Abstract In this article I analyse women asylum seekers’ claims of gendered ill-treatment under Article 3 of the European Convention on Human Rights. I argue that the European Court of Human Rights moves away from creating equal conditions of protection for women asylum seekers every time it adopts two modes of reasoning: under scrutinizing the gendered roots of risk of ill-treatment and over scrutinizing individual capacity to deal with the risk. The first mode of reasoning overlooks the social and institutional conditions that render women vulnerable to ill-treatment. The second mode over emphasizes a woman’s ability to protect herself and/or male relatives’ capacity to protect her. The two modes suggest that women asylum seekers risk ill-treatment because of personal failures/limits rather than socio-institutional failures/constraints. These modes of reasoning may oversimplify concrete risks and recreate women’s subordinate status in human rights discourse. To counter these faults, in this article I propose to reappraise the risk of gendered ill-treatment structurally and relationally. 1. INTRODUCTION Though ‘not there yet’,1 over the past years refugee law has significantly developed to respond to women’s gender-related claims of persecution. While obstacles remain2 and reform is ‘at best half-done’, gender issues have moved ‘from the margins to the centre’ of refugee law.3 International human rights law, too, has been sensitive to the plight of refugee women and women asylum seekers.4 The European Court of Human Rights (‘the Court’ or ‘the Strasbourg Court’), however, may be lagging behind. The Court has recognized the seriousness of the harms women asylum seekers often escape from but has yet to take seriously the risk assessment of these harms. In this article, I analyse women asylum seekers’ claims of gendered ill-treatment under the non-refoulement principle that the Court has read into Article 3 of the European Convention on Human Rights (ECHR).5 I argue that the Court moves away from creating equal conditions of protection for women asylum seekers every time it under-scrutinizes the gendered roots of the risk of ill-treatment and over-scrutinizes the individual capacity to deal with the risk. The first mode of reasoning overlooks the social and institutional conditions that render women vulnerable to ill-treatment. The second mode over-emphasizes a woman’s ability to protect herself and/or her male relatives’ capacity to protect her. These modes of reasoning suggest that women asylum seekers risk ill-treatment because of personal failures/limits rather than socio-institutional failures/constraints. The reasoning may oversimplify the individual risk, reduce a gender inequality problem to an idiosyncratic problem and recreate women’s subordinate status in human rights discourse. To counter these faults, I propose to reappraise the risk of gendered ill-treatment structurally and relationally. Assessing the risk structurally implies unearthing the deep-rooted gendered conditions shaping this risk. Evaluating the risk relationally involves looking at the ways in which such conditions may affect a woman’s capacity to deal with the risk. Illuminating these arguments are insights from vulnerability theorist Martha Fineman, who argues for attention to the role of the state and society in reinforcing human vulnerability,6 and from feminist scholars pointing to the role of relationships in shaping individual autonomy.7 My arguments further draw on insights from the work of the United Nations Committee on the Elimination of Discrimination against Women (‘CEDAW Committee’) and the Office of the United Nations High Commissioner for Refugees (UNHCR). I start the discussion by mapping out Article 3 ECHR case law involving women asylum seekers allegedly fleeing a risk of gendered ill-treatment in their home states. I then focus on cases in which the Court thinly examines the gendered structures shaping this risk and thickly evaluates the private capacity to deal with it. I argue that these modes of assessment may encourage protective stereotypes of women long criticized by feminists in international human rights law.8 I finish by discussing cases in which the Court engages more substantively with the gendered structures underlying women’s alleged risk of ill-treatment. 2. ARTICLE 3 ECHR AND GENDERED ILL-TREATMENT OF WOMEN ASYLUM SEEKERS Under Article 3 of the ECHR, several women whose requests for asylum failed domestically have sought the Court’s protection against what they saw as arbitrary refoulement. Though the women’s claims surveyed for this article rely on various grounds, alone9 or combined with others,10 the large majority of these claims (29 of 37)11 concern a risk of gendered ill-treatment.12 For present purposes, gendered ill-treatment is understood in three ways:13 (1) as forms of mistreatment typically inflicted on women (for example, rape)14 for reasons not related to gender (for instance, religious, ethnic reasons); (2) as mistreatment because of gender regardless of the form the mistreatment may take (women may experience ill-treatment because of gender when they do not conform to socially constructed norms of femininity and challenge women’s subordinate status in society);15 and (3) as forms of mistreatment usually experienced by women because of gender. As Heaven Crawley notes: A woman may be persecuted as a woman (eg raped) for reasons unrelated to gender (eg activity in a political party), not persecuted as a woman but still because of gender (eg flogged for refusing to wear a veil) and persecuted as and because she is a woman (eg female genital mutilation).16 While Strasbourg examples of the first17 and second scenarios18 described by Crawley appear uncommon, examples of the third scenario seem more frequent.19 The Court has found an Article 3 ECHR violation in only three of the 29 gendered ill-treatment cases. In one of them, the basis for finding a violation was primarily procedural.20 In the other two, the violations were based on a substantive assessment of the issues at stake.21 Of the remaining 26 claims, seven have been decided on the merits.22 The rest have been found manifestly ill-founded and therefore declared inadmissible.23 Nine of the 26 unsuccessful applications have been rejected nearly exclusively on credibility grounds.24 These are usually instances where the Court finds applicants’ stories to be insufficiently detailed, inconsistencies unsatisfactorily explained and documents low in evidentiary value. The majority of the 29 gendered ill-treatment cases involve alleged threats from non-state actors, including partners,25 family members26 and communities.27 While acknowledging that gendered ill-treatment may affect asylum seekers of different genders,28 this article focuses on female asylum seekers. The focus is motivated by the salience and recurrence of these instances in Strasbourg case law as it currently stands. The focus on women is moreover driven by long-standing feminist concerns about women’s marginalization from mainstream interpretations in international human rights law.29 One of the well-known ways through which such marginalization has taken place is the public/private divide, including through the distinction between acts of state and non-state agents.30 This apparently neutral distinction reinforces gender inequality when linked to the reality that harms against women are often at the hands of non-state actors.31 Historically, international human rights law and refugee law have been more concerned with harms typically suffered by men in the ‘public sphere’ than with harms generally experienced by women in the ‘private realm’.32 By now, however, it has been recognized under both international human rights law and refugee law that ill-treatment by non-state actors may amount to a human rights violation or to persecution. International human rights law recognizes that the state may be held responsible for offensive acts by private actors if it fails to act with ‘due diligence’ to prevent such acts or to investigate and punish them.33 Refugee law accepts that serious acts by non-state agents can be considered persecution ‘if such acts are knowingly tolerated by the authorities, or if the authorities refuse, or are unable, to offer effective protection’.34 The Strasbourg Court has regarded several harms commonly inflicted on women by non-state actors as contrary to Article 3 of the ECHR. Think of serious domestic violence,35 rape36 and female circumcision (commonly known as ‘female genital mutilation’ or ‘FGM’ in international human rights law).37 Moreover, the Court has accepted that Article 3 of the ECHR may apply to prevent an expulsion where ill-treatment would come from persons ‘who are not public officials’.38 Applicants fearing harm at the hands of non-state actors must show that the risk is real and that the receiving state is ‘not able to obviate the risk by providing appropriate protection’.39 States parties to the ECHR, however, may argue that applicants can relocate to a safe area within the home country and escape the risk.40 Some of these are no doubt important efforts by the Court to attend to the kinds of harm frequently inflicted on women asylum seekers. Nevertheless, inattention to the gendered structures limiting the home state ability to protect women and conditioning women’s internal relocation may undermine these efforts. Gendered structures are here understood as the institutional and socio-cultural entrenched conditions that impact disproportionately or differently on women’s risk of ill-treatment. These conditions include impunity for harms usually caused to women and widespread discrimination against women. The next two parts of this article discuss two modes of reasoning that move away from creating equal conditions of protection for women asylum seekers in ECHR law: thin examination of gendered structures and thick evaluation of private capacity to deal with the risk. The former arises from a formalistic, cursory, vague or simply non-existent assessment of the home state ability to protect women and of the socio-cultural constraints women may face in relocating internally. The latter over emphasizes a woman’s ability to protect herself or male relatives’ capacity to protect her. The last part of this article looks at modes of reasoning that move towards creating equal conditions of protection for women asylum seekers. These forms of reasoning attend to the ways in which gendered structures may render women particularly vulnerable to ill-treatment. Before developing these arguments, a few caveats are necessary. Though large enough to carry out a meaningful examination, the selected case law is by no means comprehensive. Moreover, the analysis in the next parts leaves out the above-mentioned case decided on the basis of a procedural assessment.41 It also excludes seven of the nine cases rejected exclusively on credibility grounds.42 The complexity of the issues raised in these latter cases requires separate attention exceeding the scope of this article. Two of the cases decided exclusively on credibility grounds are nonetheless discussed given their relevance to the issues here addressed.43 This leaves us with 19 cases decided on several grounds, including credibility, availability of appropriate home state protection and internal relocation.44 Unless relevant for the points made in this article, the Court’s credibility assessment in these remaining cases will not be discussed. 3. THIN ASSESSMENT OF GENDERED STRUCTURES: OBSCURING STRUCTURAL AND PARTICULAR VULNERABILITY When the Strasbourg Court evaluates whether there is a ‘real risk’ of ill-treatment, it focuses on ‘the foreseeable consequences’ of sending the applicant to her home country.45 These consequences are established in view of ‘the general situation’ in this country, as described in reliable reports,46 and of the applicant’s ‘personal circumstances’.47 The Court’s reasoning in the cases discussed in this Section is characterized by a thin assessment of the general situation in the applicants’ home countries. In particular, the Court scrutinizes formalistically, cursorily or vaguely the home state ability to protect women (‘appropriate state protection’ or ‘state protection’) and the socio-cultural conditions affecting women’s possibility of relocating within the home country (‘internal relocation’). This Section argues that this thin assessment obscures the ways in which state and societal structures may heighten female applicants’ vulnerability to ill-treatment. One way in which the Court formalistically assesses state protection consists of inquiring whether there are laws prohibiting a certain form of ill-treatment regardless of whether these laws are actually enforced. An example of this kind of thin assessment is Collins and Akaziebie v Sweden.48 Emily Collins and her daughter, Ashley Akaziebie, were allegedly escaping from female circumcision in Delta State, Nigeria.49 The Court declared their application manifestly ill-founded for various reasons,50 including that Delta and several other Nigerian states prohibited FGM by law.51 Missing however from the Court’s assessment are reports on Nigeria showing that, once state FGM laws had been passed, local authorities still had to be convinced that these laws were applicable.52 A closer look at these reports would have alerted the Court that in some states, including Delta, the extent to which the police and other authorities had been enforcing the prohibition remained unclear.53 A similar formalistic assessment of appropriate state protection can be found in R.B.A.B. and Others v The Netherlands.54 A woman and a man unsuccessfully claimed before the Court that their daughters would be circumcised if expelled to Sudan.55 The Court asserted that some provinces, including that of the applicants, had passed anti-FGM laws to protect children.56 However, it did not look at country reports indicating that in practice there were no FGM prosecutions in Sudan.57 Nor did it take into account that, according to these reports, legal sanctions were ‘unlikely to be applied where a woman has been subjected by her family to FGM’.58 In another female circumcision case declared manifestly ill-founded, Izevbekhai and Others v Ireland, the Court noted that Nigeria has ratified the Maputo Protocol, prohibited inhuman treatment in the Constitution and passed state laws against FGM.59 Unlike in the two previous cases, the Court this time at least recognized ‘the absence or low level of legal action’ to enforce existing laws.60 It acknowledged the mixed views on ‘the potential for police support of women escaping FGM’.61 Yet the Court ultimately downplayed these findings by stressing the federal government’s public opposition to FGM and the work of some Ministries against it.62 What the cases discussed so far have in common is an over emphasis on the home state formal efforts to end female circumcision when assessing state protection (for example, bans, ratification of human rights instruments, government opposition).63 A state may indeed be making its ‘best efforts’ to combat a specific type of ill-treatment but these efforts signal willingness and not ability to protect.64 Refugee law scholars convincingly reject determining the adequacy of protection based merely on state ‘serious efforts’.65 This approach, they note, ‘tends to favour form over substance’ since it overlooks the actual effect of such efforts.66 Thus, formally outlawing a certain form of gendered ill-treatment should not automatically mean that the state is in fact able to offer appropriate protection.67 Due to deep-rooted socio-cultural factors, change may extend ‘for years beyond the promulgation of laws’.68 As UNHCR notes: ‘Even though a particular State may have prohibited a persecutory practice (e.g. female genital mutilation), the State may nevertheless continue to condone or tolerate the practice, or may not be able to stop the practice effectively.’69 Another way of formalistically examining state protection issues consists of inquiring whether the applicant turned to the authorities before leaving her country, without asking whether ‘that option was reasonable’.70 In Izevbekhai and Others, despite earlier doubts about the usefulness of police support, the Court reproaches the applicant for not having attempted ‘to report any issue concerning their daughters and FGM to the police’.71 According to the CEDAW Committee, however, the fact that a woman did not make a complaint to the authorities ‘should not prejudice her asylum claim’.72 Authorities may tolerate violence against women; they may systematically fail to respond to women’s complaints; or women may simply not have confidence in the justice system.73 Only after significant changes in police practice and prosecutions may women start filing cases in larger numbers.74 An analysis of the home state ability to offer appropriate protection should thus move beyond inquiring whether protection is theoretically available. It should inquire whether there are any obstacles that may make protection virtually inaccessible to or unsafe for women.75 Attention to actual state protection is especially important when cases concern women, as access to protection is often gendered.76 The Court’s thin assessment may not only be formalistic. In R.W. and Others v Sweden,77 a case declared manifestly ill-founded, the assessment of appropriate state protection remained vague. In this case a woman and her daughters failed to convince the Court that they would risk circumcision if deported to Kenya.78 Noting that the Kenyan authorities were taking ‘active measures’ to prevent FGM, the Court concluded that ‘there is no indication that the domestic authorities would be unwilling or unable to protect the applicant’.79 The reports referred to in the Court’s decision do however contain such an indication. They note that anti-FGM laws did not exist to protect adult women (like the applicant) but only women under eighteen years (like her daughters).80 According to these reports, the Kenyan government’s willingness and ability to protect women against FGM was increasing, but only slowly.81 Apart from legislation to protect minors against FGM—which had led to only a few prosecutions82—the ‘active measures’ the government was taking ultimately remain unclear. The Court skipped the assessment of appropriate state protection altogether in A.A. and Others v Sweden.83 The case involved claims by a woman and her five children (three daughters and two sons) allegedly escaping from a violent husband/father in Yemen.84 The oldest daughter additionally claimed that her father had married her to an older man who had mistreated her.85 One of the youngest daughters feared a similar fate, as her father had allegedly attempted to marry her to a man against her will.86 Though reports indicated that there were serious structural gendered deficits in state protection in Yemen,87 these deficits do not make it into the Court’s reasoning. This silence is remarkable because the Court did not rule out that AA and one of her daughters might have suffered abuse.88 Nor did it doubt the father’s alleged plans to marry off another daughter.89 The Court found, nonetheless, that the documents in support of some of the female applicants’ claims were not authentic or sufficient90 and that these applicants would have a male network to count on for protection upon return.91 Based on these and other reasons, it ultimately upheld the applicants’ deportation to Yemen.92 Gendered structures may similarly remain thinly scrutinized in the evaluation of whether a woman can relocate to another part of her home country. Take the female circumcision case of Izevbekhai and Others, discussed above. In assessing whether internal relocation was an option for women escaping violence in Nigeria, the Court cursorily noted that ‘the Federal Government provides direct protection to such women’.93 It did not acknowledge that the state protection structures facilitating women’s relocation remained weak according to country reports.94 The Court instead concentrated on the applicants’ ‘considerable familial and financial resources’ to conclude that they could have successfully relocated to the North of the country, where FGM was rarely practiced.95 In Omeredo v Austria, also concerning a Nigerian woman allegedly escaping female circumcision, the Court concluded that the applicant could successfully relocate to another part of the country even without family support.96 The reports on internal relocation for women in Nigeria do not even feature in the Court’s reasoning. These were the same reports as in Izevbekhai and Others.97 They indicated that government support to women fleeing FGM was progressing but remained weak98 and that women would need relatives able to accommodate them in the new location.99 Against this backdrop, it is difficult to understand how Ms Omeredo’s application meets an extreme form of rejection: manifestly ill-founded.100 A thin assessment of the gendered conditions influencing women’s relocation is also present in M.Y.H. and Others v Sweden,101 a case involving a woman, a man and their daughter allegedly escaping religious persecution in Iraq. The female applicants additionally claimed that, as Christian women, they would be at risk of sexual violence.102 The Court concluded that the applicants could safely relocate to the Kurdistan region of Iraq.103 In the analysis of whether relocation was an option the religious dimension eclipses the gender aspect of the case. The Court extensively examines the situation of Christians in the Kurdistan region104 but devotes only one of the final paragraphs to the female applicants’ position as Christian women in the country.105 In this one paragraph, the Court cursorily concluded that neither the applicants’ submissions nor the available country origin information indicated that they would be at risk of gender-related ill-treatment contrary to Article 3 of the ECHR.106 This conclusion, however, does not fully reflect the country information quoted earlier in the judgment. This information described sexual and gender-based violence as one of the specific concerns remaining in the Kurdistan region despite other improvements.107 Cases like Izevbekhai, Omeredo and M.Y.H. illustrate that the Court may not consistently (thoroughly) check the criteria outlined in its own case law for an internal relocation to be acceptable under Article 3 of the ECHR.108 In relocating to another area, women may face cultural and social constraints to travelling, living alone and making a living there.109 They may have work experience but still not be able to secure a job if socio-cultural conditions severely constrain women’s access to work.110 Difficulties in securing accommodation or financial survival without family support may be part of the gendered reality in the place of relocation.111 For these reasons, the safety of the route, the availability of state protection and the socio-cultural context in the area of relocation should all be carefully examined.112 In conclusion, assessing state protection and relocation conditions thinly obscures the societal and institutional structures that (re)produce female applicants’ vulnerability to ill-treatment. The thin assessment thus misses the ‘structural focus’ that scholars encouraging a more critical understanding of vulnerability have been arguing for.113 Fineman’s vulnerability theory advocates for considering the ‘structural and institutional arrangements in assessing the state’s response to situations of vulnerability’.114 Society’s institutions, she argues, may not only fail to reduce individuals’ vulnerability; they may sometimes operate to exacerbate it.115 A thin assessment of these arrangements also overlooks what Fineman calls ‘particular’ vulnerability.116 Though she believes that all human beings are inherently vulnerable, she recognizes that they may experience vulnerability more or less acutely, depending on how they are situated within society and state institutions.117 In the cases examined in this part, the Court misses the heightened ways in which societal and institutional structures may render many women vulnerable to ill-treatment. In missing these structures, the analysis may oversimplify a wider gender inequality problem118 and the risk that the individual applicant might face.119 4. THICK ASSESSMENT OF PRIVATE COPING CAPACITY: THE IMPLICIT ‘VULNERABLE VICTIM’ APPROACH The thin examination of the gendered structures in the country of origin is in some cases coupled with a thick consideration of the private capacity to cope with the risk. In these instances, the Court over emphasizes either the applicant’s assertiveness and resourcefulness to protect herself and/or the capacity of her male relatives to protect her. This part discusses some dangers inherent in over scrutinizing applicants’ personal circumstances in this way while under scrutinizing the institutional and socio-cultural structures influencing women’s relocation and the ability of the home state to protect them. A. Women’s Assertiveness and Resourcefulness120 The Court’s assessment of women’s personal circumstances is in some instances characterized by an over emphasis on their perceived strength, independence and resourcefulness to deal with the risk. Going to courts, making arrangements to leave the country and even applying for asylum may be taken as indications of women’s independence and strength to protect themselves or their children. Take, for example, the case of A.A. v Sweden discussed above, brought by the Yemeni family escaping a violent husband/father. Without inquiring into the level of state protection available in Yemen, the Court turns directly to A.A.: [T]he first applicant has shown proof of independence by going to court in Yemen on several occasions to file for divorce from X and also shown strength by managing to obtain the necessary practical and financial means to leave Yemen.121 The Court does not deny AA protection based solely on her perceived strength and independence. Her strength and independence are part of a set of factors that the Court takes into account, including credibility issues122 and, paradoxically, the presence of male relatives to protect her.123 The Court seems to attach more decisive weight to the woman’s strength and independence in Collins and Akaziebie v Sweden. In this case, as noted above, the Court under scrutinizes the structures undermining the home state ability to protect women who refuse to undergo circumcision. Ms Collins’ personal circumstances, on the other hand, are closely scrutinized: she was around 30-years-old; had gone to school for 12 years; opposed her daughter’s circumcision and secured the father’s support; managed to obtain the necessary financial means to leave Nigeria; successfully travelled to Sweden; and applied for asylum.124 The Court concludes: Viewed in this light, it is difficult to see why, as indicated by the Government, the first applicant, having shown such a considerable amount of strength and independence, cannot protect the second applicant from being subjected to FGM, if not in Delta State, then at least in one of the other states in Nigeria where FGM is prohibited by law and/or less widespread than in Delta State.125 Underlying these conclusions is the assumption that ‘if you can manage to get from Nigeria to Sweden, you can manage to protect your child from FGM’.126 As Brems puts it in a critique of the Court’s reasoning in this case, ‘if a woman is strong enough to stand up against cultural oppression, she is too strong for outsider protection’.127 Claiming rights in court or actively seeking international protection should not be held against women asylum seekers. Doing so implicitly punishes ‘rights-conscious’ women, that is to say, women who identify their entitlements and assert their rights.128 This may discourage women from bringing gendered ill-treatment cases to the attention of the legal system, in this case, of a supranational human rights court. In this way, it may actually compound the gendered access problem discussed in the previous part. Taking ‘rights-consciousness’ as a sign of a lessened need of protection implicitly perpetuates the gender stereotype that the deserving female victim is unaware of her legal entitlements. Judicial stereotyping of this sort is problematic because, in distorting judges’ perception of who is a victim, it acts as a barrier to justice.129 A brief comparison of A.A. and Collins and Akaziebie with the Court’s reasoning in other Article 3 ECHR cases concerning female asylum seekers suggests that women might be caught in a double bind. If they appear too diligent, they may be considered too strong and therefore less worthy of protection. If they appear too slow, they may not be considered genuinely fearful. Remaining in the home country for ‘some months even after separating from her husband’130 or waiting ‘three weeks’131 before applying for asylum have negatively affected some applicants’ credibility.132 The Court’s analysis of applicants’ personal circumstances in other instances over focuses on women’s resourcefulness, namely on their education, work experience and financial means.133 In Izevbekhai and Others, as previously discussed, the Court looks at the appropriate state protection only thinly while emphasizing that successful relocation depends most importantly on ‘favourable personal circumstances’.134 The Court points to Ms Izevbekhai’s family financial resources and to her ‘second and third level education and professional experience’.135 It then puzzles over why the family, notwithstanding these considerable resources, had not attempted to relocate to Northern Nigeria where the FGM rate was lower.136 The applicant’s resourcefulness, though not the sole factor, plays an instrumental role in the Court’s conclusion that she and her husband could protect their daughters from FGM in Nigeria.137 In Omeredo, a case also discussed earlier, the Court acknowledged the difficulties that the applicant might face in relocating as ‘an unmarried woman without support of her family’.138 However, it concluded that ‘owing to her education and working experience as a seamstress’ the applicant would not need such support to rebuild her life in another part of the country.139 Ms Omeredo’s education and work experience played out decisively in the Court’s conclusion in favour of her relocation. Attention to individual resourcefulness as part of the assessment of applicants’ personal circumstances is not in itself problematic. Financial resources, education and work experience may indeed play a role in lessening one’s vulnerability to ill-treatment. These are what Fineman, following Kirby, calls ‘physical’ and ‘human’ assets.140 As she argues, these assets ‘affect material well-being’ and ‘bolster individuals’ resilience in the face of vulnerability’.141 What raises concerns, however, is the focus on individual resourcefulness without an equal attention to the broader socio-cultural and institutional structures conditioning women’s enjoyment of state protection and relocation. As seen in the previous Section, a woman may have the best education but still not be able to secure a job if she relocates to an area where widespread discrimination excludes women from job opportunities. A woman may be wealthy but still fail to secure protection if the police systematically see domestic violence as a private affair. In Izevbekhai and Others and in Omeredo, the Court seems blinded to the fact that women’s high education, financial resources or professional experience may not be due to gender equality but despite inequality.142 Hiding from view the gendered roots of ill-treatment while over focusing on individual strength, independence and resourcefulness suggests that vulnerability to ill-treatment is ‘self-made’ and thus ‘ameliorable through individual self-improvement’.143 The suggestion is that women risk ill-treatment because they lack independence, education or financial resources and, therefore, can privately lessen the risk by improving personally, educationally and financially. Feminists have actually criticized how asylum adjudicators sometimes assess internal relocation ‘from the perspective of the resources and opportunities available to the asylum applicant, rather than through the scrutiny of the actions of the State’.144 Moreover, over focusing on women’s strength, independence and resourcefulness runs the risk of reducing women’s vulnerability to ill-treatment to negative stereotypes.145 The assumption is that the deserving female victim is vulnerable to mistreatment because she is unassertive, dependent and uneducated. The Court does not explicitly describe her in these terms. In fact, it does not expressly call her ‘vulnerable’. Yet these assumptions underlie the Court’s reasoning in the cases discussed so far in this sub-section and are further confirmed in a case unexamined up until now: Sow v Belgium.146 Sow was reportedly escaping a second genital circumcision in Guinea. The Court dismissed her complaint for several reasons, including the fact that a 28-year-old woman who had a progressive education and opposed female circumcision cannot be considered ‘a particularly vulnerable young woman’ (emphasis added).147 In reaching this conclusion, the Court referred to Collins and Akaziebie and Izevbekhai and Others.148 These are two of the cases where women’s strength and resourcefulness are implicitly taken as signals of a lessened need of protection. In leaving no room for the more nuanced and realistic coexistence of victimization and agency,149 this account of vulnerability may exclude from protection women who do not fit ‘ideal’ female victim behaviour (that is, unassertive, dependent and passive).150 It may thus stack the odds against non-Western female applicants who resemble the stereotype of Western women as independent and educated or who are not ‘womanly enough’, according to stereotypes of non-Western women. As Razack puts it referring to women asylum seekers in Canada: ‘If the woman betrays too much personal strength, decision-makers can assume that her story of being victimized does not ring true because she appears to be someone who can protect herself.’151 One of the clearest examples in the Court's jurisprudence of how the applicant’s story did not ring true partly because she betrayed too much independence is I.F.W. v Sweden.152 The case, declared manifestly ill-founded on credibility grounds, involved an Iraqi woman allegedly fleeing violence from her male relatives for having dishonoured them as a result of a relationship with a man.153 Since she had run ‘her own business in Iraq’ and led ‘an independent life’, the Court finds it questionable that ‘her family has such a strong honour culture that an illegitimate relationship would lead to a risk of her being killed’.154 The gender stereotypes flowing from the implicit ‘vulnerable victim’ approach resonate with essentialist constructions of non-Western women as passive victims, which post-colonial feminists155 have condemned in international human rights law.156 As Otto has shown, one of the recurring female subjects in this body of international law has been the ‘victim subject’.157 This female subject has been produced in contrast to the male subject whose dominance she sustains: ‘the masculine bearer of “civilization” who rescues “native” women from “barbarian” men’.158 Otto thus shows how stereotypes of women as helpless victims needing protection have gone hand-in-hand with stereotypes of men as ‘protectors, supporters and saviours’ and have served to recreate women’s inferiority to men in international law.159 As will now be discussed, in a few cases, the Court has implicitly recreated these gender stereotypes of men as well. B. Women’s Male Protection Network In the majority of the cases discussed up until to now, the Court highlights women’s own capacity to deal with the risk either in the place where they come from or in other areas of their countries of origin. In the cases examined in this sub-section, the Court looks at the presence or absence of male relatives to protect women. In three of the 29 cases surveyed in this article,160 the Court focuses on whether a ‘male protection network’ is available in the home country, probably prompted by arguments made by both governments and applicants.161 In examining two of these cases (the third one will be discussed in the last part), this sub-section argues that relying on a ‘male protection network’ is fraught with perils. One of these perils is underestimating the degree of risk in cases where applicants do have male relatives in the home country. An illustration of this peril is paradoxically A.A. and Others, the case in which the Court also emphasised the woman’s strength and independence to protect herself. In looking at the family support AA would have in Yemen, the Court turned primarily to her brother and to her adult sons.162 There is some evidentiary basis for the Court’s conclusion that the brother would support her.163 However, it is difficult to find a basis for the conclusion that the sons ‘would enable her to live away from her husband’.164 The Court reached a similar conclusion with respect to one of A.A.’s daughters: ‘[She] would be accompanied by her two brothers, with whom she left Yemen and, thus, she would have a male network and be able to live away from her husband and father.’165 Rather than on evidence, the Court’s findings seem to be based on the implicit gender stereotype that men are women’s protectors. The subtext is that the sons/brothers would protect the female family members merely by virtue of being men and adults. The danger of over simplifying the risk is also exemplified by R.H. v Sweden.166 RH unsuccessfully claimed before the Court that, if returned to Somalia, she would be either forced to marry or killed by her uncles for refusing to marry against her will.167 She additionally alleged that she would risk being sexually assaulted given the severe conditions for Somali women who lacked male relatives.168 Unlike in A.A., the Court generously examined the structural state protection and socio-cultural conditions underlying violence against women in Somalia.169 Yet, in its conclusion of the country situation, the Court simplistically noted that ‘a single woman returning to Mogadishu without access to protection from a male network would face a real risk of living in conditions’ contrary to Article 3 of the ECHR.170 It concluded that RH would not return to Mogadishu ‘as a lone woman with the risks that such a situation entails’,171 as she had a brother and uncles and, therefore, a ‘male protection network’.172 Thus the Court assumed that the male relatives would protect her without considering country reports indicating that family members were among the actors committing violence against women with impunity in Somalia.173 Nevertheless, family support upon return to the home country may be an important consideration, especially where the evidence shows that applicants would risk their lives or their physical integrity without such support. As a network of relationships from which people often get help, family may be a ‘social asset’ in the face of vulnerability.174 Examining the support male relatives may offer is therefore not in principle problematic. In fact, applicants may have only male relatives left in their countries of origin. In some cases, there might be evidence that such relatives have a close relationship with the applicant and are able to offer actual support. Yet, as many of the Strasbourg Court cases herein examined illustrate, the family may not necessarily be a safe space for women. It is therefore problematic to simply assume that the mere presence of male relatives automatically equals women’s protection. In A.A. and Others, there seem to be no factual indications that this will be the case of AA’s adult sons. In R.H., there are some signs that this may not necessarily be the case of the applicant’s uncles. The danger therefore arises when the Court relies on assumptions or generalizations that equate maleness with capacity to protect. As argued above, reliance on these assumptions or generalizations may distort the risk assessment in the individual case: male relatives in actual fact may not be able to protect or may represent a threat themselves. Irrespective of the potentially negative implications for the particular case, relying on such assumptions raises more principled objections. The assumptions suggest that state protection can or should be substituted by (a patriarchal form of) private protection.175 The suggestion is difficult to reconcile with the Court’s own Article 3 ECHR principle requiring that the ‘the authorities of the receiving State are not able to obviate the risk by providing appropriate protection’.176 The ‘male protection network’ cases raise yet a more fundamental challenge, one of enduring resonance in international human rights law: how to recognize a gendered reality without reviving gender stereotypes and re-inscribing women’s subordinate status in the law.177 The Court falls into these traps when it frames the risk in terms of presence or absence of a ‘male protection network’. The framing gives the impression that the risk narrowly arises from the lack of a male protector rather than from societal attitudes hostile to the status of single women. The fact that women may need male company to lead their daily lives safely should actually point to these societal root causes. Moreover, the need for patriarchal private protection signals a state’s failure to protect.178 In obscuring these societal and institutional factors, the ‘male protection network’ frame turns the capacity to protect and the need of protection into inherent attributes of men and women. It thus revives the stereotypes of (non-Western) women as needing protection and men as protectors that, as seen above, have tenaciously underpinned international (human rights) law.179 What is more, in associating the dominant side of the dichotomy (protector) with men and the subordinate side (protected) with women, the frame reproduces men’s authority and women’s dependency in human rights discourse. Feminists have challenged the gender hierarchies produced by these types of dichotomies in international human rights law for failing to recognize women as fully human.180 The notion of ‘male protector’ has been, moreover, criticized by refugee law scholars precisely for entrenching ‘perceptions of women as inferior, vulnerable on the basis of their sex, and unable to survive without male family members’.181 To conclude this Section, the ‘assertiveness/resourcefulness’ rationale and the ‘male network’ rationale may at first sight evoke contradictory images of women. The former celebrates women’s independence to protect themselves; the latter emphasizes women’s dependency on male protection. A closer look, however, reveals that these rationales are ultimately part of the same troubling narrative. They implicitly sustain a ‘vulnerable victim’ approach according to which women suffer ill-treatment because of their personal failures/limits rather than socio-cultural and institutional failures/constraints. The limits/failures problematically attributed to women, in turn, echo the stereotypes of (non-Western) women that feminists have combated in international human rights law for recreating gender hierarchies. 5. VULNERABILITY IN CONTEXT: INTERWEAVING THE STRUCTURAL AND THE INDIVIDUAL This Section argues for assessing the risk of gendered ill-treatment structurally and relationally in order to counter the dangers identified in the previous Sections. It discusses four cases in which the Court moves in that direction. In one of them, there are hints that the Court might have given a more substantive consideration to the kind of state protection available in the applicant’s home country. In the other three cases, there is express attention to the gendered structures that may (re)produce women’s vulnerability to ill-treatment. Except for one of the four cases,182 the Court stays clear of the ‘vulnerable victim’ approach. The Court’s reasoning in the first case does not inquire into the gendered roots of the alleged risk of ill-treatment. Yet it signals a less formalistic standard in assessing the level of state protection. In examining whether the home state authorities would protect the applicant in Ayegh v Sweden, the Court does not just look at whether ‘the battering of women is a criminal offence’ in Iran.183 The Court also noted that the Iranian authorities examine such allegations.184 It is difficult to know whether the domestic authorities actually investigated women’s allegations, as there are no country reports mentioned in the Court’s decision. The Court’s statement suggests, nonetheless, a concern that goes beyond formal questions such as whether a certain form of ill-treatment has been legally banned. Mahin Ayegh claimed, among other things, that she would risk violence from her husband, who had accused her of leaving their home in Iran and being unfaithful to him in Sweden.185 The Court declared her application manifestly ill-founded.186 In the other three cases, the Court does pay attention to the gendered structures influencing the ill-treatment of women. These structures include serious institutional deficits in state protection (for example, impunity for violence against women) and harsh socio-cultural conditions (for instance, widespread discrimination against women). The Court acknowledged these sorts of background conditions in R.D. v France, a case brought by a Guinean Muslim woman fearing violence by her father and brothers for having married a Christian man.187 In assessing the state protection in Guinea, the Court took into consideration the available country reports.188 These reports highlight that violence and discrimination against women are among the most serious problems in the country and that few victims report violence due to perceptions of police ineffectiveness.189 The Court finds that the applicant’s deportation would violate Article 3 of the ECHR,190 partly because the Guinean authorities were not in a position to protect women in situations like hers.191 An illustration of a truly robust examination of the gendered structures underlying both the state inability to protect and the socio-cultural context is N. v Sweden,192 in which the Court found a violation of Article 3 of the ECHR. N was an Afghan woman escaping reprisals from her ex-husband, her own family and her society for having separated from her husband and started a relationship with a Swedish man.193 She further claimed that she would face inhuman treatment in Afghanistan given that her family had disowned her and, therefore, left her with ‘no social network or male protection’.194 Taking its cue from UNHCR reports, the Court recognized the wider context of socio-cultural conditions for women in Afghanistan, which included violence and punishment for not conforming to gender roles.195 Against this backdrop, the Court noted that N may be perceived in her society as not conforming to such roles for having attempted to end her marriage and living abroad.196 When assessing whether she would enjoy appropriate state protection in Afghanistan against possible violence from her husband, the Court emphasized the kind of gendered structures overlooked in the cases examined in Section 3: Afghan authorities do not prosecute violence against women197 and many Afghan women do not even seek help for fear of police abuse or retaliation by perpetrators.198 The focus on state protection deficits and societal constraints, such as those present in R.D. and N., brings into the assessment the gendered structures shaping the risk of ill-treatment. Bringing in these structures does not only facilitate a fuller and more accurate evaluation of the individual risk. It also avoids reducing a gender inequality problem to an idiosyncratic problem, attributable to a woman’s personal failures/limitations. Structurally understood, then, the risk of ill-treatment is shaped by entrenched factors over which the applicant and many other women may not have control. Shedding light on the institutional and societal roots of ill-treatment further encourages thinking outside the stereotypes that women are at risk because they lack education or male protection.199 In examining the scenario of N living alone if deported to Afghanistan, the Court denaturalizes the male-protector and female-protected stereotyped roles precisely by foregrounding the societal factors that construe these roles. Reproducing the UNHCR’s statement, it does acknowledge that women lacking male company face serious limitations on their freedom of movement in Afghanistan.200 However, and again following UNHCR, the Court sheds light on the socially constructed character of the need for a ‘male tutor’: discrimination against women, social rejection of single women, social stigma attached to their status and social restrictions on women living alone.201 Framed this way, the risk is more complexly understood as flowing from gendered societal structures and state protection failure rather than stereotypically from the lack of a male companion. Bringing to light structural issues sometimes may not be enough to circumvent the dangers discussed in previous Sections. Recognition of the gendered and other structures in a certain country may be thorough, but ultimately it does not enter into a legal consideration of the applicant’s individual circumstances. The Court generously recognized the gendered state protection and socio-cultural context in R.H. v Sweden, the case discussed in Section 4.202 Yet when the time came to assess the risk of returning the applicant to Somalia as a single woman, the Court overlooked this context. As seen earlier, the Court ended up framing the risk in simplistic terms: lack of male protection.203 It then narrowly assessed the risk by merely asking whether RH had male relatives in Somalia.204 The Court’s thick recognition of the severe conditions of vulnerability for women in the applicant’s home country ultimately remained theoretical. The key to avoiding the shortfall illustrated by R.H. lies in examining the structural and the individual relationally. In an attempt to challenge the individualistic conception of the human self in liberal thinking205 and human rights discourse,206 several feminists have called for understanding the self relationally. In essence, the argument is that individuals are embedded in a web of relations that may enhance or undermine their autonomy.207 These relations include not only intimate (for example, family) and structural relations (for example, informal gender norms), they also include relations with the state.208 Relevant for present purposes is that these relational accounts of the self and autonomy look at the individual in her social context and pay attention to the role of this context in shaping individual autonomy. In particular, these accounts encourage investigating how structural relations and relations with the state affect the individual capacity in the face of vulnerability. A relational approach is therefore useful to avoid locating an applicant’s vulnerability solely either in the individual or in the structural. Her vulnerability to ill-treatment is understood as arising from an interaction between the two. A relational assessment of the risk of ill-treatment would thus require evaluating the applicant’s capacity to deal with the risk in connection with the socio-institutional forces enabling or disabling her capacity. The structures within which she makes choices may render her vulnerable to ill-treatment regardless of how strong and independent she may be.209 In summary, in order to counter the dangers discussed in this article, the risk of ill-treatment should be assessed structurally and relationally. The former implies bringing under fuller scrutiny the gendered structures shaping this risk. The latter requires assessing how such structures may affect the applicant’s capacity to deal with the risk. This does not mean that the Court should find an Article 3 ECHR violation in all cases where state institutional structures are precarious and discrimination against women widespread. This simply means that the assessment of a ‘real risk’ should be guided by a thorough analysis of the full set of structural and individual factors.210 The Court’s over emphasis on certain aspects of applicants’ personal circumstances as well as its under emphasis on structural issues might be motivated by floodgate concerns. So might be the Court’s narrowing of the category of single women at risk to those without a male protector.211 The Court may not want to encourage all single women, let alone all women, living in conditions of systemic gender-based violence and discrimination to come to Council of Europe Member States. Yet a thin examination of these conditions is difficult to reconcile with the rigorous assessment of the risk that the absolute nature of the Article 3 of the ECHR prohibition demands.212 As Brems argues in commenting on the Court’s decision in Collins and Akaziebie: The Court might be afraid to encourage the 3 million girls and women who are annually at risk of FGM to flee to Europe. Yet you cannot blow hot and cold at the same time. If forcing someone to undergo FGM is a violation of article 3, the same rules should apply as in other cases of this type. Women claiming to flee FGM should get a fair assessment of the ‘real risk’ involved.213 6. CONCLUSION In this article I have proposed rethinking the assessment of women asylum seekers’ claims of ill-treatment. In encouraging a contextualized assessment of the risk, I hope to have opened up the possibility of gendering the non-refoulement analysis under Article 3 of the ECHR.214 Careful attention to the de jure and de facto context in the country of origin encourages ‘seeing’ gender inequalities in state institutions and society and assessing how these inequalities may condition a woman’s capacity to ensure state protection. Attention to context, moreover, escapes reinforcing stereotypical views of (non-Western) women as vulnerable victims who lack agency and depend on male protectors. Viewed in context, vulnerability to ill-treatment is institutionally and societally shaped and not an innate characteristic of certain groups. As Crawley argues, rather than as ‘victims of “private” male violence’, women should be viewed as holders of rights that may be ‘negated or undermined by patriarchal structures and institutions’.215 Only by digging into the depths of particular contexts, can ECHR law unearth and truly assess the gendered roots of abuses women flee from. ACKNOWLEDGEMENT I wrote this article while I was a visiting scholar at Melbourne Law School. The article has benefited from generous comments made by Dianne Otto, Eva Brems, Karin Chinnian and Corina Heri. I am also grateful to an anonymous reviewer for invaluable feedback and to Giselle Corradi and Valeska David for insightful conversations. A draft of this article was presented at the Critical Legal Conference held at Kent Law School on 1 to 3 September 2016. I thank the participants for their constructive comments. Footnotes 1 Foster, ‘Why We Are Not There Yet: The Particular Challenge of “Particular Social Group”’ in Arbel, Dauvergne and Millbank (eds), Gender in Refugee Law: From the Margins to the Centre (2014) 17. 2 See, for instance, Mullally, ‘Domestic Violence Asylum Claims and Recent Developments in International Human Rights Law: A Progress Narrative?’ (2011) 60 International and Comparative Law Quarterly 459. 3 Arbel, Dauvergne and Millbank, ‘Introduction’ in Arbel, Dauvergne and Millbank (eds), Gender in Refugee Law: From the Margins to the Centre (2014) 1 at 11. 4 See, for instance, Chapter VII ‘Migration and Asylum’, Council of Europe Convention on Preventing and Combating Violence against Women and Domestic Violence 2011, CETS 210 (‘Istanbul Convention’) and Committee on the Elimination of Discrimination against Women , General Recommendation No 32: Gender-related Dimensions of Refugee Status, Asylum, Nationality and Statelessness of Women (2014). 5 Article 3 ECHR reads: ‘No one shall be subjected to torture or to inhuman or degrading treatment or punishment.’ See, for instance, M.S.S. v Belgium and Greece Application No 30696/09, Merits and Just Satisfaction, 21 January 2011 (Grand Chamber) at para 286. 6 Fineman, ‘The Vulnerable Subject: Anchoring Equality in the Human Condition’ (2008-2009) 20 Yale Journal of Law and Feminism 1. 7 See, for instance, Nedelsky, Law’s Relations: A Relational Theory of Self, Autonomy, and Law (2012). 8 See, for instance, Otto, ‘Lost in Translation: Re-Scripting the Sexed Subjects of International Human Rights Law’ in Orford (ed.), International Law and Its Others (2006) 318; Kapur, ‘The Tragedy of Victimization Rhetoric: Resurrecting the “Native” Subject in International/Postcolonial Feminist Legal Politics’ (2002) 15 Harvard Human Rights Journal 1. 9 See, for instance, Rengifo Alvarez v The Netherlands Application No 14232/07, Admissibility, 6 December 2011 (political); N.M.Y. and Others v Sweden Application No 72686/10, Merits and Just Satisfaction, 27 June 2013 (religion). 10 See, for instance, S.S. and Others v Denmark Application No 54703/08, Merits and Just Satisfaction, 20 January 2011 (political and ethnic). 11 The selected case law is the result of an Article 3 ECHR search in the HUDOC database using the following key words in English: ‘asylum’ and ‘women’ as well as ‘asylum’ and ‘woman’. The data set includes judgments and decisions of the European Court of Human Rights up until 30 June 2016 involving women who had unsuccessfully claimed asylum at the domestic level in their own right, alone or together with family members. The following cases have been excluded from the data set: Dublin transfers, applications struck out of the list, joint applications when the alleged risk did not stem from women’s activities but primarily from co-applicants’ activities and applications decided by the now defunct European Commission of Human Rights. The inadmissibility decisions included in the data set comprise only those declaring applications manifestly ill-founded. The selected cases have been additionally identified through existing literature, the snowball method (some cases lead to others) and bi-weekly team discussions of the Court’s latest case law at the Human Rights Centre at Ghent University. 12 The gendered ill-treatment cases sometimes also involve other grounds. See, for instance, M.Y.H. and Others v Sweden Application No 50859/10, Merits and Just Satisfaction, 27 June 2013 (gender and religion). 13 These understandings are inspired by Crawley’s distinction in the context of refugee law between what she calls ‘gender-related persecution’ and ‘gender-specific persecution’: see Crawley, Refugees and Gender: Law and Process (2001) at 7–8. 14 This does not mean that these harms are not inflicted on men: see Sivakumaran, ‘Male/Male Rape and the “Taint” of Homosexuality’ (2005) 27 Human Rights Quarterly 1274. 15 Following Crawley, ‘gender’ is here understood as ‘the social construction of power relations between women and men, and the implications of these relations for women’s (and men’s) identity, status and roles’: see Crawley, supra n 13 at 6–7. 16 Ibid. at 8. 17 See, for instance, Achmadov and Bagurova v Sweden Application No 34081/05, Admissibility, 10 July 2007, at para 3 (alleged rape in connection with religious activities). 18 An example could have been Jabari v Turkey Application No 40035/98, Merits and Just Satisfaction, 11 July 2000. The woman’s sole claim concerned fears of stoning and flogging for alleged adultery in Iran. The case, however, has been excluded from the list of gendered ill-treatment cases discussed in this article because, according to information available in the judgment, Iranian adultery laws also applied to men: see ibid. at paras 31–32. Cases brought by Iranian women claiming a risk of cruel punishment for adultery together with other forms of ill-treatment such as domestic violence are examined in this article. 19 See, for instance, Ayegh v Sweden Application No 4701/05, Admissibility, 7 November 2006; and S.B. v Finland Application No 17200/11, Admissibility, 24 June 2014. 20 The domestic authorities had not interviewed the applicant and had failed to meaningfully examine her request for asylum: see Ahmadpour v Turkey Application No 12717/08, Merits and Just Satisfaction, 15 June 2010, at para 38. 21 N. v Sweden Application No 23505/09, Merits, 20 July 2010; R.D. v France Application No 34648/14, Merits and Just Satisfaction, 16 June 2016. 22 Samina v Sweden Application No 55463/09, Merits and Just Satisfaction, 20 October 2011; A.A. and Others v Sweden Application No 14499/09, Merits and Just Satisfaction, 28 June 2012; M.Y.H. and Others, supra n 12; A.A. and Others v Sweden Application No 34098/11, Merits and Just Satisfaction, 24 July 2014; R.H. v Sweden Application No 4601/14, Merits and Just Satisfaction, 10 September 2015; Sow v Belgium Application No 27081/13, Merits and Just Satisfaction, 19 January 2016; R.B.A.B. and Others v The Netherlands Application No 7211/06, Merits and Just Satisfaction, 7 June 2016. 23 Kaldik v Germany Application No 28526/05, Admissibility, 22 September 2005; Dejbakhsh and Mahmoud Zadeh v Sweden Application No 11682/04, Admissibility, 13 December 2005; Bello v Sweden Application No 32213/04, Admissibility, 17 January 2006; Ayegh, supra n 19; S.A. v The Netherlands Application No 3049/06, Admissibility, 12 December 2006; Collins and Akaziebie v Sweden Application No 23944/05, Admissibility, 8 March 2007; Achmadov and Bagurova, supra n 17; S.M. v Sweden Application No 47683/08, Admissibility, 10 February 2009; I.N. v Sweden Application No 1334/09, Admissibility, 15 September 2009; Izevbekhai and Others v Ireland Application No 43408/08, Admissibility, 17 May 2011; Ameh and Others v United Kingdom Application No 4539/11, Admissibility, 30 August 2011; Omeredo v Austria Application No 8969/10, Admissibility, 20 September 2011; H.N. and Others v Sweden Application No 50043/09, Admissibility, 24 January 2012; R.W. and Others v Sweden Application No 35745/11, Admissibility, 10 April 2012; I.F.W. v Sweden Application No 68992/10, Admissibility, 9 October 2012; Muradi and Alieva v Sweden Application No 11243/13, Admissibility, 25 June 2013; F.N. v United Kingdom Application No 3202/09, Admissibility, 17 September 2013; S.B., supra n 19; and M.M.R. v The Netherlands Application No 64047/10, Admissibility, 24 May 2016. 24 Kaldik, supra n 23 at 9; Dejbakhsh and Mahmoud Zadeh, ibid. at 10; S.A., ibid. at 6; Bello, ibid. at 6–7; S.M., supra n 23 at 10; I.N., ibid. at paras 33-38; Samina, supra n 22 at paras 62–64; I.F.W., ibid. at 4–5; A.A. and Others (2014), supra n 22 at paras 61–64 and 68. 25 See, for instance, N., supra n 21. 26 See, for instance, Sow, supra n 22. 27 See, for instance, Omeredo, supra n 23. 28 Heterosexual men, too, may experience ill-treatment for flouting gender expectations and defying patriarchal power. See, for instance, D.N.M. v Sweden Application No 28379/11, Merits and Just Satisfaction, 27 June 2013. Asylum seekers’ claims based on sexual orientation and gender identity should also be viewed as gendered: see UNHCR, Guidelines on International Protection: Gender-Related Persecution within the Context of Article 1A(2) of the 1951 Convention and/or its 1967 Protocol relating to the Status of Refugees, 7 May 2002, HCR/GIP/02/01, at para 16. An example of an Article 3 ECHR claim based on sexual orientation is M.E. v Sweden Application No 71398/12, Strike Out, 8 April 2015 (Grand Chamber). 29 See, for instance, Otto, ‘Introduction’ in Otto (ed.), Gender Issues and Human Rights (2013). On the initial marginalization of women in international refugee law, see, for instance, Edwards, ‘Transitioning Gender: Feminist Engagement with International Refugee Law and Policy 1950-2010’ (2010) 29 Refugee Survey Quarterly 21 at 22–3. 30 Charlesworth, ‘Feminist Methods in International Law’ (1999) 93 American Society of International Law 379 at 387–8. 31 Ibid. at 388. 32 See, for instance, Romany, ‘Women as Aliens: A Feminist Critique of the Public/Private Distinction in International Human Rights Law’ (1993) 6 Harvard Human Rights Journal 87; and Oxford, ‘Where Are the Women?’ in Arbel, Dauvergne and Millbank (eds), Gender in Refugee Law: From the Margins to the Centre (2014) 157 at 160–1. 33 See, for instance, CEDAW Committee, General Recommendation No 19: Violence against women (1992) at para 9. 34 UNHCR, supra n 28 at para 19. 35 See, for instance, Opuz v Turkey Application No 33401/02, Merits and Just Satisfaction, 9 June 2009, at para 161. 36 See, for instance, M.C. v Bulgaria Application No 39272/98, Merits and Just Satisfaction, 4 December 2003. 37 See, for instance, Izevbekhai and Others, supra n 23 at para 73. 38 Salah Sheekh v The Netherlands Application No 1948/04, Merits and Just Satisfaction, 11 January 2007, at para 137. 39 H.L.R. v France Application No 24573/94, Merits and Just Satisfaction, 29 April 1997 (Grand Chamber), at para 40. 40 See, for instance, S.A. v Sweden Application No 66523/10, Merits and Just Satisfaction, 27 June 2013, at para 53. On the impact of internal relocation on women, albeit in another context, see Bennett, Relocation, Relocation: The Impact of Internal Relocation on Women Asylum Seekers (Asylum Aid, 2008), available at: www.refworld.org [last accessed 12 March 2018]. 41 Ahmadpour, supra n 20 at para 38. 42 Kaldik, supra n 23; S.A., supra n 23; Bello, supra n 23; S.M., supra n 23; I.N., supra n 23; Samina, supra n 22; and A.A. and Others (2014), supra n 22. 43 Dejbakhsh and Mahmoud Zadeh, supra n 23; and I.F.W., supra n 23. 44 Due to the limited space available, I am unable to examine all 19 cases in full. 45 See, for instance, J.K. and Others v Sweden Application No 59166/12, Merits and Just Satisfaction, 23 August 2016 (Grand Chamber) at para 86. 46 N.A. v United Kingdom Application No 25904/07, Merits and Just Satisfaction, 17 July 2008, at para 119 (‘As regards the general situation in a particular country, the Court has often attached importance to the information contained in recent reports from independent international human rights protection organisations such as Amnesty International, or governmental sources, including the US State Department’). 47 J.K. and Others, supra n 45 at para 86. 48 Supra n 23. 49 Ibid. at para 10. 50 Other reasons included doubts about the veracity of the applicants’ story, Emily Collins’ capacity to protect her daughter and the decline of the FGM rate in the country as a whole: see ibid. at paras 12–14. 51 Ibid. at para 12. 52 Ibid. at para 8. 53 Ibid. at para 9. 54 Supra n 22. 55 Ibid. at para 33. 56 Ibid. at para 55. 57 Ibid. at paras 25–26. 58 Ibid. at para 30. 59 Supra n 23 at para 74. 60 Ibid. at para 75. 61 Ibid. 62 Ibid. 63 See also Muradi and Alieva, supra n 23. An Azerbaijani woman claimed (at para 5), inter alia, that if returned to Azerbaijan she would be forced into prostitution. The Court (at para 42) pointed to organizations in Baku providing shelters to women seeking protection. However, it remained silent about the serious deficiencies of these shelters. See ‘Wave Country Report 2011, Reality Check on European Services for Women and Children Survivors of Violence’, at 46–9, referred to in the judgment. Another example of the formalistic ‘best efforts’ approach to assessing state protection seems to be S.B., supra n 19. A Moroccan woman alleged, inter alia, that she was at risk of violence by her father. The Court noted (at para 36) that the Moroccan police had ‘a special unit for investigating domestic violence’ and that a ‘shelter system’ for women was in place. The existence of a police unit, however, does not automatically mean that the police are in actual fact able to protect women from violence and the existence of shelters does not mean that they offer enough places for victims. 64 Hathaway and Foster, The Law of Refugee Status, 2nd edn (2014) at 310–11. 65 Ibid. at 310–19. 66 Ibid. at 311. 67 Ibid. at 312. 68 Ibid. 69 UNHCR, supra n 28 at 11. 70 Cf. S.A., supra n 40 at para 50: ‘What can be expected of him is to have turned to the authorities if indeed that option was reasonable.’ 71 Supra n 23 at para 80. 72 CEDAW Committee, supra n 4 at para 29. 73 Ibid. 74 Engle Merry, Human Rights and Gender Violence: Translating International Law into Local Justice (2006) at 188. 75 See, for instance, Lobo, ‘Women As a Particular Social Group: A Comparative Assessment of Gender Asylum Claims in the United States and United Kingdom’ (2011–2012) 26 Georgetown Immigration Law Journal 361 at 392–3. 76 Women may be less able than men to obtain state protection: see Crawley and Lester, Comparative Analysis of Gender-related Persecution in National Asylum Legislation and Practice in Europe (United Nations High Commissioner for Refugees, 2004) at 232. See also CEDAW Committee, General Recommendation No 33: Women’s Access to Justice (2015) at para 3. 77 Supra n 23. 78 Ibid. at paras 2–3. 79 Ibid. at para 9. 80 Ibid. at para 5. 81 Ibid. 82 Ibid. 83 Supra n 22. 84 Ibid. at paras 9–13. 85 Ibid. at para 13. 86 Ibid. at para12. 87 The law protecting women against violence was rarely enforced; domestic violence was not specifically prohibited; and a minimum age for marriage was not established. Violence against women was in fact deemed a family affair and therefore usually went unreported to the police. Women faced social, economic and financial obstacles in getting divorced and difficulties in obtaining justice, as courts and police stations were not considered appropriate places for women. See ibid. at paras 39–44. 88 Ibid. at paras 81 and 87. 89 Ibid. at paras 91–92. 90 Ibid. at paras 81 and 92. 91 See critique of the Court’s reliance on female applicants’ male network below at Section 4. 92 A.A. and Others (2012), supra n 22 at para 96. 93 Izevbekhai and Others, supra n 23 at para 75. 94 Ibid. at para 49. 95 Ibid. at para 80. See discussion below at Section 4. 96 Supra n 23 at para 5. 97 From the Court’s reference (ibid. at paras 3–4) to the reports available in the Izevbekhai decision, one can infer that the same country information on women’s relocation in Nigeria was available to the Court in Omeredo. 98 Izevbekhai and Others, supra n 23 at para 51. 99 Ibid. at para 49. 100 Omeredo, supra n 23 at para 5. 101 Supra n 12. 102 Ibid. at para 43. 103 Ibid. at para 73. 104 Ibid. at paras 63–67. 105 Ibid. at para 71. 106 Ibid. 107 Ibid. at para 21. 108 The Court has held that the person must be able to travel, be admitted and be able to settle in the new area: see, for instance, J.K. and Others, supra n 45 at para 82. 109 CEDAW Committee, supra n 4 at para 28. 110 See Bennett, supra n 40 at 69–70. 111 CEDAW Committee, supra n 4 at para 28. 112 Council of Europe Parliamentary Assembly, Resolution 1765 (2010), Gender-related Claims for Asylum, 8 October 2010, at para 10.5. 113 Fineman, supra n 6 at 18. 114 Ibid. 115 Ibid. at 12–13. 116 Fineman, ‘Vulnerability, Resilience and LGBT Youth’ (2014) Temple Political and Civil Rights Law Review 101 at 110–13. 117 Ibid. 118 Ill-treatment of women is understood as part of the wider problem of gender inequality in international human rights law. See, for instance, CEDAW Committee, supra n 33 at para 11 and Preamble to Istanbul Convention, supra n 4. 119 On UK migration judges failing ‘to acknowledge and engage in gender issues and over-simpli[fying] the risks on return many women may face’: see Bennett, supra n 40 at 35. 120 I use ‘resourcefulness’ inspired by Bennett, ibid. at 30 and 80. 121 A.A. and Others (2012), supra n 22 at para 83. For a similar critique of this part of the reasoning in A.A. and Others, see Spijkerboer, ‘European Sexual Nationalism: Refugee Law after the Gender and Sexuality Critiques’, Nordic Asylum Seminar, University of Uppsala, 8 May 2015, available at: www.thomasspijkerboer.eu [last accessed 12 March 2018]. 122 Ibid. at para 81. 123 Ibid. at para 83. 124 Collins and Akaziebie, supra n 23 at para 13. 125 Ibid. at para 14 [emphasis added]. 126 Brems, ‘Strong Women Don’t Need Asylum (the European Court on FGM)’, Strasbourg Observers, Blog, 19 August 2010, available at: www.strasbourgobservers.com [last accessed 12 March 2018]. 127 Ibid. 128 Engle Merry, supra n 74 at 179, 184–8. I am thankful to Giselle Corradi for this point. 129 Cusack, Eliminating Judicial Stereotyping: Equal Access to Justice for Women in Gender-based Violence Cases, Final Paper submitted to the Office of the High Commissioner for Human Rights, 9 June 2014, at ii. See also CEDAW Committee, supra n 4 at para 31. 130 Ameh and Others, supra n 23 at para 13. 131 Dejbakhsh and Mahmoud Zadeh, supra n 23 at para 10. 132 The Court makes negative credibility inferences (ibid. at para 5) without investigating the reasons that might have prevented the applicant from applying earlier. For a critique of this sort of inference, see Hathaway and Foster, supra n 64 at 97–9. See also Crawley, supra n 13 at 210–11. 133 For an analysis of women’s perceived resourcefulness and adjudicators’ similar assumptions about their capacity to relocate internally, albeit in the context of asylum in the United Kingdom, see Bennett, supra n 40 at 29–30, 69. 134 Supra n 23 at para 75. 135 Ibid. at para 80. 136 Ibid. 137 Ibid. at para 81. 138 Omeredo, supra n 23 at para 5. 139 Ibid. 140 Fineman, supra n 6 at 13–14. 141 Ibid. at 14. 142 I am grateful to Karin Chinnian for this point. 143 Stringer, Knowing Victims: Feminism, Agency and Victim Politics in Neoliberal Times (2014) at 9. 144 Mullally, supra n 2 at 481. 145 For critiques of stereotypical understandings of vulnerability, see Fineman, supra n 6 at 8; and Peroni and Timmer, ‘Vulnerable Groups: The Promise of An Emerging Concept in European Human Rights Convention Law’ (2013) 11 International Journal of Constitutional Law–I•CON 1056 at 1064. 146 Supra n 22. 147 Ibid. at para 68 (author’s translation). 148 Ibid. 149 See, for instance, Schneider, ‘Feminism and the False Dichotomy of Victimization and Agency’ (1993) 38 New York Law School Law Review 387. For a critique of this understanding of vulnerability in defining the archetypical refugee, see Spijkerboer, supra n 121 at 8. 150 On the negative implications of assumptions about ‘appropriate’ female victim behaviour in asylum cases, see Mullally, supra n 2 at 482; and Baillot, Cowan and Munro, ‘Crossing Borders, Inhabiting Spaces: The (In)Credibility of Sexual Violence in Asylum Appeals’ in FitzGerald (ed.), Regulating the International Movement of Women: From Protection to Control (2011) 111 at 119. 151 Murdocca, In Conversation with Sherene H. Razack, ‘Pursuing National Responsibility in a Post-9/11 World: Seeking Asylum in Canada from Gender Persecution’ in Hajdukowski-Ahmed, Khanlou and Moussa (eds), Not Born a Refugee Woman: Contesting Identities, Rethinking Practices (2008) 254 at 258. 152 Supra n 23. 153 Ibid. at paras 1–2. 154 Ibid. at para 4. 155 See, for instance, Otto and Kapur, supra n 8. 156 For a discussion of feminist critiques of these essentialist constructions, see, for instance, Mullally, Gender, Culture and Human Rights: Reclaiming Universalism (2006) at XXX–XXXI and 85. 157 Otto, supra n 8 at 320. 158 Ibid. 159 Otto, ‘International Human Rights Law: Towards Rethinking Sex/Gender Dualism’ in Davies and Munro (eds), The Ashgate Research Companion to Feminist Legal Theory (2013) 197 at 199. 160 In addition to these three cases, see W.H. v Sweden Application No 49341/10, Merits and Just Satisfaction, 27 March 2014, at paras 66–67. However, as noted in supra n 11, W.H. is not examined in this article because the Court’s Grand Chamber struck the application out of the list. 161 The argument was put on the Court’s table by the parties in A.A. and Others (2012), supra n 22 at paras 61 and 70; R.H., supra n 22 at paras 10, 44 and 49; and N., supra n 21 at paras 40 and 47. 162 A.A. and Others (2012), supra n 22. The Court noted (at para 83) that AA’s mother and sisters were also in Yemen but only secondarily, pointing out that they would be ‘adding to her social support network’. 163 Ibid. at paras 82–83. 164 Ibid. at para 83. See also Spijkerboer, supra n 121 at 4 (critiquing the Court’s assumption that the male relatives will be a source of protection). 165 A.A. and Others (2012), supra n 22 at para 90. 166 R.H., supra n 22. 167 Ibid. at para 44. 168 Ibid. at paras 10 and 44. 169 Ibid. at para 70. See discussion below at Section 5. 170 Ibid. 171 Ibid. at para 73. 172 Ibid. 173 Ibid. at para 33. The dissenting judges noted that the applicant’s male relatives may be equally, or more, hostile than society to her status: see Joint Dissenting Opinion of Judges Zupančič and De Gaetano. The applicant claimed (at para 44) that the main threat would come precisely from male relatives, namely her uncles. However, the Court had serious doubts (para 72) about the veracity of her allegations, as she had not convincingly explained why she presented such allegations to the migration authorities only after more than a year since filing her asylum application. 174 Fineman, supra n 6 at 15. 175 On this point, albeit in the Australian context, see Kneebone, ‘Women Within the Refugee Construct: “Exclusionary Inclusion” in Policy and Practice – The Australian Experience’ (2005) 17 International Journal of Refugee Law 7 at 32. For a critique in refugee law of decision-makers’ reliance on non-state actors’ capacity to protect, see Hathaway and Foster, supra n 64 at 289–92. For a critique of the Strasbourg Court’s use of ‘a very wide notion of agents of protection’, see Spijkerboer, supra n 121 at 4. 176 Salah Sheekh, supra n 38 at para 137 (emphasis added). 177 On this challenge, see Otto, supra n 8. 178 Kneebone, supra n 175 at 32. 179 Otto, ‘The Exile of Inclusion: Reflections on Gender Issues in International Law Over the Last Decade’ (2009) 10 Melbourne Journal of International Law 11. 180 See, for instance, Otto, supra n 8. 181 Edwards, supra n 29 at 33. 182 R.H., supra n 22. See discussion above at Section 4. 183 Supra n 19 at para 16. 184 Ibid. 185 Ibid. at para 12. She also claimed (ibid.) that she would be sentenced to death or to cruel punishment by the Iranian authorities for adultery. 186 Ibid. at para 16. 187 R.D., supra n 21 at paras 31–32. 188 Ibid. at para 40. 189 Ibid. at paras 19–20. 190 Ibid. at para 45. 191 Ibid. at para 40. Another reason (at para 43) was that the applicant’s family had the means to find her elsewhere in the country. 192 Supra n 21. 193 Ibid. at para 47. 194 Ibid. 195 Ibid. at para 55. 196 Ibid. at para 56. 197 Ibid. at para 57. 198 Ibid. 199 In order to counter stereotypes such as these, Razack proposes precisely to see the woman asylum seeker in her particular ‘historical, social, political and cultural context’: see Murdocca, supra n 151 at 259. 200 N., supra n 21 at para 60. 201 Ibid. 202 Supra n 22 at para 70. The Court does not limit itself to inquiring into the legislative progress in Somalia. It goes further and observes that, notwithstanding a sexual offences legislation, sexual violence against women remained widespread in the country. The Court also highlights one of the main causes behind women’s inability to secure state protection: impunity. As for the socio-cultural context, it does not miss the fact that Somali women were generally discriminated against: see ibid. 203 Ibid. 204 Ibid. at para 73. 205 See, for instance, Nedelsky, supra n 7. 206 See, for instance, Mullally, supra n 156 at 71–3. 207 Nedelsky supra n 7 at 19–22 and 30–31. See also ibid. at 39 (arguing for a relational conception of autonomy). 208 Ibid. 209 See Spijkerboer, supra n 121 at 8. 210 Murdocca and Razack, supra n 151 at 259. 211 For a critique of the reluctance in refugee law ‘to embrace women qua women as a social group’, see Foster, supra n 1 at 32. 212 Saadi v Italy Application No 37201/06, 28 February 2008 (Grand Chamber) at para 128 (‘[T]he Court’s examination of the existence of a real risk must necessarily be a rigorous one’). 213 Brems, supra n 126. 214 Heaven Crawley argues that women’s persecution ‘can only be fully grasped by understanding the gendered nature of rights available to women in specific geographical contexts’: see Crawley, ‘[En]gendering International Refugee Protection: Are We There Yet?’ in Burson and Cantor (eds), Human Rights and the Refugee Definition: Comparative Legal Practice and Theory (2016) 322 at 341. 215 Ibid. at 342–3. © The Author(s) . Published by Oxford University Press. All rights reserved. For permissions, please email: firstname.lastname@example.org This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
Human Rights Law Review – Oxford University Press
Published: May 17, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera