Merging Sociology and Law: How Social Construction of Identity May Explain Differing Treatment of Women and Gay People as Social Groups in Refugee Law

By Rachel Barr 

Third-year student at the Univeristy of Michigan Law School.

The Refugee Convention only recognizes refugee status when persecution is based on race, nationality, religion, political opinion, or membership of a particular social group.[1] Gender and sexual orientation—common grounds for discrimination—are categorized under membership of a particular social group. Courts, however, often narrowly define groups of women as a social group while not doing the same for people who are gay. This paper explores possible reasons and explanations for why this difference in definition occurs. It concludes that the tendency of courts to define gender groups specifically and sexual orientation groups broadly, can plausibly be explained by the social construction of these different identities.

In re Kasinga, a seminal case in defining women as a social group, considers the application of a 19-year-old from Togo who faced the threat of female genital mutilation (FGM) and sought asylum in the United States on the basis that she would be forced to undergo FGM if she returned home. [2] The Board of Immigration Appeals found FGM to be a form of persecution, but held the particular social group to be very specific: “young women of the Tchamba-Kunsuntu Tribe who have not had FGM, as practiced by that tribe, and who oppose the practice.”[3]

Kasinga is one of many cases where gender discrimination is the underlying cause of persecution, but the court has not found persecution on the basis of gender alone.[4] Instead, courts tend to define social groups as gender plus other characteristics of the group, like certain behavior or marital status. Examples of this over specifying include Sarhan v. Holder (“women in Jordan who have (allegedly) flouted repressive moral norms, and thus who face a high risk of honor killing”)[5] and SVTB v Minister of Immigration & Multicultural and Indigenous Affairs (“single Albanian women who do not have the protection of male relatives”).[6] While in some cases courts have broadly recognized women as the social group upon which persecution is based, this is not the international norm.[7]

The same cannot be said for cases dealing with persecution on the basis of sexual orientation. In many cases, courts have found gay people to be a social group without including other situational characteristics of the persecuted individual into the social group category.[8] In doing so, courts explain that sexual orientation “is either an innate and unchangeable characteristic, or it is a characteristic that is so fundamental that [one] should not be required to change it.”[9] Little explanation about the definition of the social group, beyond that, is included; it is clear that the issue of gay people as a social group is settled.[10]

The big question is why. Why is there a disparity in the treatment of women and people who are gay? Why do courts easily recognize gay people as a persecuted social group, while courts struggle with defining a social group when persecution is based on gender?

It is first necessary to look at how courts explain their decision to narrowly define gender groups. Courts argue that because the evidence presented about persecution is specific and tailored to the situation confronted by the applicants, the evidence does not prove gender discrimination broadly, but persecution for a narrower reason.[11] There are two key problems with this explanation. First, a lack of evidence does not explain the result in cases like Kasinga as “[t]he record in Ms. Kasinga’s case documented extensive discrimination against women in Togo.”[12] Additionally, this argument reflects a misunderstanding of refugee status determination; courts are expected to define social groups by first determining the group and then identifying the risk the group faces.[13] Courts attempting to define the group based on circumstances and evidence of persecution, are misapplying the refugee definition. Moreover, a lack of evidence is an impermissible explanation as the onus is on the adjudicator to share the burden of gathering evidence necessary to make an accurate determination of refugee status.[14]

Another explanation given for defining gender groups narrowly is the size of the groups at issue. Foster argues that “a more narrow formulation [of social groups defined in part by gender] appears to be based on an implicit floodgates concern,” a concern that defining the group broadly will “suggest that all persons in such a group [e.g. all women in a country] are at risk” and, therefore, lead to an influx of refugees.[15] As a smaller group, defining a social group on the basis of being gay does not raise the same floodgate concerns. This, however, is a disingenuous argument as the other permissible grounds on which persecution can be based are large as well. For example, persecution can be found on religious grounds, and there are 1.6 billion people in the world who identify as Muslim.[16] Additionally, “race, nationality…and even political opinion are also traits which are shared by large numbers of people.”[17]

Foster also argues that the increasing focus on the social perception of the group might explain the difference.[18] The application of the social perception test “often manifests in an (apparently) arbitrary finding that being a woman, even…where gender discrimination persists, is not ‘sufficiently distinguishing’ or that a category such as ‘young single women’ may not be ‘united by a common characteristic’, or ‘singled out’ from society.”[19] The use of this test, however, does not explain the difference in treatment as gay individuals can be an even less socially-prevalent group. People who are gay may be private about their sexual orientation, and, as a result of the application of the social perception test, may not be recognized as a distinct social group. For example, courts in France have denied refugee status when gay applicants were not open about their sexuality.[20] While this potential trend toward a more stringent standard for defining any social group is deeply troubling as it raises the bar for groups making claims based on gender or sexual orientation, it does not answer the question of why these groups are treated differently.

The preceding arguments put forth by courts and scholars do not provide a compelling explanation for the differing treatment of women and gay individuals as social groups. Other academic disciplines, however, may shed light on the different treatment. Sociology holds that identity groups like gender and sexual orientation are socially constructed.[21] The construction of these identities provides a more persuasive explanation for the different treatment of women and gay people as social groups in refugee proceedings.[22]

Societal expectations of gender expression arise from societal gender roles of women as wives and mothers.[23] In essence, women are defined by their husbands and children. This is exemplified by western marriage rituals, including the bride being “given away,” the couple being pronounced “man and wife,” and the bride taking her husband’s last name; these traditions emphasize that women do not have identities separate from their husbands.[24] Women are also defined by their children; “[h]egemonic gender norms dictate that mothers are supposed to be the primary caregivers and nurturers within the family.”[25] This is evidenced by societal celebrations. Studies show that “mothers receive more attention on Mother’s Day than fathers do on Father’s Day. Families not only spent more time celebrating Mother’s Day than Father’s Day, but mothers were also more likely to receive gifts, celebrate with others, and go out.”[26] It is suggested that “Mother’s Day may be more recognized because of the greater importance placed on motherhood than on fatherhood in society or because of the centrality of motherhood to women’s identity.”[27]

It is true that gender roles are changing, but instead of completely redefining gender roles, society now expects women to be “superwom[e]n,” not only fulfilling their traditional gender roles but also embodying the role of the modern career women.[28] Despite progress, traditional gender roles continue to influence the social construction of gender in western society.

People who are gay, however, are not burdened or defined by gender norms in the same way, as the very notion of being gay challenges traditional societal gender roles. “Society simultaneously socializes everyone to adhere to traditional gender role norms and teaches gay people that it assumes [they] will violate those norms.”[29] In other words, it is assumed that people who are gay “are likely to display gendered behavior that is inconsistent with their biological sex.”[30] This is not to say that gay individuals are never viewed in light of traditional gender roles. [31] “In general, people who engage in role behaviors associated with the other gender or who possess characteristics associated with the other gender are not viewed positively [particularly “among people who hold traditional gender-role beliefs”].”[32] The lack of adherence to traditional gender roles is what society sees as core to gay identity.

This inconsistency with traditional gender norms has been reinforced by the fact that people who are gay were not able to marry in many countries until recently.[33] And, in contrast to heterosexual marriages, gay weddings often do not have the same heteronormative influences.[34] Further, since gay couples are unable to biologically reproduce together, they have been typecast as being “unfit parents.” [35] As a result, their identities are not constructed such that they are subsumed by their partners or children.

The core differences in the social construction of gender and sexual orientation may explain the differing treatment of women and gay people in refugee law. Women are not seen as distinct individuals, while gay individuals’ identities are not merged with a spouse or children. This distinction in society’s perception of female and gay identity could explain the, perhaps implicit, reluctance of courts to recognize women, in and of themselves, as a social group.

While this may explain the reason, this practice continues to be inappropriate and disingenuous to the non-discrimination values of refugee law. It puts women in a “[c]atch 22,” forcing female asylum seekers to jump through hoops and strike the right balance of generality and specificity in order for a court to acknowledge persecution on account of gender.[36] Courts, instead should follow the ejusdem generis approach when dealing with persecution based on gender, as it would find women in a particular country to be an appropriately defined social group. [37]

Coming back to In re Kasinga, the court should have defined the social group as “women in the Tchamba-Kunsuntu tribe” (in order to take account of the gender and ethnic bases for the persecution) rather than “young women of the Tchamba-Kunsuntu Tribe who have not had FGM, as practiced by that tribe, and who oppose the practice.”[38] FGM at its core is a practice rooted in gender discrimination, which is carried out against women because they are women, period.[39] The definition of social group should allow women, like gay people, to be a distinct social group and should acknowledge that women are, at times, persecuted for no other reason than their gender.


[1] Convention Relating to the Status of Refugees art. 1, July 28, 1951 189 U.N.T.S. 137.

[2] In re Kasinga, 21 I. & N. Dec. 357, 357-359 (USBIA 1996).

[3] Id. at 368.

[4] See, e.g. Sarhan v. Holder, 658 F.3d 649 (7th Cir. 2011); SVTB v. Minister for Immigration & Multicultural and Indigenous Affairs, [2005] FCAFC 104 (3 June 2005) (Austl.).

[5] Sarhan, 658 F.3d at 655.

[6] SVTB, [2005] FCAFC 104, ¶¶ 4, 17.

[7] Fornah v. Secretary of State for the Home Department [2006] UKHL 46, ¶ 31, per Lord Bingham and ¶ 54 per Lord Craighead (U.K.); Minister for Immigration and Multicultural Affairs v. Khawar [2002] HCA 14, ¶ 32, per Gleeson CJ (Austl.); James C. Hathaway & Michelle Foster, The Law of Refugee Status, 439-440 (2d ed. 2014).

[8] SB v. Secretary of State for the Home Department [2010] EWHC 338 (Admin), ¶ 2 (U.K.); Karouni v. Gonzales, 399 F.3d 1163, 1172 (9th Cir. 2005); HS (Homosexuals: Minors, Risk on Return) Iran [2005] UKAIT 00120, ¶ 146 (U.K.); Refugee Appeal No. 74665, New Zealand: Refugee Status Appeals Authority [2004], ¶ 131 (N.Z.); Hathaway & Foster, supra note 7, at 443-444.

[9] HS [2005] UKAIT 00120, ¶ 146.

[10] SB [2010] EWHC 228 (Admin), ¶ 2; Karouni, 399 F.3d at 1172; HS [2005] UKAIT 00120, ¶ 146; Refugee Appeal No. 74665, ¶ 131.

[11] Fejza v. United States AG, 489 Fed. Appx. 326, 329-330 (11th Cir. 2012); Faye v. Holder, 580 F.3d 37, 41-42 (1st Cir. 2009); Rreshpja v. Gonzales, 420 F.3d 551, 555-56 (6th Cir. 2005); Fatin v. INS, 12 F.3d 1233, 1241-42 (3rd Cir. 1993); see also Khawar [2002] HCA 14, ¶ 129, per Kirby J.

[12] Karen Musalo, Revisiting Social Group and Nexus in Gender Asylum Claims: A Unifying Rationale for Evolving Jurisprudence, 52 DePaul L. Rev. 777, 799 (2003).

[13] Applicant A v. Minister for Immigration & Ethnic Affairs [1997] 190 CLR 225 (Aus. HC), 242, per Dawson J (Austl.) (citing Chan v. Canada [1993] 3 FC 675, 692-93 per Heald JA (Can.)).

[14] Hathaway & Foster, supra note 7, at 118-121.

[15] Michelle Foster, The ‘Ground with the Least Clarity’: A Comparative Study of Jurisprudential Developments relating to ‘Membership of a Particular Social Group’ 44 (2003),

[16] Drew DeSilver and David Masci, World’s Muslim Population More Widespread that You Might Think, Pew Research Center (Jan. 31, 2017), (last visited March 29, 2018).

[17] Hathaway & Foster, supra note 7, at 438.

[18] Foster, supra note 15, at 46-47 (footnotes omitted).

[19] Id. at 46.

[20] Id. at 52 (citing Mlle F, Cour Nationale du Droit d’Asile (CNDA) [French National Court of Asylum], 513547, 25 March 2005 [Adrienne Anderson trans]).

[21] Barbara Ryan, Sex and Gender, in Blackwell Encyclopedia of Sociology Online. (George Ritzer ed. 2007), (last visited Mar. 29, 2018); William Simon, Sexual Orientation, in 4 Encyclopedia of Sociology, 2564-2575 (2nd ed. 2001).

[22] The following represents an overview of these complex issues, and is in no way comprehensive.

[23] Janet Kohen, Housewives, Breadwinners, Mothers and Family Heads: the Changing Family Roles of Women, 8 Advances in Consumer Res. 576, 576 (1981); Beth Montemurro, Add Men, Don’t Stir: Reproducing Traditional Gender Roles in Modern Wedding Showers, 34 J. Contemp. Ethnography 6, 9 (2005).

[24] Robert Barr and Jill Bley, Turning Marriage Right Side Up: Lilith on Top, Huff Po (Nov. 20, 2012) (last visited Dec. 7, 2017).

[25] Nicole Gilbert Cote and Francine M. Deutsch, Flowers for Mom, a Tie for Dad: How Gender is Created on Mother’s and Father’s Day, 25 Gender Issues 215, 216 (2008).

[26] Id. at 224.

[27] Id.

[28]  Monika K. Sumra and Michael A. Schillaci, Stress and the Multiple-Role Woman: Taking a Closer Look at the “Superwoman”, 10 PLOS One 4 (2015)

[29] Marc A. Fajer, Can Two Real Men Eat Quiche Together – Storytelling, Gender-Role Stereotypes, and Legal Protection for Lesbians and Gay Men, 46 U. Miami L. Rev. 511, 614 (1992).

[30] Diane Felmlee et. al., Fairy Tales: Attraction and Stereotypes in Same-Gender Relationships, 62 Sex Roles 226, 227 (2010)

[31] See K. Brown, The Gendered Elements of Homosexual Marriage and Society’s Reaction to the Issue, 1 ANU Undergraduate Res. J. 79, 80 (2009)

[32] Bernard E. Whitley, Gender-Role Variables and Attitudes Toward Homosexuality, 45 Sex Roles 691, 692 (2002).

[33] Gay Marriage Around the World, Pew Research Center, (Aug. 8, 2017) (last visited Mar. 29, 2018).

[34] Áine M. Humble, Same-Sex Weddings, 3 The SAGE Encyclopedia of LGBTQ Studies 977-980 (Abbie E. Goldberg ed. 2016).

35 Brown, supra note 41, at 82.

[36] Fornah [2006] UKHL 46, ¶ 113, per Baroness Hale.

[37] R v. Immigration Appeal Tribunal and Another; Ex parte Shah, [1999] UKHL, per Lord Steyn (U.K.); Matter of Acosta, 19 I. & N. Dec. 211, 233 (USBIA 1985).

[38] In re Kasinga, 21 I. & N. Dec. at 368.

[39] Eliminating Female Genital Mutilation: An Interagency Statement 1 (World Health Organization 2008); Rajat Khosla et. al., Gender Equality and Human Rights Approaches to Female Genital Mutilation: A Review of International Human Rights Norms and Standards, 14 Reproductive Health 3 (2017)


Submit A Comment

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>