30, 1957, 226 U.N.T.S. 3 (ratified by the United States on Dec. 6, 1967). The United States has also passed numerous acts to prevent trafficking both domestically and internationally. See, e.g., Victims of Trafficking and Violence Protection Act of 2000, Pub. L. No. 106-386, 114 Stat. 1464 (codified as amended in scattered sections of 18, 22, 27, 42 U.S.C.), the Trafficking Victims Protection Reauthorization Act, 18 U.S.C. 1595, 22 U.S.C. 7109(a) (2003), the Trafficking Victims Protection Reauthorization Act of 2005, Pub. L. No. 109-164, 119 Stat. 3558, (codified as amended in scattered sections of 18, 22, 42 U.S.C.), and the Trafficking Victims Protection Reauthorization Act of 2008, Pub. L. No. 110-457, 122 Stat 5044 (codified as amended in scattered sections of 6, 8, 18, 22, 28, 42 U.S.C.). They have also established the President's Interagency Task Force to Monitor and Combat Trafficking in Persons. U.S. Department of State, The President's Interagency Task Force To Monitor and Combat Trafficking in Persons, Senior Policy Operating Group, http:// www.state.gov/g/tip/rls/fs/2009/120224.htm (last visited Apr. 2, 2010). Within the Department of State there is the Office to Monitor and Combat Trafficking in Persons. U.S. Department of State, Office to Monitor and Combat Trafficking in Persons, http://www.state.gov/g/tip/ (last visited Apr. 2, 2010). Columbia Journal of Gender and Law potentially barring thousands of persons-parents, relatives, and matchmakers . . . from obtaining asylum, regardless of the severity of persecution they might face."' 0 6 Because the Supreme Court reversed and remanded Gao on the narrow basis that the Appeals Court ought not have reformulated the protected social group itself (but rather remitted to the BIA to do so), none of these arguments were ultimately addressed. What is troubling in both Kasinga and Gao is their revelation of the commitment at such high levels of the immigration executive in the United States to a long-term strategy of undermining, even openly violating, their own gender guidance. iii. THE REFUGEE CASES In total we identified forty-eight American cases where forced marriage was articulated as part of a claim to asylum or withholding of deportation covering the period 1994 to 2008 (inclusive). 107 In our analysis we compare the American cases with findings from a previous study of all available administrative tribunal and court refugee determinations from the United Kingdom, Canada and Australia on forced marriage 106 Reply Brief on Petition for a Writ of Certiorari at *8, Keisler, 552 U.S. 801, at *8 (2007) (No. 01-1264); see also Petition for a Writ of Certiorari at *21, Kcisler v. Gao, 552 U.S. 801 (2007) (No. 06-1264). 107 As with the broader study, the search terms used were "forced marriage," "forced to marry," and "pressure to marry." In the United States, the databases used were LEXIS and Wcstlaw, with searches also made of the BIA Precedent Decisions and the Hastings Center for Gender and Refugee Studies collections. A small number of 2009 cases are referred to in discussion but not included in statistical analysis as we were not able to gather cases for the entire year and the international data ended at December 2008. In the international data set, Australian cases were all obtained from the Austlii case database (www.austlii.edu.au). United Kingdom cases were obtained from the Electronic Immigration Network case database (www.ein.org.uk), the Asylum and Immigration Tribunal website (www.ait.gov.uk), LEXIS, Bailii (http:// www.bailii.org) and U.N. RcfWorld databases. Canadian cases were obtained from the QuickLaw, Canlii (www.canlii.org) and LEXIS databases. Columbia Journal of Gender an Law ("the international data set"). 108 Success rates of claims, while drawn from only a partial case set and based upon small numbers of claims, provide a rough benchmark from which to start. 1 9 The overall positive rate in the United States decisions was thirty-one percent, almost identical to our findings of a thirty-two percent positive rate from the international data set. 110 In addition to unique features of the American system noted earlier, there are several other factors, which warrant caution in drawing direct comparisons between the United States and other countries. Like the international data set, the claims made in the United States are diverse and arise from a wide range of different countries of origin (fourteen in total). Only four countries-Mali, Nigeria, India and China-gave rise to more than one claim in the United States. I"' However, unlike the international data set, the United States evinced a massive concentration of cases arising from just one country. Thirty of the American cases (or 63%) arose from China, a country of origin, which did not feature heavily in the decisions of any of 108 The international data set comprised 120 decisions in total, made up of sixty-nine decisions from Australia, forty from Canada and a mere eleven from the United Kingdom. 109 We count "positive" or "negative" decisions from the perspective of the applicant, even if(as in the case of judicial review) the decision is one of remittal and reconsideration of the claim rather than an ultimate positive determination of refugee status. In general this gives an inflated sense of "positive" outcomes, as we do not have access to the majority of the remittal determinations and some, perhaps many, of these will ultimately be negative to the applicant. When the cases are disproportionately made up of judicial review decisions, as in the United States, the positive figure is likely to be even less representative of substantive results. 10 These figures mask significant divergence across the receiving nations, with the positive rate forty-three percent in Canada and twenty-six percent in Australia. In the United Kingdom, of only eleven decisions, three were positive but two of these were in fact remittals. I The other countries of origin are Kosovo, Cameroon, Guinea, Iran, Ivory Coast, Kenya, Pakistan, Philippines, Togo and Zambia. Some countries had more than one available decision but these arose from the same claimant at different levels of the appellate system. Columbia Journal of Gender and Law the other receiving countries in the international data set.' 12 This reflects the high numbers of on-shore claims from China in the United States generally as well as the high proportion of claims in which forced sterilization and abortion were raised by virtue of their specific statutory inclusion in the United States refugee definition. We found that claims of forced marriage appeared alongside those of coercive reproductive policies more broadly in a number of cases from China, and that these claims were more likely to demonstrate changed grounds through the process and to be dismissed on the basis of negative credibility as a result. It also appears that the high profile case of Gao (until overturned a year later) led to a spike of similar claims. 113 It should also be noted that the American case set was very heavily dominated by Court of Appeals decisions,"14 with only two BIA decisions and six immigration judge decisions available. This gives an artificially high sense of success rates for forced marriage claims in the United States because positive court decisions do not result in a grant of asylum but rather to a remittal of the case back to the BIA, which may then again refuse asylum. 115 Moreover, some of the positive court decisions were made on a basis other than the forced marriage claim. 116 In 112 In fact, there were only four claims from China in the international data set, one of which was successful. The top five countries of origin in the international data set were Bangladesh, Nigeria, India, Iran and Ghana. 113 More than half of the claims from China arc post-Gao and several feature strong factual similarities. 114 Thirty-nine of the forty-eight decisions (or eighty-one percent of the decision pool) were appellate court judgments. In the international data set the proportion of appellate court decisions was only thirty-eight percent. 115 Of a total of fifteen positive decisions, eleven were from the Court of Appeals, meaning that only four of the positive decisions definitely led to a grant of asylum or withholding of removal. 116 Of the eleven positive decisions at court level, three were on another basis (such as failure to consider the consequences on return of the applicant's illegal departure or changed circumstances in the country of origin). Columbia Journal of Gender an Law addition, the high proportion of appellate decisions meant that the BIA's approach to these issues was often not made clear.117 Elsewhere we have focused on lower-level administrative tribunals for the very reason that this is where the vast bulk of decision-making occurs in refugee law. Another significant difference with the American cases was that there were none in which a forced marriage claim was brought by a lesbian or gay man. Rather, all forty-eight claims concerned people who were, or were presumed to be, heterosexual: forty-five claims were brought by women, one claim was brought jointly by a woman with her male partner and two were brought solely by men (both of which failed). This stands in striking contrast to the international data set, where forty percent of the claims concerned gay or lesbian applicants. 118 In the international data set, forced marriage claims by gay men and lesbians were important in raising the intersection of gender and sexuality norms, although these connections were not always (or even often) received and analysed in a particularly sophisticated manner. However, in the United States, the complete absence of a sexual orientation dimension in the marriage cases meant that "gender" was generally seen by decision-makers as concerning only women. 119 117 It is also possible that many claims of forced marriage at early levels are simply "lost" from the record if they were unsuccessful and not reiterated at higher levels. 118 The international data set comprised: fifty-eight percent heterosexual women, thirty-two percent gay men, eight percent lesbians, and two percent heterosexual men. This led to significant differences in the way that claims were framed and received. Marriage itself was usually the central feature of heterosexual women's claims, whereas it was often a more minor or cumulative part of a claim brought by lesbians and gay men. The lesbian cases were roughly divided, with slightly more than half of them featuring actual forced marriage or a specific threat such that forced marriage was central to the claim in a manner akin to the heterosexual women's cases, while the other half were more similar to the gay men's claims in that homophobically motivated persecution was the core element of a claim in which marriage was a general threat or more tangential aspect. '9 See also Oxford, supra note 101, at 147-51. Columbia Journal of Gender and Law A. Particular Social Group While claims of forced marriage, like other gender- related claims, could be brought on the basis of the religious or political opinion (or imputed political opinion) grounds, overwhelmingly they are framed on the particular social group ground. 120 We found that the definition of particular social group was a major stumbling block in the American cases. It was clear from the often scant reasons in at least eighteen cases- representing nearly forty percent of the available American case pool-that the Immigration Judge had held at first instance that there was no relevant Convention ground for women fleeing forced marriage. t 21 Alarmingly, in a number of cases decision- makers appear to have summarily drawn this conclusion without any written analysis or formulation of the various possible 120 The United States, like the United Kingdom and Australia, generally rejected gender-based claims as related to either the religious or political opinion grounds when such claims were occasionally made. In contrast, Canadian decision-makers frequently characterized forced marriage claims as engaging the religious or political ground under the Convention in addition to particular social group. 121 Petition for a Writ of Certiorari, Keisler v. Gao, 552 U.S. 801, app. C at *25 (2007) (No. 06-1264); Gao v. Gonzales, 440 F.3d 62, vacated sub nom. Keisler v. Gao, 552 U.S. 801 (2007); Ngengwe v. Mukascy, 543 F.3d 1029 (8th Cir. 2008) (remanded on this basis); In re A-T, 24 I. & N. Dec. 296 (BIA 2007); Berishaj v. Gonzalez, 238 F.App'x 57 (6th Cir. 2007); Xiu Yun Chen v. Gonzalez, 229 F.App'x 413 (7th Cir. 2007); Yan Dan Li v. Gonzalez, 222 F.App'x 318 (4th Cir. 2007); Hua Lin v. Gonzalez, 205 F.App'x 879 (2d Cir. 2006); Yi Meng Tang v. Gonzalez, 200 F. App'x 68 (2d Cir. 2006) (remanded on this basis); Chun Hua Weng v. Gonzalez, 185 F.App'x 77 (2d Cir. 2006); Himanje v. Gonzalez, 184 F.App'x 105 (2d Cir. 2006) (remanded on this basis); Lan Zhu Pan v. Gonzalez, 445 F.3d 60 (Ist Cir. 2006); Keita v. Gonzalez, 175 F. App'x 711 (6th Cir. 2006); Xue Qin Li v. B.I.A., 172 F.App'x 385 (2d Cir. 2006); Li Qun Chen v. Gonzalez, 153 F.App'x 49 (2d Cir. 2005); Matter of S L (N.Y.C., NY Immigration Court, Oct. 7, 1999, copy on file with authors); Jin Chao Zheng v. Gonzales, 236 F.App'x 726 (2d Cir. 2007); Xiao Feng Lin v. Attorney General, 249 F.App'x 281 (3d Cir. 2007); Xiu Xia Huang v. Attorney Gen., 286 F.App'x 604 (11th Cir. 2008). In Lan Chen v. Gonzalez, it was unclear whether forced marriage was articulated at first instance as part of political opinion claim or whether that was entirely separate. 187 F.App'x 43 (2d Cir. 2006). In Xiao Feng Lin v. Attorney Gen., the BIA found social group to be a problem, overruling an immigration judge finding that the claim was frivolous. 249 F.App'x 281 (3d Cir. 2007). Columbia Journal of Gender an Law particular social groups. 122 At the level of judicial review, courts generally did not engage with the original immigration judge or BIA failure to define a social group if there was any other basis upon which to uphold the original decision. 123 The effect of the Supreme Court decision in Gao is that a complete failure to define the particular social group or a clear error in defining it will, at most, lead to the case being remitted to the BIA, as the Court is not permitted to formulate the appropriate group. This lack of judicial guidance on social group formulation is really regrettable, most especially because the reasoning on particular social group in the available American cases was dramatically worse than the other countries examined in this study. Canada has accepted gender-based grounds for refugee claims, including forced marriage, from the mid-1990s and not a single Canadian claim by a female applicant in our study was rejected on the basis of a lack of social group. In the Canadian cases, the group was framed variously as "women,"' 124 "women who refuse to follow traditional practices"'1 25 and "women regarded as chattels."'1 26 The issue of particular social group was more contentious in Australia, although this diminished 122 See, e.g., Lizhu Chen v. BIA, 238 Fed.App'x 669 (2d Cir. 2007); Keita v. Gonzalez, 2006 US App. LEXIS 9484 (6th Cir. Apr. 13, 2006); Li Qun Chen v. Gonzalez, 153 Fed.App'x. 49 (2d Cir. 2005). 123 See, e.g., Berishaj v. Gonzalez, 238 Fed.App'x 57 (6th Cir. 2007); Xiu Yun Chen v. Gonzalez, 229 Fed.App'x 413 (7th Cir. 2007); Yan Dan Li v. Gonzalez, 222 Fed.App'x. 318 (4th Cir. 2007); Lan Zhu Pan v. Gonzalez, 445 F. 3d 60 (1st Cir. 2006); Xiu Xia Huang v. U.S. Att'y Gen., 286 Fed.App'x 604 (I Ith Cir. 2008). This was so even when, arguably, the findings on matters such as likelihood of persecution and the question of the nexus of persecution to the particular social group rested upon and therefore required first a finding of what the social group actually was. See, e.g., Lan Zhu Pan v. Gonzalez, 445 F.3d 60 (1 st Cir. 2006); Ying Lin v. U.S. Att'y Gen., 319 Fcd.App'x 777 (11 th Cir. 2009). 124 X. v. Canada, 2001 CanLIl 26862 (Immigration & Refugee Bd.). 125 Re X., 2002 CanLll 52705 (Immigration & Refugee Bd.) at *3. 126 This was the tribunal's own formulation. Re X., 2000 CanLIl 21420 (Immigration & Refugee Bd.) at *3. Columbia Journal of Gender and Law following the High Court gender and domestic violence decision Khawar in 2000.127 Although in the United Kingdom the House of Lords addressed gender and particular social group in 1999 in Shah and 1slam, 128 early level decision makers in the United Kingdom continued to hold at first instance that there was no applicable social group for women fleeing forced marriage through the early to mid-2000s. Moreover, the Home Office pursued this argument through the appellate process. 129 Yet, even in comparison to the United Kingdom, on the issue of particular social groups the United States was and remains the most stagnant, least coherent and most out of step with international developments. The early BIA approach of defining a particular social group as a group which is bound together by common characteristics which are either innate or so fundamental that they ought not be changed, 130 as later refined by the Supreme Court of Canada in Ward, 131 is now one that is widely accepted 127 Min. for Immigration v. Khawar (2002) 210 C.L.R I (Austi.). In our study, heterosexual women claiming forced marriage had a positive rate of only eleven percent in Australia prior to Khawar, compared to a thirty-eight percent positive rate subsequently. 128 Islam v. Sec'y of State for the Home Dep't, [1999] 2 A.C. 629 (U.K.). 129 This argument has been pursued to the extent of appealing positive decisions by adjudicators. See RG (Eth.) v. Scc'y of State for the Home Dep't, (2006) EWCA (Civ), 339 (Apr. 4, 2006) (Eng.). 30 "The shared characteristic might be an innate one such as sex, color, or kinship ties, or in some circumstances it might be a shared past experience such as former military leadership or land ownership ... Whatever the common characteristic that defines the group, it must be one that the members of the group either cannot change, or should not be required to change because it is fundamental to their individual identities or conscience." In re Acosta, 19 1. & N. Dec. 211, 233 (BIA 1985). 131 AttomeyGen. v. Ward, [1993] 2 S.C.R. 689 (Can.). Columbia Journal of Gender an Law internationally as well as in the United States. 132 However, the BIA and various federal courts of appeal have added their own "glosses" or additional elements to the widely accepted "innate or fundamental characteristics" approach. These include the additional requirements of "cohesion" or "voluntary association" among the group by the large and influential Ninth Circuit (an approach emphatically rejected by all of the other countries in the international data set and by UNHCR' 33) and, more commonly, requiring external "social visibility" of the group. 134 American decision makers have consistently rejected broad formulations of social group such as "women" and "young women from rural China"'135 for the above reasons. In a 2005 immigration judge decision, the claim of membership of the much narrower "social group of Guinean Fulani women who oppose forced, arranged marriages" was also rejected on the basis that: 132 For an overview of international approaches, see T. Alexander Aleinikoff, Protected Characteristics and Social Perceptions: An Analysis of the Meaning of "Membership of a Particular Social Group", in REFUGEE PROTECTION IN INTERNATIONAL LAW: UNHCR's GLOBAL CONSULTATIONS ON INTERNATIONAL PROTECTION (Erika Feller, Volker Tfirk & Frances Nicholson eds., 2003). 133 U.N. High Comm'r for Refugees, Guidelines on International Protection No. 2: "Membership of a Particular Social Group" within the context ofArticle IA(2) of the 1951 Convention and/or its 1967 Protocol Relating to the Status of Refugees, 15, U.N. Doc. HCR/GIP/02/02 (May 7, 2002) [hereinafter UNHCR, Guidelines on Particular Social Group]. The Ninth Circuit was the only circuit to require a "voluntary associational relationship." Sanchez-Trujillo v. INS, 801 F.2d 1571, 1576 (9th Cir. 1986). In Hernandez-Montiel v. INS, the Ninth Circuit retreated from this position and held that a particular social group is one united by an innate characteristic or by a voluntary association. 225 F.3d 1084 (9th Cir. 2000). 134 See Aleinikoff, supra note 132; Marouf, supra note 50. More recently the BIA has formulated this as a question of "particularity" requiring recognition by society as "a discrete class of persons." Matter of S-E-G, 24 1. & N. Dec. 579, 594 (BIA 2008). Lan Zhu Pan v. Alberto Gonzalez, 445 F.3d 60 (1st Cir. 2006). Columbia Journal of Gender and Law [t]he respondent has presented no evidence indicating that women who oppose forced marriage are a cognizable social group within Guinean Fulani society . . . The respondent did not enter into any voluntary associations based on her opposition to forced marriage, nor did she demonstrate that her abuser viewed her as a member of any such group. 36 Significantly, this analysis addressed only the two additional "glosses" and not the core test of whether such women possessed an innate or fundamental characteristic. Similar formulations such as "young Bambara women who oppose arranged marriage 137 were commonly rejected on the basis that the group was not socially visible.' 38 Both the continued adhesion to a separate requirement of social visibility 136 Matter ofA-D-A (Bos., MA Immigration Court, Sept. 19, 2005) at * 13 (on file with The Hastings Center for Gender & Refugee Studies). 137 In re A-T, 24 1. & N. Dec. 296, 303 (BIA 2007), vacated on other grounds byIn re A-T, 24 I. & N. Dec. 617 (Attorney Gen. 2008). 138 See Matter of S L (N.Y.C., NY Immigration Court, Oct. 7, 1999) (rejecting the application on the basis that there was not a cohesive group with voluntary association) (on file with The Hastings Center for Gender & Refugee Studies). Columbia Journal of Gender an Law or perception 139 and the interpretation of this requirement are at odds with UNHCR Guidelines. In its 2003 Guidelines on Particular Social Group, UNHCR notes both the "innate or fundamental characteristic" approaches and the "social perception" approaches to group analysis, and formulates them as alternatives to each other rather than as additional requirements in framing a social group. 140 Moreover, UNHCR 139 Note that the Department of Homeland Security reaffirmed these requirements in its recent brief supporting domestic violence as the basis of a refugee claim. See Supplemental Brief of Department of Homeland Security, supra note 55. The Department's proposed formulations of the particular social group in that brief ("Mexican women in domestic relationships who are unable to leave" or "Mexican women who are viewed as property by virtue of their position within a domestic relationship") are as circular as the proposed particular social group rejected by the Department ("Mexican women in an abusive relationship who are unable to leave") in terms of the role of persecution in defining the group, discussed below, and moreover requires a similar degree of contortion to characterize them as socially visible, distinct and cognizable groups. Id. at 10-15 (discussing the Department's view of the different proposed particular social groups). Surely the reason that women are viewed as property within domestic relationships or are unable to leave relationships is because women generally are disempowered in both social and legal terms in the society in question: thus the appropriate particular social group should simply be "women in Mexico." "I UNHCR, Guidelines on Particular Social Group, supra note 133, at 10-12; UNHCR, Gender Guidelines, supra note 64, at 29. It is ironic that the social perception approach reached its zenith in Applicant A v. Australia, rejecting Chinese facing forced sterilization as a particular social group, while the United States entrenched the social visibility approach at the same time that it prioritised this particular experience of persecution above others through defining it as a basis for asylum in statute. (1997) 190 C.L.R. 225; see also supra note 53. Columbia Journal of Gender and Law has repeatedly stated that the broad social group of "women in X country" should satisfy both bases. 141 In common with international standards, the United States requires that a particular social group cannot be solely defined by reference to the persecution. This offers additional challenges when the group is narrowly defined. So, for example, a group formulated as "women in Iran who are forced by their fathers to marry" is unacceptable because the defining characteristic of the group is the persecution they face. Yet, persecution may still be considered as a relevant factor in the group definition if it is not the exclusive factor in defining the group. 42 It was clear in our study that Canada took a less strict approach to this issue than the United Kingdom or Australia, while the issue was particularly difficult in the United States because of the widespread rejection of broader formulations of particular social groups on the basis that they were not socially visible or not likely to be singled out for persecution. This meant that applicants and their advisors in the United States struggled for narrower formulations, which were then in danger of being 141 UNHCR, Guidelines on Particular Social Group, supra note 133, 7, 12, 18; UNHCR, Gender Guidelines, supra note 64, 30. As noted by the Third Circuit in 1993 in Fatin, "The phrase 'particular social group' was first placed in the INA when Congress enacted the Refugee Act of 1980. Pub.L. No. 96-212, 94 Stat. 102 (1980). While the legislative history of this act does not reveal what, if any, specific meaning the members of Congress attached to the phrase 'particular social group,' the legislative history does make clear that Congress intended 'to bring United States refugee law into conformance with the 1967 United Nations Protocol Relating to the Status of Refugees, 19 U.S.T. 6223, T.I.A.S. No. 6577, to which the United States acceded in 1968."' Fatin v. INS, 12 F.3d 1233, 1239 (3rd Cir. 1993). Contemporary international guidance is relevant to whether in fact such conformity is being achieved in accordance with the original legislative intent. 142 UNHCR Guidelines on Particular Social Group note that if the social visibility approach is used, as it is in the United States, "persecutory actions towards a group may be a relevant factor in determining the visibility of a group in a particular society." UNHCR, Guidelines on Particular Social Group, supra note 133, 14. Columbia Journal of Gender an Law rejected on the basis that the group was solely defined by the persecution. 143 The case of Elizabeth Ngengwe illustrates this dangerous balancing act between narrow and broad formulations of the social group. Ngengwe claimed that she was subject to persecution as a widow by her husband's family following his death. The family had demanded that she marry one of her deceased husband's brothers (levirate marriage) or repay a bride price that her family had received on her original marriage. In 2003, before the immigration judge, Ngengwe offered both broader ("Cameroonian widows" or "widowed females who are forced into marriage because of tradition or cultural values in Cameroon") and narrower formulations of the group ("widowed females who are falsely accused of killing their husbands because they are not from the same tribe."' 44) Despite the fact that there was a State Department Country Report in evidence which indicated that as a matter of customary law, widowed women in Cameroon were required by force to marry one of the deceased's brothers, 45 the government contended before the Immigration Judge that widowed women facing forced marriage was "too broad a category to be cognfizable as a particular social 143 So, for example, the proposed particular social group "young women threatened with imprisonment for failing to oblige the demands of a government official to marry his relation" was rejected on this basis. Xiao Feng Lin v. Attorney General, 249 F.App'x 281 (3rd Cir. 2007). 114 See Matter of E S N (Kan. City, MO Immigration Court, Jan. 14, 2003) at *8 (on file with The Hastings Center for Gender & Refugee Studies). On appeal the Eighth Circuit expressed these somewhat differently, as "Cameroonian widows" and "widowed Cameroonian female member[s] of the Bamileke tribe, in the Southern region that [belong] to a family or [have] in-laws from a different tribe and region, the Bikom tribe in the Northwest province, who have falsely accused [them] of causing [their husbands'] death." Ngcngwe v. Mukasey, 543 F.3d 1029, 1033 (8th Cir. 2008). 115 Matter of E S N (Kan. City, MO Immigration Court, Jan. 14, 2003) at *15-*16 (on file with The Hastings Center for Gender & Refugee Studies). Columbia Journal of Gender and Law group" under the Act. 146 The government also contended that the characteristics of this group were not innate or immutable, as the applicant had "the power to change" by either marrying or paying back the bride price. The immigration judge accepted all of these arguments. 147 In addition, the immigration judge found that the broadest formation of "widows" was not sufficiently homogenous to be cognizable as a group and rejected the narrower formulation of "widows facing forced marriage" because it defined the group by reference to the persecution. The immigration judge also rejected the narrowest group on the basis that this amounted to "simply a widowed female, who is disliked by her in-laws" and was therefore merely personal. 48 The BIA affirmed this on review. However, on appeal, the Eighth Circuit held that it was an error to reject the broadest formulation of "Cameroonian widows" because the United States government's own country of origin evidence plainly demonstrated that they did share common immutable characteristics (gender and the experience of losing a husband) and were in fact viewed as a socially distinct group. The British experience is instructive here. In 2005, after an exhaustive review of the case law on particular social group and gender, the United Kingdom Immigration Appeal Tribunal noted "from experience that such cases often appear to become bogged down in pedantic and often unnecessary argument as to definition of the particular social group."'1 49 In that case, the tribunal took the step of itself formulating the group (as "Young 146 Id. at 9. Again note this is not in conformity with the UNHCR approach, which holds that the size of the group is irrelevant. UNHCR Guidelines on Particular Social Group, supra note 133, at 18; UNHCR, Gender Guidelines, supra note 64, 3 1. 147 The immigration judge did so without examining the second aspect of the immutability requirement, which is whether the characteristics were so fundamental to human dignity that she ought not to be required to change them. See generally Matter of E S N (Kan. City, MO Immigration Court, Jan. 14, 2003) (on file with The Hastings Center for Gender & Refugee Studies). 1481 d. at 13. 1 4 9 TB, [2005] UKAIT 00065 66(9 Mar. 2005). Columbia Journal of Gender an Law Iranian Women who refuse to enter into arranged marriages"), holding that this group was defined by its non-conformity rather than the persecutory outcome, which followed, and thus presented an acceptable basis for the particular social group.150 Thus, resistance or opposition to the oppression (which is surely implicit in the making of the refugee claim) rather than the actual experience of the persecution was centered as the basis of group membership. Ironically, this represents a belated acceptance of the position first put by UNHCR in 1985,151 restated over and over since then in various gender guidelines and articulated in the earliest of the American gender cases: that the basis of many women's claim to a particular social group will be their non-conformity with prevailing social mores. In sum, the American approach to gender-based particular social groups proved to be a major barrier to forced marriage claims. This was especially due to the rejection of broadly based groups (such as "women") because of to the American interpretation of additional requirements that the group be "visible," "particular" and "distinct." However, narrower groups were also in danger of being rejected if the formulation of the group was, or was seen to be, too reliant upon the persecution that its members experienced or if it was so specific that it was viewed by decision-makers as unlikely to be singled out by persecutors or as a "personal" experience rather than a group identity. In addition to the fact that American asylum law has consistently rejected both broadly and narrowly-framed gendered grou'ps, it is very troubling that in the United States the onus is so strongly upon the applicants themselves to frame the group, with little or no input or guidance from the relevant 150 Id. 57. Cf. Bcrishaj v. Gonzalez, 238 F.App'x 57, 62 n.3 (6th Cir. 2007) (noting that the Immigration Judge rejected the petitioner's claim that she fell under a recognizable particular social groupr constituting of"a woman who is not willing to go through a forced marriage" and doubting but not disturbing this ruling); Xiu Yun Chen v. Gonzalez, 229 F.App'x 413, 415 (7th Cir. 2007) (noting that the Immigration Judge rejected the petitioner's claim that she fell under a recognizable particular social group constituting of "young females who are against marrying" and doubting but not disturbing this ruling). 1SI See UNHCR, Refugee Women and International Protection, supra note 1, at § k. Columbia Journal of Gender and Law decision maker. In the American cases numerous applicants failed because they did not themselves frame an appropriate group, or do not frame it early enough in the process, 152 even in cases where their testimony as to the experience of forced marriage was accepted as truthful and persecution was established. 153 This failure is out of step with the formulation by decision-makers of gender-based social groups concerning marriage in comparable countries. These failures of analysis and engagement in the particular social group definition have severely retarded American asylum jurisprudence on gender more broadly and forced marriage in particular. In order to qualify as a refugee, persecution must be "for reasons of' one of the Convention grounds. Failure to properly define the particular social group also had flow-on effects in the analysis of the nexus between the Convention ground and the persecution, creating an additional doctrinal hurdle for claimants. B. Nexus: Marriage as Entirely Personal, Occasionally Commercial and Different in Foreign Places I Marriage was often understood by decision makers as a "cultural" or "traditional" experience such that there frequently was not seen to be any nexus between claims of forced marriage and a Convention ground. Pressure to marry was sometimes characterized as lacking a nexus because it was an experience that affected men also. 154 More commonly, claims were seen as lacking nexus- even when the conduct associated with the marriage was 52 See, e.g., Xuc Qin Li v. Bd. of Immigration Appeals, 172 F.App'x 385 (2d Cir. 2006). 153 See, e.g., Hua Lin v. Gonzalez, 205 F;App'x 879 (2d Cir. 2006); Berishaj v. Gonzalcz, 238 F.App'x 57 (6th Cir. 2007); Lin v. U.S. Att'y Gen., 319 F. App'x 777 (11 th Cir. 2009). '54 See, e.g., In re A-T, 24 1. & N. Dec. 296, 303 (BIA 2007) ("family pressures to accede to arranged marriages are not necessarily confined to females"), vacated on other grounds sub norn. In re A-T, 24 I. & N. Dec. 617 (Attorney Gcn. 2008). Columbia Journal of Gender an Law accepted as persecutory-because the harm or "dispute" was viewed by the decision-maker as "entirely personal."' 55 For example in the 2005 decision of AD, the immigration judge held that, "[T]he abuse the respondent suffered resulted solely from her uncle's desire to punish her for disobeying his request [to marry]." The abuse constituted a personal retaliation, not an act of persecution directed at a member of a particular social group. 1 56 In addition, in some cases the fact that the applicant was opposed to marrying a particular individual rather than opposed to arranged marriage in general was interpreted by the decision-maker as meaning that there was no nexus because the actions of the victim were based on personal preference. 157 In three different cases involving the payment of bride prices for young Chinese women adjudicated at different levels over a ten-year period, courts held that there was no nexus with a Convention ground because the marriage "dispute" was characterised as both "purely personal" and inherently commercial. In the 1999 case of SL the immigration judge stated: In this case, we have one party, the mayor, who wants to enforce the terms of a valid contract; while the other side, the respondent and her family, wants to void the terms of the contract. This case would be better litigated in civil court rather than Immigration Court. If respondent's family was persecuted after she left, it was because they breached the terms of the '5 See, e.g., Matter of S L (N.Y.C., NY Immigration Court, Oct. 7, 1999) at *14, aff'd sub non. (on file with The Hastings Center for Gender & Refugee Studies); Shu Lin v. Gonzalez, 148 F. App'x 38 (2d Cir. 2005). 156 Matter ofA-D-A (Bos., MA Immigration Court, Sept. 19, 2005) at * 13 (on file with The Hastings Center for Gender & Refugee Studies). 157 Petition for a Writ of Certiorari, Keisler v. Gao, 552 U.S. 801 app.. C at *20 (2007) (No. 06-1264); Syed v. Mukasey, 288 F. App'x 273 (7th Cir. 2008). Columbia Journal of Gender and Law contract and not because of the mayor's intention to punish them for one of the five enumerated [Convention] grounds in the Act.' 58 This decision was summarily affirmed by the BIA and again on judicial review by the Second Circuit in 2005.159 In 2003, in the oral decision delivered in Gao the immigration judge repeatedly characterised the issue as a "dispute between two families" over a "marriage arrangement" and as "some kind of a contract,"' 6 concluding that, "[H]er mother violated the oral contract that she had with this go- between, and that is what caused the anger by the boyfriend in this situation and not political opinion or a particular social group membership."' 61 This decision was summarily affirmed by the BIA. On appeal in 2006, the Second Circuit responded: To the extent that the Immigration Judge might have reasoned that the financial arrangement between the families somehow precluded a finding that Zhi's motive in targeting Gao was discriminatory, we reject this logic as antithetical to the very notion of individual rights on which asylum law is based. While Zhi may have a legitimate financial claim against Gao's parents, the possibility remains that if they continue to be I Matter of S L (N.Y.C., NY Immigration Court, Oct. 7, 1999) at *13-*14 (emphasis added) (on file with The Hastings Center for Gender & Refugee Studies). The Immigration Judge also refers to "fearing retribution over purely personal matters." Id. 159 Shu Lin v. Gonzalez, 148 F. App'x 38 (2d Cir. 2005). 160 Petition for a Writ of Certiorari, Keisler, 552 U.S. 801, app. C at *24a (No. 06-1264). 161 Id. at *25a (emphasis added). Columbia Journal of Gender an Law unable to repay his money, Zhi will force Gao to marry him. 162 Because the Second Circuit decision was vacated by the Supreme Court in 2007 on other grounds, this statement is left as obiter dicta only. Instead the more recent judicial authority from the Eleventh Circuit approves the "valid contract" approach to vitiating nexus. In this third and most recent case, Ying Lin, an Immigration Judge in 2006 accepted an applicant's claims that her parents promised her in marriage to a man who claimed her as payment for a gambling debt owed to him, yet went on to dismiss the harm experienced as "entirely a personal matter" between her family and the intended groom. 163 The BIA adopted and affirmed the Immigration Judge's decision in 2008. In 2009 the Eleventh Circuit affirmed the decision and held that: [A] nexus did not exist between the attempted involuntary marriage and rape and a protected ground, in that the testimony Lin gave did not show that she had been targeted on account of her membership in a particular social group. The involuntary marriage was for no reason other than repayment of her mother's gambling debt.164 It is significant here that the court did not see any nexus between a young woman being in a socially vulnerable position and her being treated by all parties as a form of payment. Unfortunately it appears that, despite the Second Circuit's efforts, this profound failure of analysis continues in American case law. It is striking that domestic discourse in the United 162 Gao v. Gonzalez, 440 F.3d 62, 70 (2d Cir. 2006) (emphasis in original). 163 Ying Lin v. U.S. Attorney Gen., 319 F.App'x 777, 779 (I Ith Cir. 2009). 164 Id. at 781-82 (emphasis added). Columbia Journal of Gender and Law States on human trafficking does not appear to have any impact upon the understanding of this issue in the asylum context. It is notable that, although in the international data set decision makers did at times regard marriage as "universal" or see abuse at the hands of family members as "personal", and thus failed to find a nexus with the particular social group in women's claims, such findings were far more common in the American cases. Furthermore, in the international data set decision makers never suggested in cases concerning a bride price or levirate marriage that what was at stake represented a valid contract, nor did they ever suggest that women could or should avoid persecution by repayment of such price, as discussed below. In addition, American asylum law appears to be stunted by an undue focus on the central motives of the persecutor in "singling out" the applicant when analyzing the question of nexus between persecution and the Convention ground. 165 This is out of step with international and UNHCR approaches to nexus, which stress that nexus can be satisfied either by the motives for the singling out by the persecutor or by the basis upon which there was a failure of state protection. 166 That is, a "purely personal" attack by a non-state actor upon a woman who refuses to marry should still satisfy the nexus requirement if the basis of the failure of state protection was the government's disinterest in protecting women from domestic or familial violence. In the international data set this dual nexus was well accepted. C. Persecution In 2006 the Second Circuit made arguably the strongest judicial pronouncement on forced marriage when it stated that "[Gao] might well be persecuted in China-in the form of 165 See, e.g., Syed v. Mukasey, 288 F.App'x 273 (7th Cir. 2008). 166 See UNHCR, Guidelines on Particular Social Group, supra note 133, 23; UNHCR, Gender Guidelines, supra note 64, 21. See also Min. for Immigration v. Khawar, (2002) 210 C.L.R I (Austl.); Islam v. Scc'y of State for the Home Dep't, [1999] 2 A.C. 629 (U.K.). Columbia Journal of Gender an Law lifelong involuntary marriage."' 167 Because this decision was vacated (although on another basis), American courts since have continued to regard the question of whether forced marriage constitutes persecution as "an open issue."' 168 Although an unwanted marriage was often articulated by claimants as either an integral aspect of another feared harm such as FGM or as an independent basis of the claim, in American case law it was infrequently received as either one. Applicants' assertions of forced marriage were not infrequently reframed in decisions as "arranged" marriage, 169 and expressed as "unwanted,"' 70 with persecutors restyled as "suitors"' 7 and their threats as "proposals."" 72 As with the particular social group issue, one of the most alarming trends in the American '67 Gao v. Gonzales, 440 F.3d 62 (2d Cir. 2006). 168 Ngengwe v. Mukascy, 543 F.3d 1029, 1036 (8th Cir. 2008). 169 See, e.g., In re A-T, 24 1. & N. Dec. 296, 297, 302 (BIA 2007), remanded on other grounds sub nom. in re A-T, 24 I. & N. Dec. 617 (Attorney Gen. 2008); Matter of Anon (Buffalo, NY Immigration Court, Dec. 14, 1999) (on file with The Hastings Center for Gender & Refugee Studies). See also Petition for a Writ of Certiorari, at *12, Keisler v. Gao, 552 U.S. 801 (2007) (No. 06-1264) (referring to the particular social group as "women in arranged marriages."). 7 Yan Hua Lin v. Gonzales, 246 F.App'x 746, 748-49 (2d Cir. 2007). 17' Id.; Matter of Anon (Chi., IL Immigration Court, Oct. 18, 2000) at *2, *7 (but note that this was a positive decision) (on file with The Hastings Center for Gender & Refugee Studies). 172 Xiu Yun Chen v. Gonzalez, 229 F.App'x 413 (7th Cir. 2007); Matter of A-D-A (Bos., MA Immigration Court, Sept. 19, 2005) at *4 (on file with The Hastings Center for Gender & Refugee Studies). In a related tone, see the use of forced marriage in disclaiming inverted commas in Keita v. Gonzalez, 2006 U,S, App. LEXIS 9484, at *2 (6th Cir. Apr. 13, 2006); Matter of E S N (Kan. City, MO Immigration Court, Jan. 14, 2003) at *10 (on file with The Hastings Center for Gender & Refugee Studies). Columbia Journal of Gender and Law cases was the complete failure to offer any analysis at all for the conclusion that forced marriage was not persecutory. 173 In cases involving a bride price or widow's dowry, it was striking how often the decision maker placed the onus upon the applicant to repay the sum (including extremely large sums, funds that were paid to others and sums paid many years earlier) in order to avoid persecutory consequences.1 74 For example, in the 2003 case of Ngengwe discussed earlier, where the applicant faced forced levirate marriage or repayment of her original dowry, the immigration judge suggested that since she had been in the United States for twenty months she, "Could send money to her in-law's family if she chose to do so to pay back any money that they view is owed... but she has made no attempt to alleviate the threat of future harm by paying back to them the 'bride's price'.' 1 75 Although summarily affirmed by the BIA, in 2008 the Eighth Circuit remitted this issue for reconsideration on the basis that the "IJ offered no analysis, and cited no law, on why the choice between forced marriage, death, or paying an unaffordable bride's price does not constitute persecution."'' 76 173 In only one appellate level decision we identified did the Court of Appeals find legal error and remand a case to the BIA for failing to consider the claim of forced marriage in its reasons. Notably, in that case the issue was whether the threat of forced marriage constituted a changed circumstance (justifying an out of time claim) rather than whether it constituted persecution per se. Joseph v. Gonzales, 240 F.App'x 726 (7th Cir. 2007). 174 See the three Chinese contract cases discussed above. See also Jin Chao Zheng v. Gonzales, 236 F.App'x 726, 727 (2d Cir. 2007) (finding no nexus when an official offered to waive a fine levied at the applicant's parents if she would marry his son because she "did not claim that her parents wcre unwilling to pay the fine, just that they could not afford to do so"). 175 Matter of E S N (Kan. City, MO Immigration Court, Jan. 14, 2003) at *17 (on file with The Hastings Center for Gender & Refugee Studies); see also id. at 18 ("she certainly could send money to her in-law's family to alleviate any possible threat that she might face"). 176 Ngengwe v. Mukasey, 543 F.3d 1029, 1036-37 (8th Cir. 2008). Columbia Journal of Gender an Law The case of A-T in 2007 is a particularly disturbing example of the failure to understand forced marriage as a form of persecution. In that case the BIA stated: It appears from the record that the respondent and her intended fianc6 are of similar ages and backgrounds, given the respondent's testimony that she and her cousin played together as children, and that the family used to joke that they would one day marry. Thus, if the respondent were to return to Mali and proceed with the marriage, it is not likely that she would be in a disadvantaged position in relation to her husband on account of her age or economic status. It is understandable that the respondent, an educated young woman, would prefer to choose her own spouse rather than acquiesce to pressure from her family to marry someone she does not love and with whom she expects to be unhappy. The respondent has also expressed valid concerns about possible birth defects resulting from a union with her first cousin. While we do not discount the respondent's concerns, we do not see how the reluctant acceptance of family tradition over personal preference can form the basis for a withholding of removal claim. 177 This first paragraph suggests that a forced marriage will only be harmful if there is a significant age or economic difference in the parties' relative positions, rather than constituting a human rights violation in and of itself. Somewhat 177 In re A-T, 24 I. & N. Dec. 296. 302-03 (BIA 2007), vacated sub noma. Matter of A-T, 24 1. & N. Dec. 617 (Attorney Gen. 2008) (remanding the case on the issue of FGM). The decision of the Attorney General does not make any finding on forced marriage itself. Columbia Journal of Gender and Law ironically, given that such claims have been mostly unsuccessful in the United States, it also implicitly suggests the payment of a bride price for a young woman by an older man is the paradigmatic example of forced marriage. It is also noteworthy that in the second paragraph, being forced to marry is transformed into "acquiescence" and "reluctant submission," suggesting that actual consent is not necessarily required. While the decision was vacated and remanded by the Attorney-General in 2008, this was on the basis of a failure to consider the relationship between past FGM and any future harm. The question of forced marriage was addressed merely in a footnote to the decision, which noted that the Board had appeared to make contradictory findings on whether the forced marriage and FGM were related and left this "for the Board to revisit or clarify on remand as needed."'1 78 What has been completely lost in the cases discussed above is the basic tenet that a forced marriage is persecutory because it breaches the fundamental human right to full and free consent in marriage. Yet, freedom to marry the partner of one's choosing has been repeatedly acknowledged as fundamental to human dignity in domestic constitutional litigation in the United States. While historically such challenges were to prohibitions on interracial marriage, more recently they have addressed same-sex marriage. 179 Regardless of whether legislative restrictions on marriage have been struck down or upheld, decisions in such cases have emphatically propounded the importance of marriage as both an individual right and as a voluntary social institution that fosters wider harmony and social 1 78 In re A-T, 24 1. & N. Dec. 617, 623 n.6 (Attorney Gen. 2008). '19 See, e.g., Baker v. Vermont, 744 A.2d 864 (Vt. 1999); Goodridge v. Dep't of Pub. Health, 798 N.E.2d 941 (Mass. 2003); Hernandez v. Robles, 794 N.Y.S.2d 579 (2005); Varnum v. Brien, 763 N.W.2d 862 (Iowa 2009). Columbia Journal of Gender an Law stability. 180 A contemporary Western ideal of marriage as romantic, egalitarian and companionate (as opposed to, say, dutiful, self-sacrificing or asymmetrical in power) is strongly present in such domestic case law and is never trivialized, as it was in In re A-T, as mere "personal preference." In Goodridge v. Dep 't of Public Health, a majority judgment of the Massachusetts Supreme Court declared that the state of marriage "nurtures love and mutual support" and is "at once a deeply personal commitment to another human being and a highly public celebration of the ideals of mutuality, companionship, intimacy, fidelity and family."' 8' That judgment also characterized marriage as crucial to the formation of self- identity and to individual self-fulfilment, claiming inter alia that, "[w]ithout the right to marry--or more properly, the right to choose to marry-one is excluded from the full range of human experience' 82 and "the decision whether and whom to marry is among life's momentous acts of self-definition.' 1 83 The ideas of marriage expressed in such cases-as the unique fulfilment of self-hood in a state of loving unity, and as an expression of human dignity fundamental to human rights-are conspicuous by their resounding absence in refugee cases concerning forced marriage where decision makers rarely, if ever, articulated coerced marriage (and concomitant inability to also choose to enter into a voluntary marriage with someone else) as a persecutory harm. 180 Indeed, in the past decade a number of commentators have noted that both challengers and defenders in same-sex marriage litigation in North America have utilized strikingly similar characterisations of voluntary marriage as an idcaliscd universal good. See, e.g., Nancy Polikoff, For the Sake of All Children: Opponents and Supporters of Same-Sex Marriage Both Miss the Mark, 8 N.Y. CITY L. REV. 573 (2005). For a discussion of relevant Canadian cases, see Katherine Ostcrlund, Love, Freedoom and Governance: Same-Sex Marriage in Canada, 18 Soc. & LEGAL STUD. 93 (2009). 181 Goodridge, 798 N.E.2d at 948, 954. 182 Id. at 957. 183 Id. at 955. Columbia Journal of Gender and Law IV. WHO WINS? THE SUCCESSFUL CLAIMS Despite the persistent doctrinal hurdles in applying refugee law principles to forced marriage cases as explored above, some of the decisions in our data set were positive. As noted earlier, at the broadest level the rate of positive decisions in the United States was comparable with the international data set. 184 We gathered together the positive American decisions to analyze the key elements of a successful claim. The results of this analysis are disappointing. There were fifteen positive decisions in the United States portion of our data set. We counted "positive" decisions as those in which the outcome was what the claimant sought at that stage. Of the fifteen positive decisions, eleven were judicial review decisions by appellate courts. This means that a "positive" case was often merely a remittal for redetermination of the claim rather than an actual grant of asylum. Furthermore, as many of the positive outcomes occurred at the appellate level on judicial review there is often little information about the factual background to the decision. The characteristics of successful claims varied considerably. The fifteen claimants came from nine different countries, with China as the only country of origin with more than one successful claimant. 185 The high number of claimants from China is likely not indicative of a greater openness to these claims, but instead reflects the high number of asylum applicants each year from China as well as the considerable evidence that forced marriage is an important human rights issue in China and 184 Positive results among the United States data set constituted thirty- one percent of all cases, whereas in the international set such cases constituted thirty-two percent of all cases, although as noted earlier this rate is misleadingly high. See supra notes 105, 110-1I. 85 The other countries were Iran, Pakistan, Mali, Cameroon, Zambia, Togo, Nigeria and the Ivory Coast. Columbia Journal of Gender an Law the specific statutory recognition of forced abortion and sterilization can constitute persecution in US law. 186 Of the positive decisions, Gao, discussed above, was later overturned by the Supreme Court. Prior to the Supreme Court decision in Gao, the Second Circuit issued a small series of three positive decisions relying on its original Gao reasoning. 187 The Second Circuit has not issued another positive decision following the Supreme Court decision, despite the fact that the decision did not actually rule out the possibility of finding a particular social group or persecution in the context of forced marriage. t88 In four of the positive decisions, including Kasinga, forced marriage was not mentioned in the analysis at all, only in the facts. ' 89 In a further two decisions forced marriage appears only as an incidental factor related to FGM. 190 Excluding the 186 8 U.S.C. § 1101(a)(42) (2009); see also supra note 58. The problems of forced marriage in China arc addressed by successive annual editions of the Department of State's nHuman Rights Report, as well as in each annual edition (2001 to 2009) of the Trafficking in Persons Report. See generally U.S. DEP'T OF STATE, HUMAN RIGHTs REPORT: CHINA (INCLUDES TIBET, HONG KONG, AND MACAU) (2009), available at http://www.state.gov/ g/drl/rls/hrrpt/ 2009/cap/135989.htm.; U.S. DEP'T OF STATE, TRAFFICKING IN PERSONS REPORT 2009, available at http://www.state.gov/g/tip/rls/tiprpt/2009/index.htm. . 187 Yi Meng Tang v. Gonzales, 200 F.App'x 68 (2d Cir. 2006); Himanje v. Gonzalez, 184 F.App'x 105 (2d Cir. 2006); Bao Yuci Chen v. Gonzalez, 175 F.App'x 492 (2d Cir. 2006). No other circuit of appeal produced a positive decision relying on the Second Circuit's Gao reasoning. '88 This conclusion was drawn by the Eighth Circuit in Ngengwe v. Mukasey, 543 F.3d 1029 (8th Cir. 2008). 189 In re Kasinga, 21 1. & N. Dec. 357 (BIA June 13, 1996); Matter of A-T, 24 1. & N. Dec. 617 (Attorney Gen. 2008); Yan Dan Li v. Gonzalez, 222 Fcd.Appx. 318 (4th Cir. 2007); Xuan Li Zheng v. Ashcroft, 36 Fed.Appx. 301 (9th Cir. 2002). 190 Matter of Anon (Buffalo, NY Immigration Court, Dec. 14, 1999) (on file with The Hastings Center for Gender & Refugee Studies); Uanreroro v. Gonzales, 443 F.3d 1197 (10th Cir. 2006). Columbia Journal of Gender and Law positive cases where the claim of forced marriage is not mentioned in the decision, cases which were analyzed only as FGM claims, and the four cases belonging to the Gao gap (including Gao itself), as well as one case in which the forced marriage argument was found not credible but a positive decision was granted on another ground, 191 a mere four positive decisions remain. In short, counting 15 of our 48 decisions as positive vastly over-represents the chance of 'success' for forced marriage refugee claimants. The singularly most striking factor in the positive American cases is that forced marriage was in itself never found to be a form of persecution in any decision. While only four decisions in the international data set contained a strong analysis of forced marriage alone as a form of persecution of a vulnerable group, none of the successful American decisions did so. 192 This is directly at odds with international human rights standards and with all guidelines-including the INS Gender Guidelines-on gender-related persecution. In the international data set, we found that positive claims were most often related to factors additional to the forced marriage itself. This "something more" was sometimes an understanding that forced marriage would constitute a catalyst for other harm such as domestic or sexual violence or FGM. In claims brought by gay men, forced marriage was often considered one way that their sexual orientation might become known, and therefore would lead to persecution for that reason. While such cases did not center forced marriage in itself as persecution, they at least recognized the linkages between forced "191 Yi Long Chen v. Gonzalez, 198 F.App'x 158 (2d Cir. 2006). 192 The four decisions in which there was not "something more" in the case, such as FGM or polygamy, in addition to the forced marriage are: Vidhani v. Canada (Min. of Citizenship and Immigration), [1995] 3 F.C. 60 ; Eimani v. Canada (Min. of Citizenship and Immigration), [2005] F.C. 42; Houssainatou v. Canada (Min. of Citizenship and Immigration), [2002] F.C.T. 2004; NS (Social Group, Women, Forced Marriage) Afghanistan CG, [2004] UKIAT 00328 (U.K.). In two of those four cases (Eimani v. Canada and Houssainatou v. Canada) the decisions are judicial review with little known about the facts. Columbia Journal of Gender an Law marriage and other forms of gendered harms and harms related to sexuality. 193 Like many of the positive decisions in the international data set, the US positive decisions generally involved 'something more' beyond the marriage itself, and that this 'something more' is profoundly 'other' to the experience of a Western decision maker. The most recent of these is Ngengwe, which involved the culturally distant practice of levirate marriage. 194 Likewise, in the 2007 Joseph decision, a woman from Pakistan feared a forcible marriage on return and presented a narrative which included violent attacks by her family and a history of so-called "honor killings."'195 These cases all fit into the pattern of "othering" or "exoticizing" women refugee claimants, presenting them as victims of distant and backwards "traditional" cultural practices. The tendency to exoticize gender claims is now well documented, 196 and FGM cases are the clearest example of this 193 In the international data set the decisions that actually characterize forced marriage itself as persecutory (on its own or as a precipitating factor) were: MZXFJ v. Min. for Immigration and Multicultural Affairs, (2006) F.M.C.A. 1465 ((Austl.); N95/10037, [1997] RRTA 623 (Feb. 25, 1997) (Austl.); N98/21046, [1999] RRTA 1872 (Oct. 12, 1999) (Austl.); N98/25465, [2001] RRTA 27 (Jan. 12, 2001) (Austl.); V0618399, [2006] RRTA 95 (June 22, 2006) (Austl.); 071426303, [2007] RRTA 132 (June 29, 2007) (Austl.); Re X., 2000 CanLIl 21442 (I.R.B); X. v. Canada (Immigration & Refugee Bd.), 2001 CanLIl 26862 (I.R.B.) X. v. Canada (I.R.B.) 2001 CanLIl 26862 (I.R.B.); Vidhani v. Canada (M.C.I.), 1995 CanLII 3606 (F.C.) (June 8, 1995); Traorc v. Canada (M.C.I.), 2003 F.C. 1256 (Oct. 28, 2003); Eimani v. Canada (M.C.I.), 2005 F.C. 42 (Jan. 17, 2005); Houssainatou v. Canada (M.C.I.), 2002 F.C.T. 2004 (Nov. 26, 2002); NS (Social Group) Afghanistan CG, [2004] UKAIT 00328. 194 Ngengwc v. Mukascy, 543 F.3d 1029 (8th Cir. 2008). '95 Joseph v. Gonzales, 240 F.App'x 726 (7th Cir. 2007). 96 See, e.g., Audrey Macklin, Refugee Women and the Imperative of Categories, 17 HUM. RTS. Q. 213 (1995); Sherene Razack, Domestic Kiolence as Gender Persecution: Policing the Borders of Nation. Race and Gender, 8 CAN. J. WOMEN & L. 45 (1995); Jacqueline Bhabba, Internationalist Gatekeepers?: The Tension between Asylum Advocates and Human Rights, 15 HARV. HUM. RTS. J. 155 (2002); Oxford, supra note 84. Columbia Journal of Gender and Law pattern.197 Indeed, even when the claimants explicitly linked their experience of FGM to forced marriage, decision makers did not. 198 Furthermore, in the American cases, the practices of FGM are described in extraordinary, almost prurient, detail. This is a striking distinction in comparison with the cases involving FGM we reviewed from Australia, Canada and the United Kingdom, and it adds considerably to the exoticization and othering of the claimants. It was striking that the American cases did not ever analyze forced marriage as a catalyst for other forms of forms of gendered harm; this was so even when claims of FGM were closely linked to marriage. Of the substantively positive decisions, two were written prior to 2001 when the combined effect of legislative and administrative changes in the wake of the 9/11 attacks brought substantive changes to American law in a much harsher climate for refugee claims. 199 It is remarkable that the oldest positive American decision in our data set, from 1994, comes closest to defying an exoticizing pattern. 200 The claim was brought by an Iranian woman from a politically dissident family who faced coercion to marry a disabled Iranian war veteran following the arrest and disappearance of her husband. 20 1 This decision of the Ninth Circuit contains a strong statement that forcible marriage constitutes persecution in certain circumstances: "There can be 197 See in particular Connie Oxford's finding in the United States asylum context that FGM was assumed to constitute persecution whereas domestic violence was not. Oxford, supra note 84. 198 See discussion supra Part II.A. 199 See, e.g., supra note 46; Donald Kerwin, The Use and Misuse of "National Security" Rationale in Crafting US Refugee and Immigration Policies, 17 INT'LJ. REFUGEE L. 749 (2005). 200 It is also notable that only two decisions refer to the U.S. Gender Guidelines. See Kamaleddin v. INS, 1994 U.S. App. LEXIS 6939 (9th Cir. Apr. 4, 1994); In re Kasinga, 21 I. & N. Dec. 357 (BIA 1996) (Rosenberg, J., concurring). These are the two oldest American decisions in the data set and are both positive outcomes. 201 Kainaleddin, 1994 US App. LEXIS 6939, at *2-3. Columbia Journal of Gender an Law no doubt that a government that coerces a woman to marry against her will on account of imputed political opinion has engaged in persecution. '20 2 While there were many factors besides forced marriage involved in the case, the court engaged in markedly less exoticising than is typical. Given our earlier discussion of the myriad of problems in establishing particular social group parameters in gender claims, it is worth noting that the strong analysis of forced marriage as persecution in this case was not made on the basis of a gendered particular social group but rather rested upon the political opinion ground. 20 3 In the international data set we found an understanding of "force" in forced marriage that rested upon proxies for consent such as the level of education, age, urbanity or "independence" of the applicant (with independence itself represented by the proxies of income and unaccompanied travel). Being educated, over the usual marriageable age for the country of origin, residing in an urban rather than a rural area or exhibiting "independence" through employment or past travel without parental supervision were frequently taken to mean that female applicants were not "disempowered" and thus could refuse marriage (and could also therefore relocate away from any persecution or seek state protection). 20 4 The approach of United States' decision makers to understanding consent was even narrower. In the American cases, proxies for consent comprised only two factors: payment to another for the marriage and being a child. For instance Gao, and the Second Circuit trio of cases decided in the "Gao gap"2° 5 (plus an earlier positive immigration judge decision concerning an applicant from 202 Id. at * 15. 203 Id. at * 14. 204 See Dauvcrgnc & Millbank, supra note 6, at 28. 205 Yi Long Chen v. Gonzales, 198 F. App'x 158 (2d Cir. 2006); Bao Yuci Chen v. Gonzalez, 175 F. App'x 492 (2d Cir. 2006); see also discussion of the Gao decision supra Part II; Himanje v. Gonzalcz, 184 F. App'x 105 (2d Cir. 2006). Columbia Journal of Gender and Law China) 206 all involved the claimant being "sold" into marriage as a minor or young woman through payment to family members or third parties. In this, way, the understanding of "force" in forced marriage refugee claims came to resemble the very limited discourse around forced marriage in American domestic politics, as concerning only child marriage and human trafficking. 207 Our canvass of the positive cases reinforces our conclusion that the United States' decision makers are far behind those in Australia, Canada and the United Kingdom in terms of analyzing gender-related persecution. In addition to not finding a single case with a straightforward holding that forced marriage in and of itself could constitute persecution, we also did not find any engagement with international human rights standards. Of the few cases that were successful on a substantive basis, we found that the underlying facts reflect an extreme exoticization of the women involved. It is also astonishing that we found no gay men or lesbians among American claimants, as they comprised a significant portion of our international data set (with reasonably high success rates). Our conclusions about substantive analysis in the United States cases are particularly distressing given that the procedural hurdles for claimants are far more onerous in the United States than in Australia, Canada or the United Kingdom. Even if one is able to surmount those hurdles, a claim that forced marriage is a form of persecution related to gender and sexuality appears to have little chance of success in the United States. V. CONCLUSION In March 2009, Reem Al Numery of Yemen was recognized by Secretary of State Hillary Clinton as one of eight 206 Matter of Anon (Chi., IL Immigration Court, Oct. 18, 2000) (on file with The Hastings Center for Gender & Refugee Studies). 207 See discussion supra Part I. Columbia Journal of Gender an Law "International Women of Courage. ' '20 Along with women from Afghanistan, Guatemala, Iraq, Malaysia, Niger, Russia and Uzbekistan, she was recognized for "courage and leadership" in the struggle for "social justice and human rights.'" Reem received this honour because of her fight against her own forced marriage to a thirty-year old cousin when she was twelve years old. 210 Bestowing this honour on Reem recognizes her as an individual, as well as the circumstances of others like her-of the vulnerable group of which she is a member. Yet, this recognition at the highest level of politics and policy does not carry into asylum law. This finding parallels and amplifies what we found in Britain, where a multilayered and highly nuanced domestic debate about forced marriage largely failed to influence refugee jurisprudence. In the United States the disjuncture between domestic policy and asylum jurisprudence was even starker. We found that despite the development of gender guidance by the INS some fourteen years ago and despite ongoing commitment to training around gender issues, there was a profound reluctance to accept any form of broadly based gender group in asylum law, accompanied by marginalization of all but the most extreme and exoticized forms of gender-related harm (such as FGM). Although the issues of child marriage and human trafficking have received considerable and increasing domestic attention in the United States, even these forms of forced marriage were rarely understood as persecutory harm in the United States' asylum cases. Our findings reflect the uneasy relationship between refugee law and human rights law. Refugee law has not been able to fully embrace human rights norms and unfolds against a 208 Press Release, U.S. Dep't of State, The Secretary of State's 2009 International Women of Courage Awards (Mar. 5, 2009), available at www.state.gov/r/pa/prs/ ps/ 20 0 9 /0 3/I 20070.htm. 209 Id. 210 Press Release, America.gov, Child Bride Fights Against Forced Marriage (Mar. 6, 2009), available at http://www.america.gov/st/hr-cnglish/ 2009/March/ 20090306161153 ajesromO.9992792.html. Columbia Journal of Gender and Law floodgates fear, a persistent cultural relativism and, in the United States more than any other country we studied, a foreign policy ethos of exoticized harm elsewhere. In the case of Goodridge, the Massachusetts Supreme Court said of marriage that it "fulfils yearnings for security, safe haven, and connection that express our common humanity. '2 11 This description could as readily be used to express the loftiest ideals of refugee law. Yet, when forced marriage is claimed as harm in the refugee context, the notion of our common humanity is unrecognizable. We began our investigation of forced marriage as a basis of persecution because of our interest in exploring the ways that gender and sexuality are understood in refugee law, expecting to find that analysis of these issues would have developed and become more complex in recent years. It became evident through this study that refugee jurisprudence in the United States is substantively impoverished as well as procedurally hobbled, and the protection that it offers falls well short of international standards and respect for our common humanity. 211 Goodridge v. Departmcnt of Public Hcalth, 798 N.E.2d 941, 954 (Mass. 2003).