2 ‘Brown’ feminism, decoloniality and international refugee law

International refugee law has traditionally been framed around the experiences of men, thereby rendering women invisible (Dauvergne, Reference Dauvergne, Costello, Foster and McAdam2021). The 1951 Refugee Convention and its 1967 Protocol,Footnote ³ which form the core of the international refugee legal framework, define a refugee as someone outside their country of nationality or habitual residence and who:

The use of the pronouns ‘his’ and ‘himself’ in this supposedly ‘universal’ definition reflects the deeply male-centric perspective from which international refugee law has been written and subsequently interpreted, as the case law analysis undertaken below demonstrates. Anyone wishing to access protection has to prove a well-founded fear of persecution for one of the five aforementioned Convention grounds, which are race, religion, nationality, membership of a particular social group (PSG) and political opinion. Sex or gender were not included among the grounds of persecution. Consequently, women persecuted on account of their gender were typically denied international protection (Greatbatch, Reference Greatbatch1989). Significantly, there was not a single woman among the drafters of the 1951 Refugee Convention (Johnsson, Reference Johnsson1989). During the negotiations, when Yugoslavia made a proposition to include sex as a ground of persecution, the British delegate replied that ‘equality of the sexes was a matter for national legislation’ (Spijkerboer, Reference Spijkerboer2000, p. 1).

As a result, refugee status determination (RSD) has been largely guided by the refugee male prototype (Crawley, Reference Crawley, Burson and Cantor2016). Until the 1990s, the experiences of refugee women were largely unrecognised (Greabatch, Reference Greatbatch1989). RSD was presented as a neutral process based on a universal definition when, in fact, it was largely centred and framed around the experiences of men. Gender-specific or gender-based forms of violence, such as rape or female genital mutilation, were not recognised as relevant forms of persecution (Spijkerboer, Reference Spijkerboer2000). Ahigh-profile case illustrating the patriarchal nature of refugee law concerned a man who was tied to a chair and forced to watch his wife being raped by soldiers (Pittaway and Bartomolei, Reference Pittaway and Bartolomei2001). When he lodged a refugee claim, he was acknowledged as having been subjected to torture, while his wife was not (ibid.).Footnote ⁴ Traditionally, refugee claims submitted by women fleeing gender-specific or gender-based violence were generally rejected on the basis that they fell under ‘the private sphere’ and did not amount to persecution in the sense of the Convention (MacKinnon, Reference MacKinnon1983). For instance, in Campos-Guardado, a US Court rejected the asylum claim of a Salvadorian woman who had been raped by abusers, while they shouted political slogans, on the basis, again, that rape should be considered a private act that did not amount to persecution.Footnote ⁵ The fact that the treatment was dispensed by non-state actors reinforced the perception that it fell under the private rather than the public sphere (Macklin, Reference Macklin, Martin and Tirman2009).Footnote ⁶

As a result, women have routinely faced rejections of their asylum claims because their experience of persecution did not fit the refugee male paradigm (Fiddian-Qasmiyeh, Reference Fiddian-Qasmiyeh2014). This has led feminist critiques to conclude that international refugee law is structurally constructed against or, at least, in disregard of women (Haines, Reference Haines1997). Subsequently, international organisations and national jurisdictions have taken these feminist critiques on board and have begun to introduce a more gender-sensitive approach in their interpretation of refugee law, in an effort to correct the masculine bias of the Refugee Convention. In 2011, the EU introduced an explicit mention of gender as a ground of persecution in its legal framework.Footnote ⁷ Nonetheless, this progressive inclusion of gender into the refugee legal framework has principally relied upon narratives infused with stereotypical images of vulnerability and victimhood (Oxford, Reference Oxford2005; Smith, Reference Smith2016; Crawley, Reference Crawley2022). Too often, women are automatically categorised as ‘vulnerable’, solely because of their gender (Freedman, Reference Freedman2010) without acknowledging their agency in resisting forms of gender discrimination (Crawley, Reference Crawley2022). This shift has been criticised for victimising women, instead of treating them on an equal stand with men, thereby negating the depth and complexity of their experience (Edwards, 2010a).

The second component that needs to be taken into consideration when reflecting on the exclusionary nature of refugee law is race. Refugee law is not only highly gendered but also deeply racialised. After the fall of the Soviet Union, the quasi totality of asylum seekers came from developing nations (Mayblin, Reference Mayblin2017). In this post-cold war context, refugees could no longer be used as political pawns symbolising the oppressive Soviet regimes in contrast to the ‘free’ West (Hall, Reference Hall, Davies and True2018). Former colonial powers found themselves having to analyse and accommodate asylum claims lodged by individuals coming from (their) former colonies. Yet, the 1951 Refugee Convention had been drafted and framed from an exclusionary, Eurocentric, and colonial perspective. Indeed, the Convention was initially only meant to apply to European refugees displaced by World War II; it is only in 1967 that a Protocol was enacted to lift the original geographical and temporal restrictions delimiting the scope of the 1951 Convention. Moreover, the Convention initially contained a clause stating that ‘this treaty shall not apply to the colonies’ (Abuya, Krause and Mayblin, Reference Abuya, Krause and Mayblin2021). This provision should come as no surprise, since the European colonial powers in 1951 were repressing anti-colonial resistance and had no intention of extending protection to the so-called natives. In particular, imperial powers used the concept of race to justify colonial domination by presenting themselves as the bearers of civilisation and by dehumanising indigenous peoples (Kyriakides etal., Reference Kyriakides, Taha, Charles and Torres2019). Europeans justified imperial domination and economic expropriation by presenting themselves as bringing ‘civilisation’ to barbaric people (Saïd, Reference Said1978). This particular representation of indigenous people and its ‘othering’ effect is reflected in refugee law and practice. The Western asylum model has been accused of perpetuating colonial stereotypes by (re)presenting refugees as victims waiting to be rescued by the ‘white saviour’ extending his protection (Bhabha, Reference Bhabha2002; Luibhéid, Reference Luibhéid2008). As a result, in their quest for international protection, refugees are expected to fit into a narrative framed by a Western understanding of what a ‘genuine’ refugee is and should look like (Zagor, Reference Zagor, Jenkins, Nolan and Rubenstein2014).

In this regard, Edward Saïd’s thesis on Orientalism (Reference Said1978) offers a particularly apposite standpoint to analyse the unequal power relations between the ‘West’ and the ‘East’, the coloniser and the colonised, or – here – the adjudicator and the asylum seeker. Saïd, considered as the founder of postcolonial studies (Vandeviver, Reference Vandeviver2019), analyses how dominant narratives produced by European scholars in the nineteenth and twentieth centuries present the West as a civilised and ordered place, and the East as an exotic, barbaric, chaotic space (Saïd, 1978/Reference Said2014, pp. 155–156). He argues that this type of discourse is what laid the ground for the justification of the subsequent European colonisation of 85 percent of the globe by 1914 (Saïd, Reference Said1978/Reference Said2014, p. 132). His analysis is valuable because it shows how the production of knowledge serves power and its infrastructure.Footnote ⁸

In relation to gender, colonialism has relied on the impression that European colonisation would somehow bring freedom to women of colour and liberate them from their oppressive culture and religion and from those supposedly imposing them (Abu-Lughod, Reference Abu-Lughod2015; Wekker, Reference Wekker2016; Kapur, Reference Kapur2018). This narrative, which Leila Ahmed refers to as ‘colonial feminism’ (Reference Ahmed1972), is still being reproduced in the current post-independence context. For instance, it is reflected in oppressive policies against Muslim women, such as veil bans, on the basis that such cultural or religious practices are backwards (Kapur, Reference Kapur2018); the colonial power knows better than the Muslim woman what is best for her. Postcolonial and decolonial studies,Footnote ⁹ therefore, have engaged in a deconstructive analysis by showing how this discourse is used to justify imperial invasions and civilisation missions (Mahmood, Reference Mahmood and Wallach Scott2008). Frantz Fanon, who lived through the Algerian war of independence against French colonialism, is considered one of the pioneering scholars in anti-colonial studies. His work attacks the central premises of colonialism, i.e. the idea that colonised people are inferior to their coloniser (Fanon, Reference Fanon1952 and Reference Fanon1961). This assumption can be grounded in direct racism, which purports that white Europeans are superior to other races; or by indirect discrimination, which asserts that colonial powers are more ‘civilised’ than colonised countries. Although the majority of former colonies have technically secured independence, the unequal power relationships between former colonisers and colonised are still relevant to this day (Achiume, Reference Achiume2021). The idea that non-Western people are less civilised, vulnerable to exploitation and in need of ‘rescue’ is still widespread. This discourse is notably used to justify imperial wars and projects, such as the ‘war on terror’.Footnote ¹⁰ Fanon’s work has demonstrated how French colonialism was grounded in the idea that Algerians were inferior and incapable of governing themselves (Reference Fanon1961). Drawing on this framework, Hamid Dabashi criticises the narrative that presents Muslim countries as needing to be ‘saved’ from their own barbarism and backwardness (Reference Dabashi2011). Decolonial studies demonstrate not only how racialised this discourse is, but also the damaging consequences it can have. The racial component is at the centre of the colonial and decolonial narratives sustaining oppression and emancipation. In this regard, Fanon affirms that ‘the Arab, permanently an alien in his own country, lives in a state of absolute depersonalisation’ (Reference Fanon1965, p. 53).

Nonetheless, the component of gender has been traditionally absent from early anti-colonial studies. Subsequent work by postcolonial or ‘Third World’ feminists has demonstrated how racialised women are portrayed by the colonial discourse as oppressed and ‘vulnerable’ subjects in need of saving (Kapur, Reference Kapur2018). Spivak (Reference Spivak, Nelson and Grossberg1988) was one of the first postcolonial scholars to include gender in her analysis. In an epistemological examination, she argues that the production of certain narratives inflict moral violence upon the less privileged group – which she analyses in terms of race, gender and class. For instance, the constant representation of Muslim women as oppressed and submissive is a form of moral violence – or epistemic violence – that is deeply harmful (Abu-Lughod, Reference Abuya, Krause and Mayblin2015). In 2013, former British prime minister David Cameron referred to the ‘traditional submissiveness of Muslim women’, which prompted a sharp response (BBC, 2016). This stereotype is grounded in colonial frames that victimise Muslim women as such (Khan, Reference Khan2019). And yet, this narrative is still prevalent today; it is used to justify imperial invasions in the name of Muslim women,Footnote ¹¹ or to legitimise what Spivak referred to as the white man’s mission to ‘save brown women from brown men’ (Reference Spivak, Nelson and Grossberg1993, p. 93).

On this basis, I argue that this narrative is transposed in the representation of ‘Third World’ refugee women ‘who are portrayed as victims, subjects of pity’ (Pittaway and Bartomolei, Reference Pittaway and Bartolomei2001; Honkala, Reference Honkala2017; Crawley, Reference Crawley2022). And portraying them as vulnerable is ‘a form of violence’ that reproduces patriarchal and racial stereotypes (Ticktin, Reference Ticktin2005, p. 367). This discourse contrasts with the representation of brown refugee men as violent, threatening, and dangerous to women – thereby undeserving of protection (Olivius, Reference Olivius2016; Crawley, Reference Crawley2022). This colonial gendered discourse ultimately reinforces the colonial belief that the ‘West’ is superior to the ‘rest’ and capable of extending protection and ‘liberating’ the oppressed, ‘vulnerable’ women refugees from ‘Third World’ backward regimes. Rather than realising the emancipatory potential of vulnerability, as propounded in this Special Issue (Moreno-Lax and Vavoula), the notion is employed to the opposite effect, as the next sections confirm.

3 International Guidelines and Advocacy

International guidelines have played a significant role in conveying the image that refugee women are ‘vulnerable’ solely because of their gender. Firstly, the United Nations High Commission for Refugees (UNHCR) advises on which criteria make a refugee ‘vulnerable’ and thereby (more) deserving of protection. The agency has created a list of seven categories under which a refugee can apply for resettlement, for instance, including: (i) Legal and/or physical protection needs; (ii) survivors of violence and/or torture; (iii) medical needs; (iv) Women and girls at risk; (v) family reunification; (vi) children and adolescents at risk and (vii) lack of foreseeable alternative durable solutions (UNHCR, 2011). The fourth category is the focus of this study. It raises the question of why the UN refugee agency deems that forcibly displaced women and girls ‘at risk’ are necessarily more vulnerable than men and boys at risk? Why does UNHCR equate vulnerability with gender? In its handbook on resettlement, UNHCR clarifies that the category of ‘women and girls at risk’ refers to ‘women or girls who have protection problems particular to their gender, and lack effective protection normally provided by male family members’ (2011, p. 263).

On the one hand, it is undeniable that women and girls face gender-specific types of harm and hardship, and it comes as a positive and important step forward for the UN to recognise them. On the other hand, this categorisation is over-simplifying and seems to suggest that women can only be safe in their countries of origin if and when they enjoy male protection. Unintentionally, the organisation thereby reproduces patriarchal and disempowering discourses by suggesting that women are necessarily dependent on men’s protection. It is generalising a situation which cannot be generalised. Overall, the conflation of gender with vulnerability has been criticised by feminist scholarship for reproducing disempowering narratives (Clark, Reference Clark2007; Edwards, 2010a; Turner, Reference Turner2016). All humans can become vulnerable because of certain conditions or situations they may experience (Fineman, Reference Fineman2008), rather than because of their gender per se (Welfens and Bekyol, Reference Welfens and Bekyol2021). There are a plethora of valid approaches and conceptualisations in this regard specifically applicable in the migration context (on legally-induced and legally-mediated vulnerability, see Moreno-Law and Vavoula; on the ‘route causes’ of forced displacement, see Grundler; on ‘migratory vulnerability’, see Baumgärtel and Ganty; on ‘asylum vulnerability’, see Hudson; on ‘ecological vulnerability’ impacting ‘climate refugees’, see Ippolito, all in this Special Issue). Moreover, the category of ‘women at risk’ is too simplistic, as it focuses on one’s gender status instead of one’s experiences and necessities. It is based on the premise that single mothers or widows are necessarily (more) ‘vulnerable’ because they lack male protection, regardless of other conditions, such as economic resources, or personal, family, and social circumstances. This categorisation, therefore, assumes that women’s vulnerability is the result of a lack of patriarchal protection whilst, in reality, both women and men can be exposed to similar, including indiscriminate forms of violence, e.g. in the context of armed conflict, bombings or collective punishment. In such a scenario, both men and women are equally unable to protect themselves and their families (Manderson etal., Reference Manderson, Kelaher, Markovic and McManus1998). This does not deny the fact that women may and do suffer from gender-specific forms of violence and can be ‘more vulnerable’ than men in certain situations. However, equalling vulnerability with gender by default by creating a homogenous category is essentialist. It also fails to acknowledge resistance to gender-based violence and its political importance. Furthermore, a body of academic literature has demonstrated that conflating vulnerability with women erases the gender-specific forms of violence that affect men and boys,Footnote ¹² thereby creating new exclusionary biases (Malkki, Reference Malkki1995; Parrs, Reference Parrs2018). In reality, someone’s vulnerability stems from their politically situated individual experiences, rather than their gender alone or other decontextualised, ahistorical classifications. For instance, a woman might be facing a risk of persecution because she resists gender-based violence, such as female genital mutilation. In short, the category ‘women and girls at risk’ and its definition by UNHCR suggests and generalises passivity and powerlessness – for example, by presupposing that a woman is at risk because her husband has died, she has been repudiated or otherwise shunned by the male members of her family. On the contrary, I argue that someone’s vulnerability should be assessed in relation to their circumstances and experiences, not their gender or legal status in isolation. In this connection, a decolonial approach to humanitarianism (Rutazibwa, Reference Rutazibwa2019) pushes to focus on refugees’ political agency instead of on their perceived or presumed (reified) vulnerabilities.

International guidelines and advocacy instruments furthermore reproduce epistemic violence by denying the political agency of refugee women. In 2002, the UNHCR suggested in a guideline on refugee women that:

Although this statement is not meant to be normative or prescriptive, it still (re)produces patriarchal hierarchies by presenting women as necessarily less politically implicated than their male counterparts, and by describing them as only involved in ‘low level’ political activities. It creates a political hierarchy based on gender, and it erases the political agency of refugee women by suggesting that they are only involved socially or culturally and to a lesser extent than men in the public sphere. The guidelines make no mention of women as potential political opponents, leaders, soldiers, or rebel combatants (Edwards, 2010a). Yet, there is no evidence to suggest that women are inherently or necessarily less politically engaged than men. In anti-colonial resistance movements, some women were actually more politically active than their male counterparts because colonial authorities would perceive women as less ‘suspicious’ than men. During the Algerian War of Independence, an estimated 2,000 Algerian women fighters were militarily involved in the conflict (Amrane, Reference Amrane1991). Being a woman offered them an advantage, such as the fact that going through checkpoints would be easier than for their male counterparts, and their ability to hide the bombs under their burqas (Soto, Reference Soto2009). When Kenyans fought against British colonisation in the 1950s, women were reportedly involved in the same work as men (White, Reference White1990; Clough, Reference Clough1998). In her memoir, the Kenyan female fighter Elizabeth Gachika famously wrote: ‘we were the ones who fought for freedom, not the ones who told you about the cooking’ (as quoted in Presley, Reference Presley1992, p. 136). Finally, the role of Palestinian women in fighting ongoing Israeli occupation and settler-colonialism has been extensively documented and reported (Peteet, Reference Peteet1992; Sharoni, Reference Sharoni1995). Even before Israeli occupation, when Palestine was under British colonisation, Palestinian women played an active role in anti-colonial resistance. Biyadesh reports the story that took place in his native village in 1936, when the British army imprisoned all the men, which prompted women to attack military barracks at night, with only rocks as a weapon (as quoted in Alsaafin, Reference Alsaafin2014). This action succeeded and led to the release of all village men by the British army.

In these examples, women were politically involved as much as men, if not more. Therefore, making blatant and sweeping generalisations about women having ‘lower political activities’ compared to men not only lacks nuance, it furthermore contributes to reproducing epistemic violence by denying women’s agency and relegating them to a second-class status subaltern to men. Although it is correct to highlight that gender might impact political involvement, drawing up a priori hierarchies based on gender does not seem to service the purpose of advancing women’s rights. Thus, following criticism, UNHCR removed the aforementioned paragraph from its guidelines on gender.

Aside from reproducing patriarchal premises (Hall, Reference Hall, Davies and True2018), I argue that these stereotypes are also grounded in colonial narratives, because they consolidate the idea that native women are apolitical subjects (Locher-Scholten, Reference Locher-Scholten2000). The following section turns to examine how Courts contribute to reproduce this discourse and the narratives that depoliticise the experience of ‘Third World’ refugee women, relying on ‘vulnerability’ assumptions in their reasoning.

4 Gendered colonial discourses regarding refugee women in case law

The progressive introduction of gender considerations within international refugee law has principally relied on a narrative that presents refugee women as ‘vulnerable’ victims. This discourse largely relies on the assumption that refugee women cannot be political actors both because of their gender and their ethnicity or national origin. This section relies on the decolonial framework laid down earlier to deconstruct the way in which national jurisdictions depoliticise the experiences of refugee women by (re)producing certain (vulnerability-based) narratives.

The first case relates to Nada, a Saudi Arabian woman in her twenties who sought asylum in Canada.Footnote ¹³ She faced persecution from authorities for refusing to abide by the Kingdom’s gendered laws, including the obligation to wear a full-face veil. She reported that she left the house one day without covering her face and was, as a result, stoned, spat on and subjected to verbal harassment (Akram, Reference Akram2000). In addition, Nada listed other patriarchal laws that severely affected her freedom: as a woman, she was forbidden from driving, she was unable to study the field she wanted, she could not travel without the authorisation of a male relative and she additionally feared the religious police who would beat her with sticks or jail her if she refused to comply with the dress code. Subsequently, Nada introduced a claim for asylum on the political ground of the Refugee Convention and argued that she faced persecution from Saudi Arabian authorities based on her political beliefs and her feminism. The Canadian Immigration and Refugee Board, however, flatly rejected her claim on the basis that ‘it was not credible that an Arab Muslim woman would disagree with the authorities of a Muslim State’ (Akram, Reference Akram2000, p. 25). The Canadian Court’s reference to the fact that Nada was ‘an Arab Muslim woman’ to justify the dismissal of her claim shows how deeply racialised and gendered this decision is. It reflects the Western perception of Muslim women as passive and echoes David Cameron’s comments about the so-called ‘submissiveness’ of Muslim women (BBC, 2016). Moreover, the decision appears to imply that somehow only white/Western women are capable of expressing political opinions. The decision is even more troubling when considering that Canada has been a pioneering country in advancing gender equality – which does not seem to apply equally to all ethnicities. Inevitably, it does raise the question of whether the outcome would have been similar, had the claimant been a white/Western(ised) woman. When rejecting Nada’s claim, the Canadian Immigration and Refugee Board ruled that she should just ‘comply with the laws of general application she criticises’ and ‘show consideration for the feelings of her father who, like everyone else in her large family, was opposed to the liberalism of his daughter’ (Wheelwright, Reference Wheelwright1993, p. 11). This case echoes Howard’s argument that non-Western women are expected to renounce their human rights for the ‘greater good of the collectivity’, and that such renunciation would be costless because their identity is merged with that of the group (Reference Howard1993, p. 332).

Moreover, the Court’s ruling reflected the implicit notion that Islam, and probably religions in general, are seen as inherently and fundamentally opposed to feminism. Whilst most religions are systemically dominated by men and have patriarchal structures, religious beliefs are not necessarily incompatible with feminist convictions. Western feminism’s dominant view that views religion as an antithesis to feminism has been challenged by cultural relativists, who highlight the role played by religious organisations in enhancing female emancipation (Hua, Reference Hua2010; Nyhagen, Reference Nyhagen2019). Brown (Reference Brown2012) even believes that, in some cases, secularism can be a barrier to female empowerment in certain contexts.

Finally, the Board considered that Nada’s feminism did not constitute ‘political opinion’ under the Refugee Convention. Ultimately, this raises the question of how political opinion should be understood, especially when applying an intersectional framework that considers both race and gender. From this perspective, not only is the Board’s decision a form of epistemic violence, but it also illustrates the devastating consequences that stereotypes can have. In this case, it led to the unfair rejection of the claimant’s refugee status, leaving her unprotected. Nada’s case was rejected because her story did not fit the Western paradigm of how racialised women should ‘behave’ or ‘act like’.Footnote ¹⁴

Other cases have confirmed this pattern by illustrating how Courts deny refugee women’s agency on the pretext that their feminist beliefs are not a valid form of political opinion. In particular, Muslim women who face excessive punishment for violating dress or moral codes see their asylum claims rejected on the basis that their claim is not political (Akram, Reference Akram2000; Ticktin, Reference Ticktin2005). In 1993. In Fatin v. INS, a US Federal Court refused to accept that an Iranian woman’s refusal to wear the chador amounted to a political opinion.Footnote ¹⁵ Instead, the Court considered that her refusal to comply with patriarchal dress codes was simply an act of non-compliance and, thereby, rejected her claim for international protection. Women’s resistance against the codification of patriarchal dress code based on certain religious practices, and the severe consequences associated with such resistance, was not recognised as valid grounds for refugee status recognition by adjudicators. Although scholars describe the 1990s as ‘the decade for refugee women’ (Oxford, Reference Oxford, Arbel, Dauvergne and Millbank2014, p. 167), the steps taken at that point towards the inclusion of women did not make room for women’s political agency – at least, not entirely.

Indeed, case law demonstrates how reluctant Courts can be in granting asylum when claims of political opinion do not fit Western presuppositions of non-Western women – in particular the stereotype that they are ‘submissive’, vulnerable and fearful of their male relatives. By contrast, Courts appear eager to grant protection to women when their stories fit ‘white savour’ narratives. In 2006, for instance, the UK Asylum and Immigration Tribunal granted refugee status to an Iraqi woman on the basis that she was ‘Westernised’, educated, refused to wear the hijab and spoke fluent English.Footnote ¹⁶ The judges concluded that the appellant had ‘a history of collaboration with the West’ on the basis that she used to work for an engineering company that negotiated oil contracts with the Iraqi government and Western corporations (para. 74). In short, the claimant was granted refugee protection because she fitted into all the Western stereotypes of what a ‘perceived political opinion’ is. The decision, subsequently, raises the question of whether the claim would have been accepted if the claimant did fit into the view that sees feminism as a Western secularist concept. Moreover, the decision made no reference to the claimant having her own political opinion, despite being granted refugee status – instead, it ruled that she faced persecution due to the persecutory state’s ‘perceived’ political opinion that it ascribed to her, thereby diminishing the importance of her personal agency. It thus reproduced disempowering narratives of a passive victim rather than an active rights-holder challenging prevailing forces. Although the concept of ‘perceived’ political opinion emerged in the case law is not specific to women, it arguably reinforces the idea of passiveness and lack of agency when applied to them.

More recently, in a 2018 case involving an Afghan woman, a Dutch Court ruled that women who adopted a so-called ‘Western lifestyle’ were entitled to refugee status if they could prove that this lifestyle was linked to their political or religious beliefs.Footnote ¹⁷ The Court did not, however, define what a ‘Western lifestyle’ means. In the case Fatin v. INS mentioned earlier, the Court rejected the political claim of the applicant but accepted to grant her refugee status based on a separate gender claim: the fact that she belonged to ‘the social group of the upper class of Iranian women who supported the Shah of Iran, a group of educated Westernized free-thinking individuals’.Footnote ¹⁸ This case clearly illustrates the colonial idea that Western freedom is the only valid form of freedom. It furthermore reinforces the orientalist dichotomy of civilised versus uncivilised, ‘free world’ versus ‘oppressed world’ trope.

Ultimately, this framing forces refugee women to fit into certain narratives and stereotypes in order to have their asylum claims recognised. Women considered worthy of protection are those who fit into white saviour standards and prove that they have the so-called ‘Western lifestyles’. In the case Sophia (1999), as reported by Oxford (Reference Oxford2005), the claimant was a Jordanian woman who faced forced marriage and constant domestic abuse. To the question of what would happen if she were to be returned to her home county, Sophia answered ‘I would kill my husband’. In the final submission, the lawyer replaced this response with ‘my husband would kill me’. This case illustrates, perhaps through an extreme example, how the asylum system makes little room for agency and forces women to present themselves as vulnerable victims. It refuses to acknowledge the full extent of their anger and desire to stand up for themselves.

In the case AK6, a German Administrative Court granted refugee status to an Iraqi woman who argued she faced persecution due to her ‘Western way of life’.Footnote ¹⁹ The Court ruled that the applicant ‘gave the court the impression of a self-confident young woman, oriented towards a Western lifestyle, who would like to stand on her own feet and who is a Sunnite on paper only’. This decision is partly positive because it acknowledges the woman’s agency and choice in living a particular lifestyle. However, it does so by unnecessarily opposing it to Islam (‘she is a Sunnite on paper only’) thereby implying that being a Muslim, a woman can only be free if she is nonreligious and Westernised.

Accordingly, I argue that this disempowering narrative, perpetuated by Courts, denies the full autonomy of Muslim refugee women on equal footing with their white, Western, non-religious counterparts, paying lip service to the feminist cause. This case also sheds a light on the gendered stereotypes framed behind the concept of ‘particular social group’ (PSG) as applied to brown refugee women, to which the next section turns. It will demonstrate how the PSG Convention ground reproduces and prolongs the gendered colonial discourse that has already been explored.

5 The ‘particular social group’ (PSG) Convention ground and its disempowering effects

In this section, Iconsider the use of the PSG category within the overall context of colonial discourse and the representation of brown refugee women as victims. Initially, the PSG category was used to correct the absence of gender or sex in the definition of a refugee per Article 1(A)2 of the 1951 Convention. Canada was the first country to adopt a broad interpretation of the ‘membership of a particular social group’ category to encompass women in the Ward case.Footnote ²⁰ Progressively, more countries began adopting a similar approach and used the PSG category to accommodate claims related to gender-based persecution. In particular, states started recognising gender-specific forms of violence, such as rape, female genital mutilation and forced marriages (Spijkerboer, Reference Spijkerboer2000) . In 1991, UNHCR stated in its Guidelines that ‘women fearing persecution or severe discrimination on the basis of their gender’ should be considered members of a PSG (1991, p. 19).

Nonetheless, the academic literature has pointed out that, throughout the years, the PSG Convention ground has been over-used at the expense of the political opinion one (Edwards, 2010b; Honkala, Reference Honkala2017). UNHCR itself has acknowledged that ‘in some cases, the emphasis given to the social group ground has meant that other applicable grounds, such as religion or political opinion, have been over-looked’ (2002, p. 7). Drawing on this, I argue that women who obtain the refugee status based on the PSG category are essentially approached as vulnerable victims. They are considered as worthy of protection typically when fleeing domestic violence, forced marriage, female genital mutilation, or other forms of gender-based violence, of which they are portrayed as passive recipients.Footnote ²¹ Constantly, decisions highlight that women are victims in need of rescue, rather than active rights-holders who chose to resist and oppose gender-based forms of violence.

As mentioned in the introduction to this article, the 1951 Refugee Convention and its 1967 Protocol were drafted from a white male-centric perspective and were largely gender-blind. Notably, the five Convention grounds under which a refugee can claim persecution do not include sex or gender. However, following feminist critiques against the mainstream international refugee law regime (Indra, Reference Indra1987; Crawley, Reference Crawley2001), national jurisdictions began incorporating gender in their refugee status determination processing. First, practice challenged the public/private distinction which considered that only persecutory acts committed by the state were to be recognised as falling within the remit of the Refugee Convention (Charlesworth and Chinkin, Reference Charlesworth and Chinkin2000). As already stated, national jurisdictions began to recognise gender-based violence, such as female genital mutilation,Footnote ²² rapeFootnote ²³ and domestic violence,Footnote ²⁴ as valid forms of persecution for refugee status qualification. Non-state actors who committed gender-based persecution were also recognised as potential actors of persecution in cases whereby the state did not or could not provide protection to women (Anker, Gilbert and Kelly, Reference Anker, Gilbert and Kelly1997). Jurisdictions started acknowledging the political nature of so-called ‘private’ acts of persecution that are gender-based (Hall, Reference Hall, Davies and True2018). Second, the fifth ground: PSG, emerged as an effective alternative to the lack of references to sex or gender in the Convention. In the first set of Guidelines related to gender that UNHCR adopted, the organisation stated that women ‘fearing persecution or severe discrimination on the basis of their gender should be considered a member of a social group for the purposes of determining refugee status’ (1991, p. 1). APSG is to be understood as a group of persons who share ‘common and immutable characteristics’Footnote ²⁵ or ‘innate and unchangeable characteristics’,Footnote ²⁶ including women.

However, PSG is not to be considered as the exact equivalent of sex or gender. Being a woman on its own is not enough. The applicant needs to prove that she belongs to a social group with (some additional) other immutable characteristics. In Gomez v. INS, a US Court recognised for the first time that a PSG could provide a valid basis for an asylum claim related to gender-based violence; nonetheless, the decision specified that the PSG was not constituted by sex or gender itself, but by the additional caveats constituting the group of ‘women who have been previously battered and raped by Salvadoran guerrillas’.Footnote ²⁷ Similar decisions by other jurisdictions that incorporate gender into their understanding of PSG include examples such as ‘widows at risk of forced marriage’Footnote ²⁸ or ‘unmarried women from families whose traditional self-image demands a forced marriage’.Footnote ²⁹ In the landmark case Shah and Islam, the UK House of Lords ruled that the applicants belonged to a PSG defined as ‘women who had offended against social mores or against whom there were imputations of sexual misconduct’ and were not protected by the state.Footnote ³⁰ These groupings have been criticised for forcing women to fit into artificial categories (Edwards, 2010b), also referred by Justice McHugh as ‘intellectual constructs’.Footnote ³¹

What is more, against this backdrop, I argue that reliance on the PSG category as substitute for gender has led to the depoliticisation of refugee women’s experiences (Anker, Reference Anker and Lambert2017). As Edwards points out, ‘PSG has now become the default ground for women’s claims, even when one or more of the other grounds may be equally or more applicable’ (2010a, p. 28). Crawley argues that this overemphasis on PSG in relation to refugee women reinforces ‘the existing and paradigmatically masculine normative structures of international refugee law’ (Reference Crawley and Indra1999, p. 326). Dauvergne, in turn, highlights that, although some cases have considered the transgression of social norms as the expression of a political opinion, these are only a few cases in comparison with those based on the PSG ground (2021). UNHCR has also observed an excessive focus on the PSG ground regarding asylum claims introduced by women, which has led to other grounds, such as religion or political opinion, being largely ignored (2002). Indeed, national practice tends to consider refugee claims from female asylum seekers as automatically falling under the PSG category, despite applicants explicitly claiming one of the other four grounds (Edwards, 2010b). In particular, the asylum claims based on the ‘political opinion’ ground tend to go unrecognised. One of the early arguments from the feminists who opposed amending the Refugee Convention to introduce a sixth Convention ground of ‘sex’ or ‘gender’ was that doing so might lead to a categorisation and essentialisation of women’s experiences under one homogenous category, giving the impression that women’s persecution is fundamentally different from men’s (Freedman, Reference Freedman2015). I argue that we eventually fell into this trap by largely equating women with PSG.

For instance, in the case of Shah and Islam,Footnote ³² the Court could have simply considered that opposing certain social mores is the expression of a political opinion, instead of creating an artificial PSG.Footnote ³³ Women who transgress social norms are regularly denied the ‘political opinion’ ground and automatically classified under the PSG category. In September 2020, a French Court granted refugee status to a woman from Burkina Faso who fled forced marriage.Footnote ³⁴ The Court considered that she belonged to the PSG of ‘women fleeing forced marriages’ in a society where, according to the judges, ‘forced marriages were so widely practiced that they became a social norm’.Footnote ³⁵ But if they are a social norm, why was the claimant not granted refugee status based on the political opinion ground?

Social norms are common standards regarded as socially acceptable or appropriate behaviour, the breach of which has social consequences (Munday and Chandler, Reference Munday and Chandler2011). Therefore, opposing a social norm is the expression of a political opinion because the individual is aware of the consequences attached to breaching it. I argue that, denying the political opinion ground to refugee women who transgress social norms, and recognising instead their refugee status based on their ‘gender’ or other ‘immutable characteristics’ – i.e., something inherent to them that they cannot control – Courts are depoliticising their experiences and denying them their autonomy. Whilst many women are victims of patriarchal social norms, only a few decide to resist and oppose them because they deeply believe in gender equality – which represents, in itself, a political opinion. Ithus posit that, by transgressing social norms and standing up against patriarchy, these refugee women are expressing their legitimate political opinion in the sense of the Convention. Many women are victims of patriarchal social norms but simply abide by them. Others dare and decide to transgress them, fully aware of the consequences, because they believe in women’s rights and gender equality. That, in itself, is the expression of a powerful political belief. It follows the lead of some feminists who consider that all kinds of violence against women is political (Bilaud and Direnberg, Reference Billaud and Direnberger2021; Goodwin-Gill and McAdam, Reference Goodwin-Gill and McAdam2021). Therefore, I argue that all resistance to that violence is also inherently political, endorsing Macklin’s argument who suggests that opposition to institutionalised discrimination should be considered the expression of a political opinion (Reference Macklin1996). Reducing refugee women to passive apolitical actors simply fleeing generic gender-based violence does not serve justice.

In 2011, an atheist Iranian woman was subjected to physical and psychological violence by her husband and sought asylum in Greece.Footnote ³⁶ She was granted refugee status based on the PSG category, while her political opinion claim was denied. The Court ruled that the applicant was:

It is difficult to understand why the applicant was not granted refugee status on the political opinion ground in this case, since resisting sexual violence and transgressing social or cultural norms constitutes, again, the expression of a political opinion. In this regard, a distinction must be drawn between gender-specific and gender-based persecution. On one hand, gender-specific persecution encompasses practices, such as female genital mutilation or so-called honour crimes; on the other, voicing non-conformist opinions and resisting gender-based patriarchal cultural practices is the expression of a political opinion. The 1951 Convention ground of ‘political opinion’ encompasses opinions on gender roles, and both a woman and a man can be persecuted for holding feminist views. I argue that transgressing patriarchal social norms and facing harm should be considered as constituting a political opinion for the purposes of the Convention. Case law confuses gender with women and ignores the fact that men who do not conform to social norms and standards of masculinity can also be exposed to persecution as a result (Connell and Messerschmidt, Reference Connell and Messerschmidt2005). The focus should not be on gender per se, but on the disapproval of social norms as a form of political opinion.

In Shah and Islam, Lord Hope suggested that ‘[t]he reason why the appellants fear persecution is not just because they are women. It is because they are women in a society which discriminates against women’.Footnote ³⁷ Yet, as Chief Justice Gleeson pointed out in Khawar, there is not one single society in the world that does not discriminate against women.Footnote ³⁸ However, there are some women who resist gender-based persecution because of their strong political beliefs in gender equality, and who, subsequently, face persecution as a consequence thereof. They are courageous political activists, not vulnerable victims worthy of pity. The prevailing approach of Western Courts reflects a highly orientalist perspective, which portrays Muslim women as fleeing a barbaric religion and backward culture, who come to seek safe haven in a ‘civilised’ country that strives for gender equality. This is what Lord Hope seems to suggest by referring to Pakistan as ‘a society which discriminates against women’ (ibid.). This discourse fits into the colonial dichotomy of East versus West, Orient versus Occident, barbaric versus civilised – problematised, as mentioned earlier, by Saïd (Reference Said1978).

Moreover, it also (re)produces the colonial discourse that represents Muslim women as victims of their own religion and in need of saving, rather than self-sufficient political actors who decide to oppose certain religious and cultural practices. It furthermore perpetuates essentialist and orientalist stereotypes about Islam by presenting it as inherently static, barbaric, and patriarchal. This is notable in the aforementioned case about the Iranian woman whereby the Greek Court ruled that the claimant’s persecution was due to ‘practices of Islam’ that were ‘part of a situation that cannot be changed’.Footnote ³⁹ Similar patterns can be identified in other national jurisdictions. In a case involving an Armenian woman from Turkey, the Canadian Immigration Board found that she belonged to a particular social group made of ‘single women living in a Moslem country without the protection of a male relative (father, brother, husband, son)’.Footnote ⁴⁰ In another case, the same Board ruled that the claimant belonged to a PSG of ‘young women without male protection’.Footnote ⁴¹ Once again, this case reproduces the stereotype, fostered also by UNHCR in its policy Guidelines and advocacy materials, that women need a male protector to survive. In Mrs N., S., and S., a woman fleeing female genital mutilation applied for asylum in France.Footnote ⁴² The Court granted her refugee status based on the PSG ground, considering that ‘in a population in which genital mutilation is widespread, children and women that are non-mutilated represent a particular social group’ (ibid., para. 8).Footnote ⁴³ Here, the judges portray the applicant as a powerless victim of gender-based violence and fail to acknowledge her agency in opposing and refusing this violence. It is not the absence of past mutilation which characterises her, but rather her refusal to be subjected to mutilation at any point in the future. By classifying her plight under the PSG ground, the asylum system fails to acknowledge her voice.

I argue that this default application of the PSG category to female asylum claims reproduces a disempowering discourse that robs women of their political agency. Ultimately, it remains doubtful whether the progressive introduction of a gender-sensitive interpretation to correct the masculine bias of refugee law has done complete justice to women. Of course, the PSG category has been successful in correcting the absence of sex/gender in the Refugee Convention, and it has allowed the recognition of women fleeing gender-based violence as refugees. However, there has been an over-reliance on the PSG ground, by automatically and excessively applying it in cases involving women. Isuggest that, to advance equality between men and women applicants, the political opinion Convention ground should instead be favoured. Accordingly, the transgression of social norms should be considered as the expression of political opinion rather than a characteristic of the specific PSG.Footnote ⁴⁴ In particular, Iposit that all violence against women should be considered political by default, because it reflects political views deeply embedded in gendered and patriarchal opinions.

6 Courts denying political agency to (non-Westernised) refugee women

The previous section of this article has explored how Courts tend to depoliticise the experiences of refugee women transgressing social norms by automatically classifying them under the PSG category, instead of recognising and acknowledging their political opinion as a basis for protection. Even when female asylum seekers manage to secure refugee status based on the political opinion ground, Courts still find ways to depoliticise their experiences by attributing their political opinion to their partners or other members of their family. The present section explores this pattern by focusing on a recent case from a French Court which had to assess the asylum claim of a refugee woman from Syria,Footnote ⁴⁵ looking at the doctrine of ‘imputed’ political opinion.

The claimant was a Syrian woman from Damascus, referred to by the Court as ‘Mrs K’. She applied for asylum in France and claimed she faced a well-founded fear of persecution from the Syrian authorities because of her political opinion. She fled to France with her husband who also submitted an asylum claim. Referred by the Court as Mr A, he claimed that he faced a well-founded fear of persecution because of his refusal to do his military service and thereby serve in the army of Bashar Al-Assad. Mr Ahad been able to avoid his military service because he studied at a higher education organisation affiliated with the Scientific Studies and Research Centre (SSRC), a Syrian government agency coordinating scientific activities – including nuclear and military operations.Footnote ⁴⁶ Both Mrs Kand Mr A’s asylum claims were examined jointly by the Court.Footnote ⁴⁷

Mrs Kargued before the adjudicator that she faced a well-founded fear of persecution because of her own political opinion. She explained precisely how she started developing dissident political views early in life. In high school, she would write and share anti-regime posts on her social media. She had also been accused of destroying a portrait of Bashar Al-Assad, thereby making her being perceived as an opponent of the regime. In 2013, her brother was kidnapped and detained for several months before her family was informed of his death. In 2016, she was stopped during a roadside check and was informed that the regime had been searching for her since December 2014. After being arrested, she was released shortly thereafter because of her husband’s status, i.e. as someone studying at a government agency. She was nonetheless given notification by the authorities that she had to come to the police station within 72 hours. It is at this point that she decided to immediately flee with her husband.

The French Court rejected Mrs K’s claim, considering her story not credible, pointing out that, although she had been arrested by the Syrian authorities, she had been released shortly thereafter. On the other hand, the Court did believe Mr A’s account and ruled that he had a well-founded fear of persecution based on political opinion because of his refusal to do the military service. Regarding Mrs K, the Court finally decided to recognise her refugee status based on the fact that her husband had a dissident political opinion. The Court ruled that the Syrian regime could ‘impute’ to Mrs Kthe political opinion of her husband. In short, the judges were of the view that Mrs K’s own political views were not relevant or sufficient, and that any future risk of persecution facing her should be tied to her husband’s political stand, to which hers, as his wife, would be assimilated by the Syrian authorities.

The Court’s reasoning is disturbing considering that the reason the couple fled Syria in the first place was because Mrs Khad been given 72 hours to report herself to the police. In fact, her husband was probably less likely to be persecuted because of his status as a former student of a government agency. When Mrs Kwas initially arrested and detained, she was released because of her husband’s position. Yet, the judges turned the story upside down and decided to deny her claims on the ground that her story was not ‘credible’. Still, they granted her protection on the basis of her ‘imputed political opinion’ as an extension of her husband’s. To my mind, the credibility assessment in this case seems thus grounded in gender and racial stereotypes. The Court reached such a conclusion because Syrian women – and probably all Muslim or ‘brown’ women in general – are perceived as oppressed, submissive and dependent on their male relatives. Anyone who steps out of this cliché is perceived as a liar. Although the unfairness of the ‘credibility assessment’ process affects all refugees, the difficulty to believe specifically brown women appears to follow a colonial narrative of the Third-World woman as an apolitical and submissive subject. As Hua explains, Courts refuse to consider refugee women as potential political actors because it would be ‘contrary to the construction of Muslim difference as backward and antithetical to feminism’ (Reference Hua2010, p. 392). Within this framework, Muslim women can only fit into the white saviour discourse that portrays them as vulnerable victims, not human rights heroes. In addition to the fact that this discourse reproduces colonial misconceptions that constitute epistemic violence, Ialso emphasise the harmful consequences it has on refugee women in practice. The reproduction of such a gendered and racialised narrative leads to the unfair rejection of their asylum claims. Although they might be able, in certain cases, to obtain subsidiary protection and other forms of humanitarian status, it is not equivalent to refugee status. As Zagor points out, the recognition of refugee status is not just about obtaining a residence permit; it is about having one’s political experience recognised and acknowledged (Reference Zagor, Jenkins, Nolan and Rubenstein2014).