Reconfiguring the Deserving Refugee: Cultural Categories of Worth and the Making of Refugee Policy

Abstract

Studies on asylum give little explanatory power to the role of categories of worth in how lawmakers formulate asylum law in lack of a clear policy framework for determining eligibility for asylum status. This article contends that during periods of policy upheaval, distinctions of worth shift to forefront lawmaking: lawmakers renegotiate the moral boundaries between categories of deserving and undeserving refugees to give content in ambiguous law. In the United States, lawmakers drew on the concept of immutability—the notion that to be worthy of protection you must be targeted on account of traits beyond your control to change—to distinguish between “undeserving” Central Americans fleeing civil wars and gang violence, and “deserving” women subjected to gender violence. Understanding how categories of worth inform the formulation and implementation of law in periods of policy upheaval advances understandings of asylum policy and expands scholarship on the role of ideas about worth in processes of institutional change.

1. Introduction

In the early 1990s, new claims involving gender-related harms gained recognition and prominence in the field of U.S. asylum adjudications. These claims, which include harms such as rape, female genital cutting, and domestic violence, do not involve standard forms of government-sponsored persecution of well-recognized political, religious, and ethnic minorities. The increasingly inclusive approach of high-level government officials and adjudicators toward this new group of claims posits a puzzle to immigration scholars: at the same time that support by lawmakers and public officials across the political spectrum grew for this class of novel claims, greater restrictions were imposed on immigrants' and asylum seekers' access to legal status and other social benefits, and public sentiment toward asylum seekers became increasingly hostile.

This article contends that the development of gender asylum is reflective of a broader reconfiguration in the meaning of worthiness in the context of asylum claims: classifications of worthiness for asylum gradually ceased to center on a person's motivation to flee and came to focus on the immutable trait on account of which a person is persecuted. This article explores how classifications of worthiness shape the social construction of law during periods characterized by a policy breach where institutionalized legal scripts cease to apply. When faced with a policy vacuum, lawmakers draw on moral distinctions to give content to ambiguous legal categories. In the process, the meanings of both law and worth change. After the Cold War, the moral concept of immutability provided lawmakers a conceptual template for differentiating between deserving asylees and a growing category of undocumented illegal immigrants who, under the Refugee Act of 1980, were for the first time eligible to apply for asylum. Central Americans fleeing the violence of civil wars in the 1980s and gangs after the early 2000s were categorized as undeserving of asylum protection because the traits on account of which they suffered persecution are not immutable. Conversely, crimes that target persons on account of their sexuality and/or gender came to define immutability in the context of asylum. In turn, a hierarchy of harms is created whereby harms violating a person's sexuality and body gain precedence over harms targeting persons on account of profession, socioeconomic status and living environment.

The article examines asylum policy in the transition from a Cold War policy framework in which communism served as the primary criterion for distinguishing between refugees and other categories of immigrants (refugee status was mostly restricted to persons fleeing communist countries), to a post-Cold War regime in which individual human rights figured more prominently, at the same time that increasing numbers of immigrants enhanced both the public's and administration's anxieties about national security and border control. This research investigates how lawmakers negotiated these conflicting pressures when applying asylum law in the absence of a clear policy framework for distinguishing refugees from other categories of immigrants. My analysis focuses on how sectors of the state involved in formulating policy through court decisions apply the category “persecution on account of membership in a particular social group”—one of the five categories of persecution enumerated in the refugee definition. The Refugee Act of 1980 defines a refugee as any person with a well-founded fear of persecution on account of race, religion, nationality, membership of a particular social group, or political opinion. The drafters of the Act did not define any of the five persecution grounds. For the categories of “race,” “religion,” “nationality,” and “political opinion,” lawmakers could rely on shared institutionalized knowledge rooted in Cold War politics and American political culture. This was not the case with “membership of a particular social group,” a category that had no clear referent in U.S. political culture and that lawmakers perceived as ambiguous. Following the collapse of the communist-centered policy framework guiding U.S. refugee policy throughout the 20th century, and in lack of an alternative programmatic framework upon which to rely, the category's inherent ambiguity made it ever more important: legal actors involved in the making of asylum policy strategically made use of the open-endedness of the term to renegotiate the meaning of asylum status. Examining how lawmakers defined the category membership of a particular social group thus provides a unique opportunity to investigate empirically how legal actors set to construct asylum law in the absence of an established policy framework. Moral categorizations are central to this process.

In what follows, I review existing accounts on the role of ideas and values in law-making, outlining a framework for understanding the interrelations between conceptions of worth and law-in-action during periods characterized by policy change. I then draw on an array of legal documents including case law, legislative debates, memoranda, and administrative guidelines on how to interpret and apply asylum law in the post-Cold War period, in addition to interviews with high-level immigration officials and lawyers, to demonstrate the utility of this framework for understanding changes within the institution of asylum, specifically as it explains the growing recognition of gender-based asylum claims. I demonstrate how the moral concept of immutability, defined in relation to gender violence, provided lawmakers with a template for translating the conflicting political pressures of attempts to curb illegal immigration and calls for broader protections to victims of human rights violations. I conclude with a discussion of the implications of these findings for immigration scholars as for theorists concerned with the role of moral categories in the social construction of law.

2. Theoretical Discussion

A burgeoning body of scholarship demonstrates how moral classifications inform social action. Scholars have shown how people draw on, and produce, moral distinctions to make sense of, and justify, institutional constraints and established strategies of action (Boltanski and Thévenot 2006; Lamont 1992; 2000; Swidler 1986; 2013). The moral classifications agents draw on to make sense of their judgments and commitments vary across lines of nationality, race, socioeconomic status, and gender (Amiraslani et al. 2017; Lamont and Thévenot 2000; Saguy 2003). Ordinary people and state officials are influenced by internalized moral schemas in ways that they are largely unaware of and unable to articulate (DiMaggio 1997; Vaisey 2009). Studies on the street-level work of bureaucrats and state officials show how decision makers' practices and interactions with clients are shaped by shared perceptions of clients' worth; decision makers often screen out information that does not respond to their preestablished moral classifications (Hasenfeld 2000). Thus, the state is more than a bureaucracy with rules and procedures, as state actors draw on categories of worth to make sense of complex persons in ways that cannot be reduced to the organization's or their own self-interests (Fassin 2015).

A central insight of this literature is that law and social policy must resonate with the predominant moral classificatory schemes in a particular society if they are to be enduring. People typically have a hard time justifying policy provisions that fall outside the scope of collectively shared distinctions of worth (Campbell 1998; Skrentny 2006; Steensland 2006). Some scholars have shown how embedded moral schemas constrain the formulation and implementation of new rights when these threaten to undermine the existing moral order (Blair-Loy 2009). In Steensland's (2006) study on the failure of the proposed Guaranteed Annual Income legislation, passage of proposed legislation failed precisely because it threatened to undermine deeply embedded moral distinctions between the deserving and undeserving poor. Even when new legal rights are successfully passed, institutionalized norms, and values constrain how they will be enforced. In the case of the Family Medical Leave Act, despite the existence of the law, deeply held beliefs and expectations about what constitutes good and productive work created a barrier to the implementation of new workplace rights such as family leave (Albiston 2005).

At the same time, in the process of applying rules and performing procedures the meaning of these moral classifications is negotiated and redefined. Legal institutionalists contend that processes of legal compliance are informed and constrained by embedded moral values at the same time that they transform the meaning of these values (Edelman 1992; Edelman and Talesh 2011). For example, in the case of equal protection law, managers devised practices of compliance that eventually transformed the original meaning of equitable racial and gender representation to a managerial ideal according to which varying backgrounds and viewpoints promote organizational productivity (Edelman et al. 2001). This line of important work contributes to our understanding of how moral schemas inform, and in turn are informed by, law-in-action.

Recently, an increasing number of studies have examined the relation between moral schemas and law in “unsettled times” (Swidler 1986), when policy undergoes fundamental change, and state officials find themselves faced with new social, political, or economic problems for which the current policy framework offers no clear solutions (see Campbell 2002; Dobbin 1993, Hall 1993). Scholars show that during periods characterized by significant social transformation, distinctions of moral worth shift from the background to foreground policy debates and have a direct effect on policy and action (Campbell 1998; Steensland 2006; Swidler 1986). Others focus on the mechanisms that allow certain moral beliefs to exert extraordinary political influence during moments of social turmoil and national crises: Somers and Block (2005) propose the idea of “epistemic privilege” to describe ideas with the causal power to undermine, dislodge, and replace a previously dominant ideational regime. Swidler (1986: 278) argues that during unsettled periods, ideologies have a direct effect on policy and action by providing “explicit, articulated, highly organized meaning systems.”

However, to fully understand the role of moral categories in the process of constructing ambiguous law when there is no clear policy framework to guide policy makers on how to make sense of new social problems, it is necessary to study the process of legal construction: how legal actors draw on, negotiate, and redefine moral scripts found in their broader environment, to resolve conflicting political pressures. In the case of U.S. asylum after the Cold War, my data suggest that the collapse of the Cold War framework, which had guided U.S. refugee policy from the outset, generated a policy void: there was no clear alternative policy framework for how to distinguish refugees from the growing category of undocumented immigrants, who under the Refugee Act of 1980 were for the first time eligible to apply for asylum. In lack of a shared policy framework for distinguishing refugees from other categories of immigrants, adjudicators and policy makers struggled to define the meaning of the Refugee Act's ambiguous legal definition of a refugee. Unable to default to Cold War policy prescriptions, and in lack of sufficient international guidance on the meaning and scope of the term persecution and the five protected grounds, legal actors crafted the meaning of ambiguous asylum law through a process of “moral bricolage”¹: they drew on, and recombined, moral scripts found in their broader organizational environment, and used these as primary criteria for negotiating the conflicting political pressures of humanitarianism and exclusionary state policies in the process of defining asylum law.

Accounts of 20th-century U.S. asylum and immigration policy document the role of foreign policy (Boyle and Busse 2006; Coutin 2001; McKinnon 2016; Sales 2002), anti-immigration forces (Capetillo-Ponce 2008; Chavez 2013), political culture (Gibney 2004; Pratt and Valverde 2002; Yoo 2008), and shared stereotypes (Flores and Schachter 2018; Luibhéid 2013; Volpp 1996) in forming distinctions between categories of “worthy” and “unworthy” immigrants across national, ethnic, and gendered lines. Such distinctions have an effect on immigrant behavior and identity (Andrews 2018; Coutin 2003; Menjívar 2006) and result in the reinforcement of patterns of inequality between the Western hegemony and the Third World (Bhabha 2002; Razack 1995).

Gender asylum is one area of asylum law where the role of conflicting state interests, socioeconomic values and humanitarian obligations in constructing refugee deservingness is especially apparent (Bhabha 1996; 2002). Scholars of gender and immigration document how sexual norms and gendered hierarchies shape how state actors draw the line between groups of deserving and undeserving immigrants (Luibhéid 2013) and work to legitimize policies that criminalize immigrants and sanction non-Western regimes (Bhabha 2002; McKinnon 2016). Given this thorough documentation, it is striking that, to date, we know little about how the parameters of worthiness for asylum came to be reconfigured after the Cold War. Through analysis of particular social group jurisprudence, and interviews with lawyers and policy makers, I show how subsequent to the end of the Cold War and the collapse of the Cold War refugee policy framework, definitions of worthiness for asylum shifted from a person's political or religious motivation to flee, to an immutable trait on account of which one is persecuted, with gender becoming a central criterion for defining immutability in this context. In turn, adjudicators, lawyers, and state officials drew on this new conception of worth, centered on immutability, to give content to ambiguous asylum law.

3. Data and Methods

This article develops a framework for analyzing how groups of legal actors drew on moral scripts found in their environment to demarcate the meaning and scope of asylum eligibility in the United States after the Cold War. The empirical analysis for this article draws on multiple sources of data including case law, legislative debates, agency committee transcripts, agency training guidelines, administrative data on asylum applications filed between 1995 and 2015, in addition to interviews with high-level immigration officials and lawyers, to illustrate how different groups of legal actors negotiate the boundary between categories of deserving and undeserving refugees.

I analyzed all Board of Immigration Appeals (BIA) precedent decisions concerning the meaning and intended application of the category “membership in a particular social group.” I focus on the BIA because it is the highest administrative body for interpreting and applying immigration laws. Board member judges are politically appointed by the attorney general. Their decisions constitute binding precedent on all lower-level asylum adjudicators in the country. In this respect, they have a central role in both the formulation and implementation of asylum policy. The Department of Justice (DOJ) provides online access to all BIA precedent decisions. Using the search terms “asylum,” “well-founded fear of persecution,” and “particular social group,” I identified over 100 precedent decisions pertaining to asylum. I then categorized these into the following subgroups: “human rights violations,” “gang related,” “gender,” “definitions of persecution,” “social group,” “political opinion,” “race,” “nationality,” and “religion.” I analyzed these decisions with the goal of identifying changes in how adjudicators draw on, and at times redefine, worthiness for asylum when trying to make sense and define ambiguous asylum law. By analyzing a diverse array of BIA' decisions, from 1980 and through 2018, I was able to identify how developments in “particular social group” jurisprudence coincide with broader developments in asylum law pertaining to the meaning of humanitarian relief and persecution.

I supplement analysis of case law with (1) legislative debates preceding the enactment of the Refugee Act of 1980 about the purpose and intended meaning of asylum relief, (2) policy memoranda issued by the U.S. Citizenship and Immigration Services (USCIS), an agency housed within the Department of Homeland Security (DHS), and by the DOJ, on how to implement new regulations governing asylum, and (3) training modules issued by the Refugee, Asylum, and International Operations Directorate of USCIS on how asylum officers should interpret the five grounds of persecution when applying asylum. The training module on “particular social group” is not publicly available. I obtained access to this training module, titled “Officer Training Guidelines on “Nexus-Particular Social Group,” after filing a Freedom of Information Request in February 2016 to USCIS.

I supplement these legal documents with administrative data on asylum applications which I obtained by filing a Freedom of Information Request in June 2015 to USCIS. The administrative data include case-specific information on all asylum applications processed by the Asylum Office between 1995 and 2015. The data are based on information recorded in the Refugee Asylum and Parole System, which tracks the processing of asylum applications at the Asylum Office. For each principal applicant filing an asylum claim, basic sociodemographic information is recorded in addition to the applicant's country of origin and alleged basis of persecution (i.e., race, religion, nationality, membership in a particular social group, or political opinion).

I also conducted interviews with the former Chair of the BIA (1995–2001), the former Director of U.S. Citizenship and Immigration Services (2014–2017), in addition to 15 immigration lawyers who actively participated in developing the meaning of the legal category “membership in a particular social group.”² Interviews with immigration officials and lawyers averaged about 90 minutes each. Interviews were tape-recorded and transcribed word-for-word when participants consented. I used open-ended and inductive interviewing techniques to identify the criteria interviewees drew on when constructing ambiguous legal categories in lack of a clear policy framework (Lamont 2009). I developed coding categories inductively to capture how asylum officials and immigration attorneys drew on moral distinctions to cast content into the category “membership of a particular social group.” This includes codes referring to the criteria interviewees used when distinguishing between different groups of asylum seekers, their interpretations of asylum law, and competing notions of worth.

4. Refugee Status, U.S. Immigration Law, and Foreign Policy

During the Cold War, refugee status was mostly restricted to persons fleeing communist-dominated countries (Loescher & Scanlan, 1998 Reimers 1992). Accordingly, communism (e.g., being a resident of a communist-dominated country) provided a means to distinguish both legally and conceptually between “refugees” and all other categories of immigrants. The Refugee Act of 1980, which codified the UN definition for a refugee, repealed the prior communist-centered definition. The new law was largely the result of mounting political pressures to align U.S. law with the UN Convention definition and its 1967 protocol to which the United States was a signatory state. While the signing of the Protocol in 1968 was initially viewed by U.S. administrators as a purely symbolic act, it provided an opening for legal mobilization: throughout the 1970s cases bringing attention to the discriminatory treatment of asylum seekers from noncommunist countries found their way into the court system. These cases eventually worked to upend the former classificatory system, setting the stage for the Refugee Act of 1980 and the emergence of a radically new logic guiding refugee policy (Hamlin and Wolgin 2012).

The Refugee Act of 1980 was not intended as a revolutionary change (Hamlin 2012). Quite the contrary, policy makers involved in the drafting of the Act believed it to be “an extension of the Cold War by other peaceful means.”³ Policy makers viewed the enactment of the new law as a symbolic gesture that showed the United States' commitment to the UN Convention and to international human rights law; most lawmakers at the time shared the assumption that the United States would continue to be a country of overseas resettlement (rather than a country of first asylum, permitting refugees to enter its territory), and that the majority of refugee admissions would continue to predominantly be from communist-dominated countries. Judge Bruce J. Einhorn, a former United States Immigration Judge in Los Angeles from 1990 to 2007, who also participated in the drafting of the Refugee Act of 1980, explains that at the time no one perceived the law as introducing a substantially new definition of a refugee. Among the drafters “the widespread conception was…that the great majority, the overwhelming majority of those that would come would be those that would flee Soviet-style oppression.”⁴ Judge Paul Wickham Schmidt, who between 1979 and 1981 was acting general counsel of the Immigration and Naturalization Service (INS), later becoming the Chairman of the BIA (1995–2001), and an Immigration Judge at the U.S. Court in Arlington Virginia (2003–2016), recalls that during the proceedings leading to the enactment of the new Refugee Act of 1980, the guiding assumption of the drafters was that “we will get to select the people we want…people were [not] thinking of the mass migration with individuals ending up on the U.S. border…A lot of the assumptions were based on: ‘we will be the country that helps countries of first asylum. We will get to select the people we want.’”⁵

The reality ended up much different from what the drafters had anticipated. Increasing numbers of persons traveling to the United States to claim asylum essentially turned the United States into a country of first asylum (as opposed to being only a country of overseas resettlement) and required lawmakers and adjudicators to grapple with the challenge of “making the individualized and fine-grained determinations of likely persecution” which the Refugee Act of 1980 required (Beyer 1992: 264–268). This case-by-case adjudication process signaled a sharp shift from the group-based admission system prevalent during the Cold War. Under the 1980 Act, a refugee is defined as any person who has a well-founded fear of persecution on the basis of “race, religion, nationality, membership of a particular social group or political opinion.” For the first time, U.S. adjudicators were charged with determining on a case-by-case basis which narratives of harm fit within one of the five enumerated persecution grounds. Moreover, the Act ordered the Attorney General to develop new procedures for asylum screenings for applicants already on U.S. territory. Prior to the Refugee Act of 1980 applicants could only apply from designated locations outside the United States.

Where possible, adjudicators relied on established political knowledge concerning the meaning and scope of the enumerated persecution grounds. Judge Schmidt explains that even though prior to 1980 Congress and the courts had never formally defined the terms “race,” “religion,” “nationality” and “political opinion,” their meanings were well established within U.S. political culture and garnered wide-political support.⁶ Specific fact patterns such as being persecuted for publicly voicing one's political opinion, holding a particular religious belief, or being harmed because of skin color or national/ethnic affiliation, often triggered use of ordained schemes without government representatives and adjudicators having to spend considerable time debating the meaning of the screening categories themselves. Interviews with asylum officers and immigration judges suggest that when applying the grounds of race, religion, nationality and political opinion, asylum officers, and immigration judges could more easily rely on institutional scripts which associated refugee status with state-sponsored persecution of religious and political minorities rooted in Cold War politics. In this respect, rather than signaling a sharp break from Cold War refugee policy, the new regime drew from, and to a certain extent preserved, understandings of refugee status formed during the Cold War.

Conversely, lawmakers had no institutional scripts upon which to rely when it came to the meaning of the legal category membership of a particular social group. The term was first introduced during the Conference of Plenipotentiaries on the Status of Refugees and Stateless persons (1951), which was convened for the purpose of completing the drafting of the Convention Relating to the Status of Refugees. Stating that it was “essential that the text be as clear as possible” the Swedish representative introduced an amendment to include a reference to “persons who might be persecuted owing to their membership of a particular social group…”⁷ At the time, no attempt was made, either by the Swedish representative or by any other state representative, to define the meaning and scope of the proposed category. The conference adopted the amendment by a vote of 14 to 0 with 8 abstentions (Jackson 1999). The legislative history of the 1980 Refugee Act similarly sheds little light on the social group aspect of the refugee definition.

Following the implementation of the Refugee Act of 1980, lawyers and refugee advocates started to make use of the open-endedness of the term to argue for the recognition of various types of new and previously unrecognized forms of harms. Debates over the intended scope of the “particular social group” category were in essence debates over the scope of asylum protection in the post-Cold War era. A close reading of policy documents and case law, supplemented by interviews with government officials, judges, and lawyers, suggests that during this moment of policy breach lawmakers drew on moral scripts from their broader institutional environment—namely, the concept of immutability—to define the meaning of “membership of a particular social group,” and to demarcate the boundary between categories of deserving and underserving asylum seekers.

5. Constructing Asylum Law in the Absence of Institutionalized Scripts: The Case of “Particular Social Group”

The first time the BIA set to define the meaning and scope of the category particular social group was in Matter of Acosta (1985). Acosta is one of the most cited decisions in asylum law, setting the shape of particular social group jurisprudence not only in the United States but also, in other asylum-receiving countries (Dauvergne and Millbank 2003; Hamlin 2012).⁸ The case involved a claim filed by a native of El Salvador who was persecuted by guerrillas on account of his profession as a taxi driver refusing to participate in guerrilla-sponsored work stoppages. At the time, Acosta's asylum claim was by no means unique. In the mid-1980s, increasing numbers of applicants fleeing civil war in Guatemala and El Salvador claimed asylum. Reports on the violence and killings in these countries problematized attempts to categorize Central American applicants as mere economic immigrants seeking to enhance personal gain. At the same time, their acceptance as potential asylees threatened to conflate the category of asylees with the category of undocumented (illegal) immigrants. It is against this background that the Board proceeded to determine whether Acosta was eligible for asylum.

The question before the Acosta court was whether the respondent's claim to have suffered persecution on account of his membership in a group comprised of “CO TAXI drivers and persons engaged in the transportation industry of El Salvador,” was cognizable for asylum purposes (Matter of Acosta 1985: 233). To answer this question the Board proceeded to define the meaning of “membership of a particular social group.” In the absence of a clear policy framework for organizing differences between asylees and other categories of immigrants, the Board proceeded to redefine the meaning of “deservingness” for asylum, and then used deservingness as a primary criterion for demarcating the scope of asylum status.

According to the Board, a definition of the category “particular social group” must first and foremost “preserve the concept that refuge is restricted to individuals who are either unable by their own actions, or as a matter of conscience should not be required to, avoid persecution” (Matter of Acosta 1985: 234). Emphasizing that “dissent or disagreement with the conditions in another country or a desire to experience greater economic advantage or personal freedom in the United States,” does not suffice for asylum, the Board held that an applicant is “worthy” of asylum if he or she is persecuted for an immutable attribute that the applicant cannot change or should not be required to change (Matter of Acosta 1985: 221–222). Accordingly, the Board interpreted the phrase “persecution on account of membership of a particular social group” to mean “persecution that is directed toward an individual who is a member of a group of persons all of whom share a common, immutable characteristic, i.e., a characteristic that is either beyond the power of the individual members of the group to change, or is so fundamental to their identities or consciences that it ought not be required to be changed” (Matter of Acosta 1985: 233).

The concept of immutability first appeared in the U.S. Supreme Court's equal protection jurisprudence Frontiero v. Richardson (1973) where a plurality of the Court reasoned: “Since sex, like race and national origin, is an immutable characteristic determined solely by the accident of birth, the imposition of special disabilities upon members of a particular sex because of their sex would seem to violate the basic concept of our system that legal burdens should bear some relationship to individual responsibility” (686). While the Board did not directly cite the Frontiero decision, the language it uses is reminiscent of this very logic. The logic behind the immutability standard is that certain traits such as race, sex and national origin are not blameworthy “on account of being accidents of birth.” This reflects the deeply embedded moral understanding in U.S. political culture that it is unjust to discriminate against someone because of a trait that bears no relationship to individual responsibility (Clarke 2015).

The Board rationalized its new emphasis on immutability on the well-established doctrine of ejusdem generis (“of the same kind”), stating that “membership of a particular social group” should be understood in light of the other four enumerated categories of “race,” “religion,” “nationality,” and “political opinion,” all of which describe “persecution aimed at an immutable characteristic” (Matter of Acosta 1985: 233). A critical reading of U.S. asylum policy suggests, however, that this explanation is partial at best. Analysis of reports of international refugee conventions dating back to the interwar period indicates that at the time deservingness for asylum was considered to be a function of an individual's political and religious motivation to flee. That is, the focus was not on the immutable nature of the trait on account of which an individual was persecuted, but rather on whether the motivation to flee was religious/political rather than economic. That the Board's use of the immutability standard signals a new development in U.S. asylum law is further indicated by the fact that the Board's definition of particular social group in Acosta diverged from prior definitions of the category, all of which emphasized shared background, habits, interests, values, and social status, as defining factors of particular social groups for asylum purposes (Goodwin-Gill 1983; Grahl-Madsen 1972).⁹ Referencing these sources, the Board held that they do not provide a workable definition of “membership of a particular social group” and proceeded to establish its own definition centered on the notion of immutability.

The moral intuition underlying the concept of immutability, that it is unjust to discriminate on the basis of traits that are not chosen and that one is helpless to change, was central to U.S. political and legal culture (Clarke 2015). The unstated moral assumption was that a person is deserving of state protection only when he or she is truly powerless to prevent the source of persecution. After the Cold War, Board members drew from their broader legal environment to recombine the concept of immutability with that of the political refugee. In the process, a new (moral) framework for determining who is deserving of state protection from persecution was created. Specifically, this conception of justice, centered on notions of individual responsibility, provided lawmakers with a means for negotiating between a burgeoning human rights politics that prioritized the individual's plight over state interests, and political interests in curbing illegal immigration and securing the national border: a person was now considered deserving of asylum only if persecuted for a trait that cannot be changed and for which he or she is not responsible.

The precise answer as to why lawmakers start drawing on the concept of immutability in the mid-1980s as a central criterion for defining asylum protection is beyond the scope of this article. My goal is rather to demonstrate how lawmakers in the post-Cold War period use the moral concept of immutability as a primary template for demarcating the scope of asylum protection. An examination of asylum policy in the decades following the Acosta decision suggests that starting in the second half of the 1980s, differences between eligible and ineligible applicants came to be phrased in moral terms: asylum seekers deserved protection, in contrast to other immigrants seeking legal membership, because they were targeted for traits immutable to their individual identity. The meaning of immutability in this context was, however, far from clear.

6. Immutability and the Construction of the Deserving Refugee

In 1985, “immutability” was still a rather vague concept in U.S. asylum law. In Acosta, the Board held that neither the characteristics of being a taxi driver or refusing to participate in guerrilla-sponsored work stoppages are immutable “because the members of the group could avoid the threats of the guerrillas either by changing jobs or by cooperating in work stoppages” (1985: 234). According to the Board, while “[i]t may be unfortunate that the respondent either would have to change his means of earning a living or cooperate with the guerrillas in order to avoid their threats…because the respondent's membership in a group of taxi drivers was something he had the power to change he has not shown that the conduct he feared was persecution on account of membership of a particular social group” (1985: 234). Following the Acosta decision, the Board denied protection to increasing numbers of Central Americans claiming persecution on account of their membership in a particular social group under the rationale that the characteristics for which they were targeted are not the type of factors that are immutable or fundamental to a person's individual identity or conscience. These characteristics included age, living environment, socio-economic status and former professional status (Matter of C-A-L-, BIA 1997; Matter of V-T-S-, BIA 1997; Sanchez-Trujillo v. INS, 9th Cir. 1986).

During this early period, it was, however, not yet clear what attributes did qualify as immutable in ways that warranted asylum protection. As I will show, gender became central to how immutability in the context of asylum was defined. Over a period of several decades, forms of gender-based violence targeting women, children and sexual minorities were incorporated into mainstream refugee scholarship and law as examples of persecution inflicted on account of a trait (e.g., gender) that is both immutable and fundamental to individual identity. In this respect, while negotiations over the meaning of worth are not particular to gender asylum, they provide an exemplar of the dynamics involved in the making of U.S. asylum law in the post-Cold War era.

Existing studies provide various answers to the question of why certain forms of gender-based violence were successfully incorporated as bases for asylum protection, perhaps more so than other types of harms. These include successful bipartisan mobilization and the creation of new legal protections for girls, women and sexual minorities from sexual violence. Scholars show how these legal protections may work in practice to legitimize the deportation and/or incarceration of nonwhite immigrant men by framing them as barbaric violators of women's rights (Bernstein 2008; Fassin 2013; Ticktin 2011), and to justify discriminatory U.S. international defense and diplomacy (Grewal 2005; McKinnon 2016). Building on this body of work, I ask not why gender-based violence (as opposed to other forms of harms), but rather how gender-based violence came to define what immutability in the context of asylum means, with implications for how the line between categories of deserving and undeserving asylees is drawn.

7. Gender Asylum

In the early 1990s, asylum advocates started drawing on an emerging feminist discourse which framed sexual violence as persecution inflicted on women on account of their immutable gender, to argue for the recognition of gender persecution as a basis for asylum under the category of membership in a particular social group.¹⁰ Gender-based persecution was traditionally not considered grounds for asylum. In the United States, the few claims involving gender persecution that were recognized as grounds for asylum were initially brought under the political opinion category (Bhabha 1996).

In 1992, the Women's Refugee Project, a joint project of Harvard Law School's Immigration and Refugee Clinic and Somerville Legal Services, was founded for the stated purpose of providing legal assistance to women asylum seekers. A premier goal identified in the work of the Women's Refugee Project was to establish a theory of gender-based violence as persecution on account of membership in a particular social group (Anker 1995; Goldberg 1993; Kelly 1993). The Women Refugee Project drew on a notion that started to gain traction within international feminist circles, that women are persecuted on account of their gender, a trait framed as both immutable and fundamental to individual identity (as opposed to being victims of social, religious, and cultural mores). Women's inability to change or escape their gender positioned them as vulnerable victims unable to avoid persecution at all cost.

Over the course of a few years, the Project became a major setting for lawmaking on gender asylum; first through its central participation in the development of the Considerations for Asylum Officers Adjudicating the Asylum Claims of Women (1995), and later in its engagement by the DOJ in the training of immigration judges, asylum officers, and supervisors on issues related to women's asylum claims. During this period, the number of law journal articles on the topic of gender asylum proliferated (Neacsu 2003). In 1993, an article published in the Georgetown Immigration Law Journal asserted that evaluation of gender-related cases under the social group category would yield more accurate results than their evaluation under the political opinion category. According to this view, “speaking out against and opposing the social structure of society is political; but women are not persecuted for their beliefs that they should be free from violence. Nor is violence directed at them for personal reasons. Rather it is because of their gender” (Bower 1993). This new framing of gender-based violence appeared in multiple prominent law journals with legal scholars increasingly advocating for a reconceptualization of women asylum claims as victims of their own (immutable) gender (Anker 1995; Bunch 1990; Kelly 1993).

In the mid-1990s, in the case Matter of Kasinga (1996), the administration drew on this very framing of gender persecution to define immutability for asylum purposes.¹¹ The case involved a 19-year old native and citizen of Togo by the name of Fauziya Kassindja, who testified that she fled coerced female genital cutting and forced marriage. Kassindja was denied asylum by an immigration judge who claimed that female genital cutting is part of a tribal culture and thus does not constitute persecution under asylum law. Her case, however, caught the attention of Karen Musalo, the acting head of the Human Rights Clinic at American University, who agreed to handle the appeal pro bono. Shortly after becoming involved in the case, Musalo turned to various human rights and feminist organizations. Among them was Surita Sandosham, the executive director of Equality Now, a women's rights organization and a leading player in the emerging global anti-female genital mutilation (FGM) campaign.¹² In the coming months, there was a steady progression of positive media in favor of Kassindja who was largely portrayed as a victim of a horrific practice aimed at subordinating women (Martin 2005). This in turn, sparked a public debate over which harms ought to be protected by the institution of asylum, and whether or not female genital mutilation should garner asylum protection.¹³

In 1996, the BIA set to formally resolve this debate. In a nearly unanimous decision, it held that genital mutilation constitutes precisely the type of harm that deserves asylum protection, granting Kassindja asylum. In its ruling, the Board cited Nahid Toubia, an advisor to the World Health Organization and director of the Global Action Against FGM Project at Columbia University School of Health, and various other human rights health organizations, to establish that female genital mutilation is not merely a cultural practice but rather a form of oppression aimed at violating and exploiting a woman's sexuality. Drawing on advocate briefs, media representations, and established academic knowledge in the fields of health and anthropology, the Board held that the “characteristic of having intact genitalia is one that is so fundamental to the individual identity of a young woman that she should not be required to change it (1996: 367).” It is on this basis, concluded the Board, that female genital mutilation constitutes persecution on account of membership of a particular social group.

What came to be of great importance to the future framing of immutability in the asylum context, was the Board's decision to focus solely on genital cutting, deliberately overlooking Kassindja's alternate claim of forced marriage (Dauvergne and Millbank 2003). Through its focus on female genital cutting, the Board generated a particular understanding of immutability associated with bodily (sexual) traits framed as innate and fundamental to individual personhood. Exford (2005) shows how this framing came to be institutionalized within asylum jurisprudence such that a range of service providers such as doctors, psychologists and lawyers, in addition to legislatures and judges, actively encourage female applicants to pursue FGM claims, and often fail to inquire about other forms of gendered harm.¹⁴ In turn, a new hierarchy of harms was created according to which violation of one's (sexual) body and identity gain precedence over harms targeting persons on account of socially acquired characteristics. This hierarchy of harms is grounded in a particular understanding of worthiness according to which persons discriminated against on account of innate bodily traits are more deserving of protection than those persecuted for social characteristics they have the power to change or avoid.

Significant in this regard was the administration's support of Kassindja's claim. Judge Schmidt, chairman of the Board in 1996 and the author of the majority opinion in Matter of Kasinga, explains that “Kasinga wouldn't have come down the way it did but for the fact that David Martin [the General Counsel for the INS] basically stripped the case of controversy…and David Martin wouldn't have argued that position unless [Attorney General] Janet Reno agreed with it.”¹⁵ The administration's acceptance of Kassindja's claim is perhaps less surprising considering its unequivocal acceptance of sexual orientation as a prototypical example of immutability for asylum purposes just a few years prior. In 1990, the Board, in accordance to the opinion of the Immigration and Naturalization Service, held that sexual orientation constitutes an immutable trait and a basis for a particular social group (In re Toboso-Alfonso, BIA 1990).¹⁶ In its decision, the Board cited social and behavioral scientists who stated that sexual orientation is an immutable characteristic that is highly resistant to change. On June 19, 1994, then-attorney general Janet Reno, designated the decision on Toboso-Alfonso to be precedent in all proceedings involving the same or similar issues.¹⁷

Accordingly, far from being a pure legal exercise, the Kasinga decision was orchestrated to make a broader statement about the types of harms that deserve protection under the open-ended rubric of membership of a particular social group. The decision was a highly publicized event and the Board had to make unprecedented arrangements for media presence in the courtroom and for a bank of cameras outside the building to film comments of the parties immediately after the argument. The case was covered in all of the major media outlets including the New York Times, the Nation, the Washington Post, the Los Angeles Times and USA Today. Ted Koppel devoted his entire Nightline show on the evening after the Board hearing to an interview with Kasinga, a graphic background piece of FGM, and a review of the legal arguments in the case. Kasinga also appeared on CNN, CBS and was interviewed by dozens of newspapers (Martin 2005).

Following Kassindja's victory, Karen Musalo founded the Center for Gender and Refugee Studies at Hastings University, the first center devoted to providing national guidance on how to adjudicate claims involving gender-based harms from women. According to Lisa Frydman, an attorney at the Center:

Kasinga was the catalyst for gender asylum. This is not to say that no one ever thought about it but Kasinga was really a catalyst. I mean, Kasinga is why this center started. After the Kasinga case, Karen was contacted by tons of attorneys saying: “now I am going to file a gender case. What shall I do?” and this is why the center came up.¹⁸

In future cases, activists used the framing proposed in the Kasinga decision as a template for defining persecution on account of “membership of a particular social group.” Immutability, now defined in relation to female genital mutilation, became the basis of a newly emerging moral boundary between categories of deserving and undeserving asylum seekers. In subsequent years, legal actors negotiated and reconstrued this moral boundary in relation to new claims, shedding light on how political anxieties, values, and interests intersected with moral categorizations to redefine asylum law.

8. Domestic Violence: Renegotiating Immutability

One of the most contested questions related to gender asylum was whether the framework presented in Kasinga is applicable to other forms of gender violence, most notably domestic violence. A couple of months subsequent to the Kasinga decision, Immigration Judge Mimi Schooley Yam granted Rodi Adalí Alvarado-Peña's request for asylum. Alvarado-Peña, a citizen and native of Guatemala, experienced extreme abuse at the hands of her husband. Holding that Alvarado-Peña was credible, the immigration judge ruled that Alvarado-Peña was persecuted because she is a woman. Specifically, the judge drew a comparison between female genital cutting and domestic violence and held that Alvarado-Peña was eligible for asylum because her gender and prior association with her husband are immutable characteristics she could not, or should not, be expected to change (see Matter of R-A- 1999: 911). The Immigration and Naturalization Service appealed the immigration judge's grant of asylum and in June of 1999, the Board reversed the immigration judge's decision, rejecting the assertion that the framing used in Matter of Kasinga for female genital cutting can be extended to domestic violence. According to the Board, whereas female genital mutilation is motivated by a quest for male domination and the subjugation of female sexuality, domestic abuse is rooted in “such things as the Latin American patriarchal culture, the militaristic and violent nature of societies undergoing civil war, alcoholism and sexual abuse in general” (1999: 910). Significantly, the debate over whether victims of domestic violence are deserving of asylum centered on the question of immutability.

The Board's decision sparked a national debate over whether domestic violence should be considered a basis for asylum. Karen Musalo became sole counsel for the case and domestic violence became the heart of CGRS's work on gender asylum policy reform. During the next 2 years, dozens of members of the House of Representatives and the Senate, in addition to over 51 organizations and 49 law scholars and advocates, sent letters to Attorney General Janet Reno requesting that she reverse the Board's decision.¹⁹ The general argument made was that Alvarado-Peña was persecuted on account of her gender and that gender “is both immutable and fundamental to an individual's identity” (Brief of Amici Curiae, p. 16). This argument was shared by a remarkably diverse coalition of social activists and policy makers, including abolitionists, feminists, evangelical Christians, and both conservative and liberal government officials, formed in support of Alvarado's grant of asylum.²⁰ Despite their conflicting ideas about sexuality and gender, conservative-minded organizations came to work together with liberal feminist groups and human rights activists to promote the idea that domestic violence is a crime inflicted on women on account of their (immutable) gender and that on this basis women are deserving for asylum. While there was considerable controversy over whether domestic violence is inflicted on women on account of their gender or rather a private criminal act motivated by societal factors, the notion that to be deserving of asylum protection one has to be persecuted on account of an immutable trait, was not once questioned by social activists, adjudicators and state officials.

In December 2000, the DOJ published proposed regulations on the definitions of “persecution” and “membership in a particular social group,” in an attempt to provide an analytical framework for gender-related asylum claims (Proposed Rules December 7, 2000). The proposed rules specifically noted that the Board was wrong to conclude that Alvarado-Peña was not persecuted on account of her “membership in a particular social group.” According to the proposed rules, where the victim of domestic violence can establish that “the abuser is motivated to harm her because of her gender or because of her status in a domestic relationship,” she can establish that she was persecuted on account of her membership in a particular social group since “marital status could be considered immutable” (Proposed Rules 2000: 76593). On January 19, 2001, Attorney General Janet Reno vacated the Board's decision and ordered the Board to issue a new decision in Alvarado-Peña's case after the DOJ's issuance of proposed rules on the subject of gender asylum.

The proposed rules were never finalized, and in 2003, Attorney General Ashcroft, appointed by the Bush Administration, referred Alvarado-Peña's case back to himself for review and provided an opportunity for additional briefing. In a fascinating move, the DHS provided an analytical framework of its own for adjudicating claims of domestic violence. Consistent with the analytical framework laid out in the proposed rules, DHS contended that the particular social group in Alvarado-Peña's case should be defined by the respondent's (immutable) status within the domestic relationship. According to DHS, marital status may be considered a fundamental attribute of one's identity and in some circumstances one that would be impossible to change (DHS R-A-Brief 2004: 20).

In response to the supporting DHS brief and growing public pressure, in 2005 Attorney General Ashcroft remanded Matter of R-A- back to the Board, directing it to reconsider its decision once the proposed rule was finalized. In 2008, Attorney General Michael Mukasey certified Matter of R-A- to himself, with order not to wait for the finalization of the regulations. In 2009, the DHS submitted a brief in a separate case involving a claim of domestic violence by a Mexican woman, L-R-, in which it reaffirmed the framing proposed in the 2004 DHS Brief (DHS L-R-Brief 2009). That same year, in December 2009, Alvarado-Peña was at last granted asylum. In 2014, the Board issued its first precedent decision in which it held that in certain instances domestic violence can constitute a basis for membership in a particular social group since “marital status can constitute an immutable characteristic where the individual is unable to leave the relationship” (Matter of A-R-C-G, 2014: 393). In a subsequent unpublished case from 2015, the Board broadened its 2014 holding to also include domestic abuse among nonmarried couples.²¹ The Asylum Officer Training Guidelines on Membership of a Particular Social Group similarly note that abuse serious enough to amount to persecution can also occur within other domestic relationships that are not spousal like.²² Accordingly, over a period of more than 10 years, consecutive administrations debated how to approach asylum claims based on domestic violence. In the process they reaffirmed a new moral boundary centered on innate traits defined by gender and sexuality.

9. Discussion and Conclusion

This study aims to show how in the process of defining the meaning of “membership of a particular social group,” agency officials, judges and lawyers actively renegotiated the boundaries of refugee deservingness. Political and foreign policy interests, geopolitical changes, and an emerging coalition of progressives and conservatives mobilized against gender violence, all shaped the way immutability in the context of asylum came to be framed. Whereas the original meaning of immutability, rooted in equal protection jurisprudence, referred to traits beyond one's control to change, when it came to asylum, the administration rejected a literal interpretation of immutability in favor of one that emphasizes the degree to which a trait is fundamental to individual identity and bodily integrity. This notion of immutability, I argue, served as a moral marker for differentiating between deserving asylum seekers and undeserving undocumented immigrants—two categories which in the post-Cold War period were under increasing threat of being conflated. Following the collapse of the Cold War policy framework whereby communism was the primary criterion for distinguishing refugees from other categories of immigrants, and in lack of an alternative framework, distinctions of worth centered on the moral concept of immutability provided a template by which to recreate the asylum seeker as legally and conceptually distinct from other immigrant categories. Over the years, various qualifying markers such as particularity and distinction were added to the definition of “particular social group,” yet these failed to clarify the meaning of the law or to ensure its consistent application.²³

The centering of the “particular social group” standard on the moral conception that to be deserving of asylum you have to be persecuted for traits fundamental to your body and identity has shaped who can get asylum. Analysis of administrative data obtained on affirmative asylum applications processed by the asylum office between 1995 and 2015, suggests that in the last decade, gender-related claims pursued under the particular social group category constitute the primary means for gaining asylum from certain Western African countries including The Gambia, Guinea, and Mali. In all three countries, the majority of asylum claims are framed as “persecution on account of membership in a particular social group.” The fact that in all three countries the vast majority (90%) of all particular social group applications granted are filed by women, strongly suggests that in a majority of cases, these applications involve claims based on female genital cutting and other gender-related harms.²⁴

Despite the broadening of the “particular social group” category to include a range of previously unrecognized harms, the growth of gender asylum is far from a story of unidirectional progressive expansion. At the same time that immigration and government authorities issued provisions protecting women from a whole range of gender-based harms, they also enacted laws restricting the entry of refugees from politicaland religious-based persecution. Central American immigrants have seen their protection options contract while female victims of sexual violence have received protections unavailable a couple of decades ago (Fassin 2013). In 1996, the same year that the government expanded asylum law to include female genital mutilation as a basis for asylum, Congress passed the Illegal Immigration Reform and Immigration Responsibility Act, which placed many new restrictions on immigrants and on refugees in particular (Public Law 104-208, 1980).

The development of asylum policy is thus indicative not of a move toward greater inclusion but rather of a reconfiguration in the parameters of worth. In the early 2000s, the numbers of applicants from Mexico and the Northern Triangle countries in Central America started to significantly increase.²⁵ The majority of applicants from these countries escape persecution by gangs and present claims that largely fall outside the scope of conventional forms of government sponsored persecution. According to Leon Rodriguez, former DHS director (2014–2017), the sharp increase of applicants from Mexico and the Northern Triangle countries seeking asylum on the basis of gang violence imposed not only an administrative strain on the system, but also a “conceptual strain” due to the nontraditional nature of a majority of these claims.²⁶ Reports issued by the State Department, by human rights and advocacy organizations, and by major news outlets, documented the escalating violence, the unprecedented number of civilian killings, and the government's inability to grant effective protection to its citizens.

The majority of claims from Central America, however, are unsuccessful. Fears of floodgates, racism, xenophobic domestic politics, and foreign policy interests, all led consecutive administrations to adopt a hostile approach toward persons fleeing gang-violence in Central America and Mexico (Frydman and Healy 2017). Analysis of case law and policy directives involving gang-related claims from Central America suggests that immutability provides a conceptual template by which to frame Central American applicants' ineligibility. Starting in 2007, the Board issued a series of decisions in which it concluded that applicants persecuted by gangs on account of attributes such as socioeconomic status, former profession, and level of education did not deserve protection under the Act because they failed to show that they were persecuted on account of immutable traits. For example, the DHS training guidelines on how to apply the category “membership of a particular social group” state that groups persecuted on account of their wealth or socioeconomic status are not eligible for protection “because wealth is not immutable,” (Matter of A-M-E- & J-G-U- 2007), just as groups defined by professional status are not protected because this [owning a small business] is not “the kind of group characteristic that a person either cannot change or should not be required to change.”²⁷

While conventional categories of persecution rooted in the interwar period based on political opinion or religion continue to shape contemporary asylum policy, a breach in the prevailing communist-centered framework created situations where policy makers had to apply asylum law without being able to default to institutionalized policy prescriptions. Starting in the 1980s, policy makers renegotiated moral distinctions between categories of deserving and undeserving refugees to bring content to ambiguous law. Immutability, as it came to be defined in the asylum context, provided a conceptual template for reorganizing differences between “deserving” and “undeserving” asylum seekers: women and sexual minorities are worthy of asylum protection because they are targeted for traits perceived to be immutable and fundamental to their body and individual identity. Conversely, Central Americans fleeing the civil wars in the 1980s, and gang violence in the 2000s, are constructed as unworthy precisely because they are targeted for mutable traits framed as not fundamental to their personhood. In turn, a hierarchy of harms warranting asylum protection is created whereby persecution inflicted on account of sexual and gendered characteristics takes precedence over harms inflicted upon socioeconomic and environmental-related traits.

Asylum constitutes a core arena through which state agents demarcate the boundaries of legal membership and citizenship. Examining the process through which the boundaries of asylum deservingness are negotiated advances not only our understanding of U.S. asylum policy but also of the processes through which groups draw on moral boundaries to construct notions of “us” and “them” (Bail 2008). My findings suggest that categories of moral worth play an important role in how social actors bridge emerging gaps between abstract rules and complex cases during periods characterized by a policy breach, where actors can no longer rely on preexisting policy prescriptions to chart specific courses of policy action (Campbell 2002). After the Cold War, judges, administrators and legal advocates engaged in a process of moral bricolage; they drew on moral scripts that they found in their broader institutional environment to redefine the meaning and scope of ambiguous law; categories of moral worth served as primary criteria for classifying complex persons and cases. In this respect, while political interests, power relations, and local culture, influence what groups will be placed on which side of the binary, shared distinctions of worth shape how groups' eligibility for asylum will be framed. In the process, the very meaning of worth is redefined in ways that have unanticipated policy implications.

On June 11, 2018, then Attorney General Jefferson Beauregard Sessions, issued a decision involving a claim of domestic abuse filed by a Salvadoran applicant by the name of Ms. A.B., in which he asserted that the “particular social group” category would no longer be used as a basis for protection for victims of domestic violence and gang violence perpetrated by nongovernmental actors (Matter of A-B- 2018). The decision is expected to have significant implications for thousands of immigrants worldwide and from Central America in particular, as claims involving domestic violence and gang violence constitute the majority of applications from Mexico, El Salvador, Guatemala, and Honduras in recent years.²⁸ The significance of the Attorney General's decision, however, is not only in the expected detrimental effects on the lives of those seeking asylum, but also in what it implies about how the moral boundaries of the institution of asylum are drawn. In Matter of A-B-, the attorney general set out to recast the meaning of immutability for purposes of defining the category “particular social group.” After asserting that “the asylum status does not provide redress for all misfortune,” Sessions held that the immutability standard was misapplied in domestic violence cases, since domestic violence is a private criminal activity (2018: 318). The attorney general's decision has led to considerable controversy, with 11 amicus briefs filed in support of Ms. A.B.²⁹ It is perhaps too early to determine whether Matter of A-B- will lead to a contestation of the moral boundary defining contemporary U.S. asylum. However, in order to understand how this decision and others like it will influence the future development of asylum policy we have to pay close attention to the ways anxieties about illegal immigrants and political interests are mediated through embedded distinctions of worth.