‘Hearing the Right Gaps’

UK Asylum Decision-making: Procedures and Challenges

While the EU Qualification Directive substantiates certain subsidiary forms of international protection, particularly under Articles 3 and 8 of the European Convention on Human Rights, securing full status as a ‘refugee’ continues to require an asylum applicant to establish to a ‘reasonable degree of likelihood’ that he is a person who ‘owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country’ (Article 1A(2) 1951 Geneva Convention on the Status of Refugees, as modified by its 1967 Protocol). To be successful, therefore, an applicant’s claim must be deemed by decision-makers both to be credible and to correlate sufficiently to the definitional criteria for refugeehood (Home Office, 2010b). In the UK, both of these elements will be evaluated primarily on the basis of the accounts provided by claimants during two interviews, which are conducted by UKBA personnel at different stages in the application process. The first of these – the ‘screening interview’ – occurs at an early stage, and largely involves the communication of basic information regarding the applicant’s mode of travel and arrival into the UK, country of origin and the alleged nature of her persecution. At this screening stage, a decision will often also be taken as to the applicant’s suitability for detained fast track procedures (for analysis of women’s experiences of these procedures, see Human Rights Watch, 2010). Whilst non-detained applicants can expect to undergo a second ‘substantive’ interview within one month of lodging their claim, those who are detained will face an accelerated process within which they will be interviewed by the UKBA again in a matter of days. Either way, the purpose of this second interview is to obtain further information regarding the nature of the applicant’s claim, on the basis of which, and in conjunction with any evidence that may be available via country of origin experts or medical reports about the applicant, the UKBA Case Owner will make an asylum decision.

In 2010, 75% of applicants in the UK were refused either asylum or any form of subsidiary protection at this initial UKBA decision stage, a percentage that is roughly in line with statistics for previous years (Home Office, 2010a). Most unsuccessful applicants will go on to lodge an appeal against this refusal to the independent Immigration and Asylum Chamber First Tier Tribunal (hereafter ‘the tribunal’), which will lead to an in-person hearing presided over by an Immigration Judge. ¹ Many applicants will be assisted at this tribunal hearing by a legal representative, barrister or immigration adviser. The availability of such assistance has, however, become increasingly difficult in recent years as a result of legal aid restrictions and the associated application of a stringent merit test by solicitors, as well as the recent collapse of the UK’s two leading not-for-profit immigration advice firms (Refugee and Migrant Justice and the Immigration Advisory Service). ² If required by the applicant, language support will also be provided at the tribunal by an official interpreter. In this formally adversarial forum, the UKBA will usually – although, again, as a result of increasing budget and operational constraints, not always – be represented, either by a Presenting Officer or the Case Owner who made the initial refusal decision. Although applicants may be examined and cross-examined at the tribunal, the appeal hearing itself proceeds relatively quickly (typically being completed within a couple of hours), relying heavily on prior submission of paper documents, including personal statements from the appellant, the UKBA interview transcripts and UKBA Reasons for Refusal Letter, as well as any additional expert/medical reports that may have been commissioned by the UKBA or legal representative, where applicable.

In 2010, 27% of such appeals against refusal of leave to remain heard at the tribunal were allowed (Home Office, 2010a). There are a number of factors that can effect a change in outcome for an appellant, including the emergence of new evidence not available at the time of the initial decision or a change in the levels of risk and security associated with a country of origin. At the same time, however, previous research (Ceneda and Palmer, 2006; Thomas, 2011) has raised concerns that some aspects of the asylum process, such as the short timescales for UKBA decision-making and the tightly structured question/answer techniques used in substantive interviews, may reduce the prospects for quality decision-making at first instance. The suggestion that this may have a disproportionately negative impact on women’s claims is supported, moreover, by recent research which found that, in contrast to the average overturn rate of 27% outlined above, some 42% of initial refusals in women’s cases were overturned at the appeal tribunal (Asylum Aid, 2011). Several commentators have gone further, alleging that there exists, particularly within the UKBA, an institutional ‘culture of disbelief’ or ‘culture of denial’, which promotes a sceptical appraisal of applications from the outset (Souter, 2011). While we believe that simplistic assertions of the existence of a ‘culture of disbelief’ risk trivialising the myriad structural, cultural, political, economic, emotional and bureaucratic factors that can interact in framing the processes and outcomes of asylum decision-making, as will be discussed below, our findings do lend some support to the above-mentioned specific concerns about institutional structures and practices around timescales, evidence-gathering and opportunities for disclosure, at least in relation to claims lodged by female asylum-seekers that involve an allegation of rape.

On Methods, Data and Limitations

The fieldwork for the present study was conducted from August 2009 to December 2010. To assist in providing background context and in formulating questions and topics for closer examination, a small number of UKBA substantive interviews were observed by each member of the research team and a small-scale sampling exercise was undertaken across three regions of initial UKBA decisions in cases involving female asylum-seekers. For the main part of the study, the researchers conducted a series of 104 semi-structured interviews with a range of professional actors in the UK asylum process, specifically, immigration judges (n = 20), UKBA personnel (n = 24), legal representatives (n = 25), non-governmental organisation (NGO) personnel (n = 21) and interpreters with experience in translating at UKBA interviews or tribunal hearings (n = 14). These interviews provide a key source of data for the present study, and were combined with a series of 48 observations of appeal tribunal hearings, the vast majority of which involved a female appellant who had disclosed an experience of sexual violence. Of these cases, 31 were identified and referred to us by legal representatives, voluntary sector organisations or the Tribunal Service, and were observed with the consent of the appellant and her legal representative. Of these 31 cases, the team successfully secured access to surrounding case files (which included witness statements, interview records, UKBA refusal letters, etc.) in 12 cases. The remaining 17 hearings observed in the study were identified as a result of random sampling, and since many of these involved unrepresented appellants, some of whom (n = 10) had their cases considered within the detained fast track system, this also enabled the research team to explore a range of issues associated with divergence in the levels of legal and voluntary sector support provided, as well as in the opportunities for securing expert reports and for presenting a full and thorough narrative account.

Although this data provides rich and important insights, there are, of course, a number of limitations that must be borne in mind. For one thing, although this should not be read as implying a minimisation of the experiences and impact of sexual violence on male asylum applicants, the primary focus of the present study was quite deliberately upon female claimants. There were a number of reasons for this, amongst them the fact of the reportedly higher incidence of rape amongst this group, ³ the ability to extrapolate from a stronger body of pre-existing research in other contexts (e.g. criminal justice) in relation to the experiences of female complainants, and the need for the distinctive gender issues at stake in male and female victimisation to receive their own careful attention. In addition, it should be noted that, while it is important to understand more about women’s direct experiences of the asylum process, this research does not include interviews with female asylum applicants themselves. Again, this was the result of a deliberate decision on the part of the research team. Given the already arduous nature of re-telling narratives of abuse, sexual violence and torture, we took the view that it would be possible to obtain information as to the procedures for facilitating and handling such disclosures (which were, or were not, utilised by professional actors in the asylum system) without speaking directly to the women involved. However, these women’s voices are not entirely absent from the research – they are present (to some extent at least) in the context of the observed asylum hearings and surrounding personal statements within case files, and are represented (albeit in a mediated fashion) through the accounts provided by legal representatives, NGOs and others within our interviews. At the same time, we realise that, to the extent that the direct inclusion of these women’s voices is limited, a potential criticism of this study is that it – somewhat ironically – repeats the problematic cycle of silencing that we seek to identify and challenge. Giving voice to women asylum claimants is doubtless a legitimate focus for future research and reflection (see, for example, the Testimony Project, at http://www.testimonyproject.org/). Equally, however, we remain confident that, in terms of the specific focus and aims of the current project, our data provide significant and crucial insights into the views and techniques of the professionals who, whether they recognise it or not, through their conduct profoundly influence the production of the narratives upon which women’s claims for international protection are ineluctably based.

Disclosures of Rape in the Context of UK Asylum Decision-making

Although it is a subject that has ignited considerable critical concern (Black Women’s Action Project and Women Against Rape, 2006; Bögner et al., 2007), no official statistics exist to determine the proportion of asylum claims lodged in the UK by female applicants that involve an allegation of rape. ⁴ Equally, remarkably little is known regarding the role and significance that any such disclosure might assume in terms of the concrete determination of refugee status. In this section, we briefly outline some of the responses provided by study participants, which both illustrate the considerable divergence of views as to the scale of the problem and indicate the multiple, and often intersecting, ways in which disclosure may be relevant.

When asked to estimate the scale of disclosures of rape in relation to their own caseloads, participants in the present study provided a very wide variety of responses – ranging from one judge who estimated the incidence at 5% or less to one legal representative who told us that ‘nearly all’ of his cases involved such disclosures. This variation did not appear to be related – as one might have expected – to the gender of the respondent, their locality or their job description. Indeed, within the same region, one male UKBA respondent put the incidence of rape at 60–70%, while a female colleague estimated it at no more than 20%. Equally, amongst respondents within the voluntary sector, estimates ranged from ‘four clients in two years’ and ‘1% or lower’ to ‘almost all women’s cases’. In broad terms, though, and notwithstanding such variations, it was clear that the majority of respondents considered rape to be a common feature within women’s asylum applications, placing the incidence at above 50%. Many respondents emphasised, moreover, that prevalence would vary significantly depending upon country of origin, and the geographical generalisation was regularly made that cases involving applicants from ‘African countries’ (in particular, the Democratic Republic of Congo, Somalia and Zimbabwe) had extremely high incidences of sexual violence.

Despite this potentially high incidence, however, the relevance of the rape claim itself to the asylum determination process was often perceived by participants to be limited. Refugee jurisprudence, including the EU Qualification Directive, has confirmed that rape is a serious form of abuse, which – if meeting the other criteria associated with refugeehood, such as a relevant Convention reason and a lack of prospects for internal relocation or protection against recurrence if returned ‘home’ – would constitute persecution and suffice for the granting of asylum. Despite this, however, respondents in the present study – including judges – typically continued to depict rape as likely to simply be ‘part and parcel’ of a package of abuses that would have been suffered by an applicant, and unlikely in itself to be the central or sole basis for a successful claim.

The extent to which this approach reflects a gendered interpretation of nominally neutral standards for refugee protection has not gone unnoticed, and has itself been the subject of extensive criticism (Anker, 2002; Crawley, 2001; Kneebone, 2005). This body of work has highlighted the perceived inability of refugee law to encompass the often private harms suffered by women, such as rape, domestic abuse, forced marriage and honour crimes, owing to their invisibility from the public (male) realm of ‘real’ or ‘political’ persecution. When combined with the failure of Country of Origin Information to provide dedicated data on the re-settlement prospects of women returning to regions of their country far from family and support networks, this trivialisation of gendered forms of persecution has created in the doctrine of internal relocation an often insurmountable barrier to the awarding of international protection (Bennett, 2008). Moreover, the very reliance by even the best intentioned of refugee advocates on an image of refugee women as passive victims of private, apolitical harms can perpetuate the patriarchal violence women have often sought to flee (Kneebone, 2005).

At the same time, however, it is important to note that the fact that respondents often appeared to downplay the relevance of a rape allegation as a central plank of the doctrinal evaluation of an asylum claim does not suggest that respondents were oblivious to the myriad other important functions that such an allegation might play in terms of bolstering a positive decision. It was acknowledged, for example, that, as a component of the broader ‘factual narrative’ of the case, a disclosure of sexual violence may help to establish that an applicant has undergone ‘serious harm’ in the past, which in turn will be taken as an important indicator of future risk. More generally, some respondents noted that a disclosure of rape could assist decision-makers to, in the words of one judge, understand an applicant’s ‘psychological ability to formulate her claim’, thus acting to explain or justify an otherwise disjointed or incoherent narrative. Similarly, it was noted that incidences of rape – where they are judged to be credible – may contribute positively to assessments of the applicant’s personal or ‘overall’ credibility in relation to other, potentially more pivotal, aspects of her claim (for critique of such ‘broad’ assessments, see Kagan, 2003; Sweeney, 2009).

In summary, then, although their exact scale remains uncertain, there is good reason to believe that disclosures of sexual assault are not uncommon within women’s claims for asylum in the UK. Moreover, it is clear that these disclosures, appropriately of relevance within the doctrine of the Refugee Convention itself, can perform a variety of other functions in terms of framing the response that will be adopted by asylum decision-makers. In the following sections, we explore the processes by which such disclosures are facilitated, the expectations that influence the ways in which such disclosures are heard and evaluated, and the factors that operate to impinge upon the prospects for both initial disclosure and fuller narration throughout the asylum application process.

Disclosure at the Screening Interview: More Hope than Expectation?

While the UKBA’s guidance acknowledges that the screening interview is designed to elicit only basic information about an applicant’s substantive reasons for claiming asylum, in several of the cases that we observed in our study, the fact of non-disclosure at this stage was nonetheless used to undermine an asylum claimant’s testimony regarding rape. In one case, for example, a UKBA Presenting Officer asserted in his submissions to the tribunal judge that the UKBA did not ‘find it credible that neither the appellant nor his wife would have stated at their screening interview that his wife was raped twice at gunpoint’. He maintained this position, moreover, despite the appellant’s explanation that he had been expressly told to only give the basic details of his case at screening and so, to comply with this instruction, he had not mentioned the incident at this stage.

This insistence on the feasibility of early disclosure at the screening interview was belied, however, by the responses given by many of our interview respondents when asked where and when women were likely to first disclose. Here, many respondents, including a significant proportion of UKBA personnel, recognised that an experience of rape was in fact unlikely to be disclosed at the screening interview stage. The reasons given for this echoed many of the findings made by the Chief Inspector of Immigration during an unannounced visit to the UKBA’s Asylum Screening Unit in Liverpool in 2009. The Inspector’s report highlighted obstacles ranging from a ‘noisy and chaotic environment’ to a lack of privacy and childcare facilities, and gave specific examples of poor staff practice, including one case in which a UKBA employee sent a text message whilst conducting the screening interview and another in which part of a screening interview was inadvertently broadcast over the office’s tannoy system (Vine, 2009). Paralleling these concerns over privacy, confidentiality and professionalism, an appellant whose case was observed during the present study explained to the tribunal that, having claimed asylum at the port, she felt unable to give full details of her reasons for seeking protection, which included a substantial history of sexual abuse, at the screening interview because there was a queue of people from her own country waiting to speak with UKBA officials, who she feared would have been able to hear every word she said.

The particular Screening Unit that faced such staunch criticism from the Chief Inspector has now closed, and an arguably long-overdue programme of review and improvement of screening processes is to be undertaken by the UKBA in 2011–2012. ⁵ While this will hopefully generate significant improvements, there remains an extent to which the insistence by some of our participants that sexual violence ought to be disclosed at the screening interview may reflect more hope on the part of the UKBA than realistic expectation. Several respondents identified structural issues around the questioning and purpose of screening that could impede disclosure of sensitive information, even in a more appropriate physical environment. As one UKBA Case Owner put it when describing the interview, ‘it is quite harsh and it’s just a matter of just completing forms and getting the information down; and they are asked questions “well why did you leave?” and I think if it seems quite harsh then they’re not going to give a lot information at that point.’

Disclosure at the Substantive Interview: The Moment of Truth

Although, as noted above, some interviewees and professional actors that we observed in the present study continued to insist that disclosures of rape ought to be made at the screening interview, there were many others who were more open to acknowledging the constraints – in terms of both physical environment and timing – of this forum. For a large number of these respondents, however, this recognition entailed that the subsequent substantive UKBA interview emerged as a fundamentally suitable, not to mention entirely expected, site for disclosure. Almost every respondent in the present study concurred that failure to disclose any relevant aspect of the claim at this substantive interview would raise significant credibility issues for UKBA decision-makers. The expectation of full disclosure at this stage was particularly strong, moreover, in situations where the applicant had been in receipt of voluntary sector or legal support in advance of her asylum interview. Thus, as one UKBA Case Owner commented, ‘especially if someone has had voluntary support as well, I think people would question why they haven’t disclosed it … Even though I think it’s widely known there are issues’.

Interviewees repeatedly emphasised that it would have been explained to the applicant that she must disclose everything during her UKBA interview and so considered that, as one Case Owner put it, ‘the overwhelming need would be to remain in the UK and that would sort of trump the need to withhold sensitive information’. In a context in which it is widely acknowledged that the quality of advice and support provided to asylum seekers is variable, and in which restrictions on legal aid funding often make it difficult for legal representatives to devote the time required to develop effective relationships of communication and trust with applicants, this presumption that asylum claimants fully appreciate the purpose of the substantive interview, let alone the importance of full disclosure therein, is questionable. Moreover, such perspectives rely on an arguably problematic, not to mention often somewhat unrealistic, logic that automatically positions the potentially traumatised asylum seeker as an empowered and rational actor.

At the same time, however, it is important to point out that, while all participants agreed that the substantive interview was vitally important, the perception that it constituted an ideal environment in which to provide sensitive information was not universally supported. One legal representative observed that it is not ‘really a situation that’s conducive to finding out exactly what happened to someone’. Meanwhile, UKBA personnel in one region described the rooms in which these interviews are conducted as often inappropriate, since ‘you can hear other people, what other people are saying in the rooms next to you’. Further concerns were expressed, moreover, regarding the interview structure and questioning methods typically adopted, whereby UKBA personnel rely on a direct question and answer format, transcribing claimants’ responses verbatim, contemporaneously and by hand. ⁶ In line with previous research regarding the impact of such techniques (for example, Lamb et al., 2000), our observation of a small sample of substantive interviews in the present study suggests that their deployment can stifle free-flowing narration by applicants. This is problematic given that free narrative has been found, in a forensic context, to be a particularly effective means of obtaining more accurate information from vulnerable interviewees (Bull, 2010; Powell and Bartholomew, 2003). Not only are periods of delay and disruption built into the structure of the interview, which can lead to applicants losing the momentum in providing accounts of traumatic experiences, but this format can also diminish applicants’ perceptions of being fully listened to. Moreover, although applicants are officially advised that they can, and should, give full and detailed answers, there is a risk that they will instead truncate their responses in order to avoid over-burdening the transcribing interviewer, generating omissions that can later emerge as problematic in terms of establishing their credibility.

In regard to the specific issue of rape allegations, further potential obstacles to disclosure at the substantive interview were identified. Indeed, while it was widely accepted that failure to ask the ‘right’ questions could significantly impede or even prevent disclosure, several respondents reported that UKBA personnel would often deliberately avoid asking questions about a possible experience of rape – even in cases where, within the account provided by the applicant, there were indications suggesting that it may have occurred. As one UKBA Case Owner put it ‘some interviewing officers don’t like to ask specific questions – how were you raped, how many men, what exactly did they do’. Such apparent reluctance to probe was certainly exemplified in one appeal hearing that we observed during the present study in which the claimant recounted having been held in detention in the Democratic Republic of Congo on three occasions and having been raped during the third period. It was apparent from the written record of the applicant’s substantive interview that the male UKBA officer (who, having forgotten his spectacles, wore sunglasses throughout) did not ask for any further detail regarding the alleged rape, and nor did he ask any questions about the treatment the applicant had received during her previous detentions, despite country evidence indicating that rapes in such conditions were common. ⁷ The resulting lack of detail within the record of this claimant’s substantive interview was later relied upon to reject her claim and to support the UKBA’s assertion that she had invented her claims of detention and abuse. ⁸

Clearly not all UKBA Case Owners will be so reluctant to probe for detail, and many indicated that they will often revert from a closed to more open questioning style at the end of the substantive interview in order to afford applicants an opportunity for additional disclosures. Yet this in itself may come too late and be of limited assistance. As one Immigration Judge observed, ‘especially with the Home Office introduction saying only answer the questions, they [asylum applicants] may really feel inhibited about bringing up something that they wish to add, which they haven’t said before. And at the end of the interview, they say “are these all the reasons you want to adduce for your claim?”, and by then, they’re sort of exhausted or they, perhaps, form an impression of the interviewer and they don’t want to make any further claim’. This concern was again re-affirmed by an appellant in one of the cases we observed who, when advising the tribunal that she had had difficulty understanding the questions posed by her UKBA interviewer, confirmed that she had not reported this at the close of the interview when asked if she had understood everything because by then she was ‘fed up, just wanted to get out of the interview at all costs, because I did not know what was going on around me’.

For some participants, one factor that could militate against this general expectation to disclose at the substantive interview arose where there had been a failure on the part of the UKBA to provide a woman claimant with a female interpreter and/or Case Owner. Many decision-makers accepted that an absence of such ‘gender-matching’ could reasonably and legitimately explain a woman’s unwillingness to speak of sexual violence. Whilst widely accepted as ‘good practice’ in regard to the eliciting of reluctant disclosures within the criminal justice context, as has also been shown to be the case in that arena, the faith thereby placed in the mere fact of gender-matching may be unfounded. This approach repositions the ‘emotional labour’ of dealing with rape disclosures as suitable only for women (Martin, 2005: pp. 167, 190), regardless of their individual skills, beliefs or personal characteristics. Such stereotypes of women as necessarily more receptive and sympathetic listeners to narratives of rape have, however, been challenged by social science research, which has indicated that women may be more sceptical appraisers or harsher critics of women’s contributory responsibility (see, for example, Anderson et al., 2001; Cowan 2000; Ellison and Munro, 2009b). What is more, this tendency to see an all-female environment as a panacea risks doing further damage to the credibility of women who fail to disclose where there are no men present. As one UKBA Case Owner explained, ‘in the interview, it comes up; if it doesn’t then you would obviously want to know why they didn’t tell you, especially if you are a female interviewing officer’. Problematically, then, such an approach ignores the myriad other factors, aside from gender-matching, that might impact upon a women’s capacity or desire to disclose.

Reliance on the mere fact of the presence of a female interviewing officer as a mechanism by which to ensure rape disclosure may also function to inappropriately absolve UKBA interviewers of any responsibility for facilitating narration in other ways. Many of the more experienced legal representatives interviewed in this research recounted strategies that they use to enable disclosure, including asking women to write down in their own language things that they find difficult to verbalise, and responding to unspoken cues or being attentive to opaque references to, for example, ‘suffering from back pain’ or ‘falling unconscious’. By contrast, the majority of UKBA respondents in the present study expressed confidence not only that ‘genuine’ experiences of rape would be disclosed by the time of, or during, the substantive interview, but that these would be recounted to them in a clear and direct way, without the need for probing. When asked whether they had encountered cases where they suspected that a rape may have occurred but it was not disclosed, the majority of legal representatives answered in the affirmative whilst most of the UKBA personnel did not. Meanwhile, although several judges accepted that such situations do arise, many emphasised the limits of their role – as one put it, ‘you suspect that it’s there, but we aren’t to go digging around’; and another noted the UKBA ‘didn’t go down that route and it’s not our position to … (J)ust as you wouldn’t ambush an appellant, you don’t ambush the Home Office. It’s not our role’.

Some UKBA personnel did exhibit signs of reflecting more consciously than others on strategies that they might adopt to create an atmosphere that would be more conducive to sensitive disclosures, for example, by offering breaks, tissues or water as required when the applicant appeared distressed. However, it was evident that, in many cases, such allowances could remain inadequate, particularly given the power asymmetries inherent in the situation and the potentially limited opportunities for narration afforded by the overall environment and questioning structure. Alcoff and Gray (1993) have argued that the very set-up of the interview strongly echoes the classic ‘confessional’, with an expert ‘confessor’ – here the UKBA official – eliciting disclosure in order to then judge its veracity and validity according to criteria external to and uncontrolled by the applicant. Basic tactics designed to create ‘comfort’ in such a context are, therefore, unlikely to ever be fully effective. At the same time, however, the minimalist approach often taken to this task in the case of UKBA personnel indicates that the prognosis is even less promising – in contrast to legal representatives and NGOs who typically recounted the need for multiple interviews and appointments before trust and ‘comfort’ could be established, several of the UKBA personnel interviewed in this research suggested that ‘two to three hours’ of time with an applicant was ample for the purposes of building up a relationship of trust that would ensure that disclosure was possible.

Disclosure at the Appeal Tribunal – A Third Opportunity or an Opportunity Missed?

While, for a variety of structural and inter-personal reasons, the UKBA screening and substantive interviews may thus fail to create an environment that is conducive to rape disclosure, it is apparent that they continue to attract considerable levels of confidence from decision-makers, many of whom insist that applicants who postpone disclosure until the appeal stage will be likely to encounter substantial levels of disbelief. The impact of this was expressed particularly vividly in the present study by one legal representative, who maintained that ‘I always say to my clients if you don’t mention it at your asylum interview, then you can never mention it; and I don’t want to know about it because anything added later is embellishment – always. I have never seen a determination come back that didn’t go down that route if something new came up at the hearing’.

Importantly, however, and contrary to this legal representative’s assertion, such a staunch view was not in fact shared, or at least not acknowledged, by many of the immigration judges that we interviewed. Indeed, most stated that they had little time for the sort of blanket arguments they attributed as commonly put forward by the UKBA, whereby the mere fact that a disclosure of rape was made late, particularly where it was postponed until the tribunal stage, was asserted to significantly undermine its credibility. Several judges even displayed some sympathy with the predicament of women who, out of a sense of desperation, decided at the ‘11th hour’ that they had no choice but to disclose an experience of rape that they had previously sought to avoid mentioning, denied or opaquely alluded to only in passing. Equally, however, there were indications in the comments made by such judges that where disclosure did occur at the tribunal, then the manner of the appellant’s disclosure could assume a disproportionate importance.

Of the judges who recounted having dealt with initial disclosures of rape within their courtrooms, three specifically commented on the demeanour of the appellant, indicating that a ‘dramatic breakdown’ or a ‘flooding out of material’ would be viewed as more credible than a calmly laid out account. It has been well established in previous research that reliance on demeanour as a marker of (in)credibility is problematic, particularly in relation to accounts of traumatic experiences such as rape and in contexts that require cross-cultural verbal and non-verbal communication (Good, 2007; Kälin, 1986). It also goes against guidance issued to asylum decision-makers, which insists that they ‘should not be influenced by subjective factors, for example, if the applicant appears nervous or fearful … or entirely calm and rational’ (Home Office 2010b: p. 15). Nonetheless, it was apparent in this study that demeanour did influence the approach taken by several judges to assessing disclosures made at the tribunal. As one put it, ‘she smiled very sweetly and said “and I was raped” … clearly in the way she said it, it was clearly a lie’.

Even where judges resisted this temptation to base credibility judgments on the demeanour that accompanied a disclosure made in their courtrooms, moreover, it was apparent that their suspension of disbelief as to the delay exhibited by the appellant in revealing her experience of rape was highly conditional upon her being able to explicitly articulate ‘good reasons’ for not having disclosed during the UKBA substantive interview. To this extent, then, while judges often expressed disapproval of UKBA arguments that automatically treated tribunal-stage disclosures of rape with scepticism, that position remained highly contingent on the appellant being able to convince judges that there were appropriate and compelling reasons that could justify the delay. In effect, this served, therefore, to re-position the substantive UKBA interview, notwithstanding its shortcomings, as the most important and most credible site for rape disclosure. Perhaps more problematically, moreover, the ways in which such ‘good reasons’ were constructed by the judges in the present study raised a number of concerns and indicated a potential to operate in ways that might obscure rather than explain disclosures of sexual violence.

Culture and Shame

‘Culture’ and specifically the shame felt by women from ‘different cultures’ regarding sex and sexual violence were widely cited by respondents to explain, and in some cases justify, late or partial disclosures (see also Einhorn, 2009). NGO personnel frequently reported that, in contexts of domestic abuse, women from cultures where marital rape was not recognised would be less likely to label their experience in this way, or to consider what had been done to them as abusive, such that a direct disclosure may never be forthcoming. Even where rape was perpetrated outside the domestic context, several respondents cited instances where women feared that their husbands would abandon, divorce or harm them if they were to discuss having been raped with immigration officials. Such women, it has been noted, may fear even wider repercussions: ‘It is almost as if, when a woman is raped, a bomb is planted but has not gone off. When the woman talks about it, the bomb explodes and destroys relationships, and also destroys the fabric of the community of which she is a part’ (Tankink, 2006: p. 8). Reflecting a commendable effort to mitigate these effects, several judges in this study recounted cases where, having been alerted to these difficulties by legal representatives, they ensured that family members were not present in court during a wife’s testimony.

Even if instrumental to justifying late disclosures of sexual violence, depictions of the dual forces of shame and culture frequently operated in the present study not to elucidate the circumstances of individual women but to produce stereotyped assumptions about ‘other’ nations. ‘Culture’ was often ascribed as belonging only to the applicant, with few respondents recognising that they too had a culture that might influence both a woman’s ability to disclose and their ability to ‘hear’ the narrative. This cultural ‘tunnel vision’ interacted with a sense that the shame felt by these ‘other’ women was somehow bound to be greater than that which would be experienced by UK-raised victims, particularly on account of the former’s less liberated ‘lifestyles’ and social structures. As one judge put it, ‘the shame that attaches to rape is clear and obvious and I think it’s stronger in … the countries and cultures that have less sophisticated or enlightened attitudes to rape and sexual violence, then the level of shame is higher’. Given continuing concerns in the UK about low conviction rates for rape, not to mention ongoing evidence of the multiple detrimental effects that sexual victimisation can generate for its western victims, this assumption is not only misplaced but based on inaccurate perceptions of prevailing attitudes to sexual violence ‘at home’.

Of course, in a context in which decision-makers are faced with applicants from a huge range of nationalities and ethnicities, reliance on less than complete perceptions of cultural mores is, to some extent, inevitable. Using culture to explain and distance oneself from accounts of sexual abuse may also act as a ‘coping’ mechanism for actors who are continually confronted with traumatic narratives (Baillot H, Cowan S and Munro V, ‘Second hand emotion: exploring the contagion of trauma and distress in the asylum law context’, work in progress). For women applicants, however, these overly general approaches to culture can be disadvantageous. For one thing, they may privilege only those women whose accounts do not deviate from understood patterns of oppression, which themselves are often constructed by observers in stereotypical and monolithic ways (Baillot et al., 2011). They thus risk depicting experiences that do not conform to the norms of submissive women ‘entangled in backward practices’ as ‘culturally unbelievable’ (Gedalof, 2007: p. 90), and perpetuate images of women as ‘abject others’ who must produce ‘performances of abject status’ to have a chance of obtaining international protection (Johnson, 2011: p. 62). Certainly, many of the UKBA personnel observed in the present study relied on evidence that women applicants had managed to remain in the UK for long periods, despite their lack of secure immigration status (often through working illegally or relying on friends), as a basis upon which to argue that they were ‘resourceful’ enough to be returned ‘home’ without risk and thus were not in need of international protection. Equally, there were indications that inflexible or ingrained perceptions of what constitutes a specific culture could diminish professionals’ awareness of the reality and impact of rape on women’s lives, with the result that such disclosures risked being dismissed as too commonplace to be important. As one UKBA Case Owner put it, ‘in that culture [Somalia] it happens so often that now it is just like talking about what happened on any particular day’.

Trauma and Stress

NGO and legal representative respondents repeatedly cited trauma as a major factor that could delay or prevent disclosure. As one NGO worker put it, ‘there could be a whole range of reasons why delay is, there is a delay in reporting. And just good old basic shock, the women could still be in shock. If women have got Post Traumatic Stress, you know, they’re still trying to process what has happened’. These same respondents described cases where the level of trauma was such that women found themselves unable to disclose even after having received full support and encouragement, or they felt able to disclose only in a specifically tailored clinical setting. Concerns regarding the impact of trauma on recall and memory were also widely expressed by respondents, and the difficulties posed in this regard were evidenced throughout the tribunal observations.

In a large number of the appeals we observed, appellants made reference, usually during questioning by UKBA personnel, to the fact that they had had difficulty remembering events, or had not been ‘in a good state of mind’ during the substantive interview, such that they had muddled dates, times or sequences of events. Despite clinical evidence corroborating the potential for such difficulties to arise in the wake of traumatic events, and particularly when asked to recall them in a stressful environment (Bögner et al., 2007, 2010; Cohen, 2001; Herlihy et al., 2002), it was clear in the present study that UKBA representatives afforded such explanations limited credence, continuing to pursue questioning around such alleged inconsistencies in account in order to frame them not as evidence of trauma, but as evidence of evasion or fabrication. This was confirmed by many of the UKBA decision-makers that we interviewed, who though keen to acknowledge that an experience of sexual violence may have been ‘traumatic’, nonetheless did not permit this recognition to challenge their basic view that such trauma would not ultimately prevent full and accurate disclosure. This view was exemplified by one UKBA Case Owner who insisted that, ‘other than the general traumatic experience you have in disclosing something horrific that’s happened to them. We try as much as we can to appease that … Generally, you know, you tackle it with a bit of tact and sensibility … people seem to open up as long as we kind of don’t go straight in there … generally people do open up and start telling what’s happened to them’.

While, on the one hand, such an approach can be seen to trivialise the significance of trauma and stress, on the other hand, there was a concurrent tendency whereby such accounts – in order to be afforded any greater level of significance – had to be medicalised. It was often assumed that, if an applicant had indeed been raped, then she would ‘need’ to seek therapeutic support in order to recover from this experience, and a failure to do so in itself was seen by some as legitimately generating suspicion. As one UKBA Case Owner put it, for example, ‘it [rape] is psychologically massively damaging and you would expect to see some psychiatric reports, some counselling going on there. If there are no psychiatric reports, no counselling whatsoever … then I would argue that it is very unlikely that this person has undergone such a treatment as that’. Such heavy reliance on medical evidence is problematic not least because it instantiates a presumed reaction to sexual violence that may not be universally shared and applies a heavily westernised ‘therapeutic’ lens to its resolution. This can effectively oblige women to conform to expected help-seeking behaviours, even where this may not be their wish. Thus, some of our legal representative respondents described situations in which they had ‘persuaded someone to have treatment’, not out of a concern for that applicant’s health but primarily, or exclusively, in order to obtain the sort of supporting evidence for her claims of sexual violence that decision-makers expected. At a practical level, moreover, there are further difficulties with this kind of approach, since several of our respondents acknowledged that restrictions on legal aid and the lack of time in which to gather evidence, particularly for unrepresented applicants or those within the detained fast track system, can make it virtually impossible to provide this type of corroboration.

This sense that women might be pushed into forms of treatment that they themselves have not chosen compounds concerns regarding a second strand of trauma-related discussion in the present study, which relates to the potential for re-traumatisation throughout the asylum application process. Alcoff and Gray (1993: p. 269) have maintained that ‘the survivor who reports sexual violence may feel empowered politically, but that does not generally outweigh the pain and humiliation of disclosure and its recollection of the frightening and agonizing assault and abuse’. Similar concerns were also evidenced in the context of our data. Several respondents described the process of applying for asylum and/or appealing against an initial refusal of leave to remain as ‘really frightening’, ‘very daunting’, ‘extremely stressful’ and ‘utterly horrifying’ for the women involved. Meanwhile, appellants themselves supported this assessment through their words and conduct – in one case we observed, for example, an appellant explained in her witness statement that ‘having to talk in detail about these matters [past experience of corrective rape] in this asylum claim has been very difficult for me mentally and it has further compounded the difficulties I was already having’. In other hearings, mirroring the distress that previous research has shown can often be associated with complainants giving testimony on rape in the criminal courtroom (see, for example, Lees, 1996), we observed appellants who had to be assisted out of the tribunal after breaking down in tears, or who spent most of their hearing with their head in their hands. Legal representatives and NGO workers in particular were adamant that disclosing to UKBA officers was extremely emotionally difficult, and that appealing a refusal at the tribunal could be equally distressing, often provoking a manifest ‘deterioration in the [applicant’s] emotional state’. One legal representative went so far as to suggest that ‘much of the system is really in itself quite damaging … people can be traumatised by the system’.

Respondents emphasised that sources of possible re-victimisation lay not only in the risk of being disbelieved by others, but in the conditions of the applicant’s treatment throughout the entire process. Practice guidance is in place, of course, to ensure the proper treatment of vulnerable applicants (Blake and Arfon-Jones, 2010). This is certainly followed in many cases, and there were instances of good practice recounted in our research. At the same time, however, our study lends some support to claims from previous research (Ceneda and Palmer, 2006), which has suggested that best practice is not uniformly adopted. Indeed, in one case that we observed, a highly vulnerable appellant had been detained in secure facilities, with hard-fought-for clinical evidence eventually being raised at the tribunal that indicated that, not only should she not have been detained given her psychological condition, but that her time in detention had in itself worsened her prognosis for recovery. Likewise, in another case, a young woman, who had allegedly experienced an extensive history of trafficking and sexual abuse, and whose social worker had testified to the appeal tribunal that she was the most vulnerable young person that he had worked with in his 10-year career, was nonetheless subjected to rigorous cross-examination by a UKBA Presenting Officer regarding the specific details of her abuse, captivity and health problems. Such questioning was never interrupted by the judge, notwithstanding the appellant’s visible distress, crying and hyper-ventilating.

Despite these concerns, there were a number of respondents, typically UKBA personnel and judges, who concluded that ‘as far as the process goes, I don’t think it’s that bad’ and noted that applicants on the whole ‘stand up to it pretty well and they can cope’. Such assertions typically presumed that the applicant would have had the benefit of a legal representative or other support worker who would provide guidance and assistance to vulnerable claimants. Not only is this presumption not always borne out in practice, but it was apparent in the present study that it was often the legal representatives themselves who spoke most candidly about ‘dragging’ or ‘teasing’ out a narrative from a reluctant and vulnerable claimant. As one representative admitted, ‘I’m probably worse than the Home Office, I probably push them far harder than they do … in fact I know I do. I’m quite horrible to them, but for their own good’. To this extent, legal representatives themselves perpetuate an ‘ineluctably coercive’ form of disclosure, thereby further undermining its therapeutic potential (Mackinley, 1997: p. 70).

Vocabulary and Narration

Differences in vocabulary and narration were referenced by some respondents as having the potential to obscure disclosure, and to thereby prevent narratives of sexual violence from being duly noted at key moments in the asylum process. At the same time, however, many others failed to reflect upon these potential linguistic complexities or refused to accept that they could explain apparent late or non-disclosure. In relation to vocabulary, respondents were asked if women ever used words or expressions that were unusual in order to describe their alleged experiences of rape. The question appeared to perplex some respondents, who had apparently never considered this possibility. Interpreters, however, provided ample examples of different phrases that might be used as euphemisms for sexual violence – ‘he hurt me’, ‘they destroyed my respect’, ‘my honour is tainted’, and so on. Several interpreters confirmed that, during the process of translating for asylum applicants, they might take it upon themselves to explain or modify these expressions, in order to transform them into the ‘right English’ for the benefit of listeners. Equally, however, there was some evidence that other interpreters adopted a rather different approach, and deliberately tried to avoid using ‘this horrible word [rape]’, even where it may have been used by the applicant, because it was so aggressive. Subtle transformations in emphasis and expression may be inevitable to some extent whenever discourse is mediated via a third party; they may even be desirable to the extent that they enable interpreters to more accurately convey the applicant’s intended meaning, in contexts where too rigid an insistence upon literal translation could be both unrealistic and a hindrance to true communication (Good, 2007). At the same time, however, the scope for miscommunication through this mediation is significant and can have extremely damaging consequences in an asylum arena within which, in order to ensure credibility, inconsistency is to be avoided. This was well exemplified in one case we observed where a miscommunication brought about by the use of an interpreter was taken to reflect an inconsistency in the substantive account and used by the UKBA to support a negative credibility finding. Here, the applicant – who alleged that police officers had exposed their genitals to her during an interrogation – initially told the UKBA that she had been sexually assaulted, but then told a doctor that she had not. At her appeal hearing, she explained that this discrepancy arose as a result of confusion regarding the meaning and translation of the questions addressed to her during the various interviews: ‘in the translation there seems to be a little difference but I could not get the gist of it. When the doctor asked me whether I was sexually assaulted, I did not quite understand, so I understood that harassment meant sexual intercourse and I am sorry. At the interview I said yes and then to the doctor I said no because I was not sure exactly what was meant’. While the appellant went on to advise that this error had been acknowledged by the interpreter – ‘I spoke to the interpreter in the presence of the immigration officer and the interpreter accepted that he had made a mistake’ – in his closing submissions, the UKBA Presenting Officer maintained that there was a ‘genuine discrepancy’ in the appellant’s evidence, which the Home Office ‘continue to think is significant’ since it both undermines her credibility and ‘detracts from the claim that she was taken to the CID headquarters and questioned there’.

Meanwhile, the issue of narration emerged in the present study as posing even more subtle difficulties. Blommaert’s analysis of the narrative inequalities that can operate to claimants’ disadvantage within asylum procedures illustrates how the ‘home narratives’ produced by applicants, although often vital in ensuring that their accounts are fully understood by decision-makers, can be dismissed as irrelevant ‘noise’, on the basis that they do not conform to dominant patterns of discourse (Blommaert, 2001). There was some evidence in our research that supports these concerns. One of our interpreter respondents, for example, described having observed similar ‘cultural non-communication’ when interpreting for Farsi speakers, who would speak of Persian history and their family’s status rather than provide direct factual responses to questions posed. She alleged that this was often perceived by the Home Office as ‘evasiveness’ on the part of applicants, when in fact they were simply trying to ‘paint a whole picture’ of themselves and of their circumstances in order to contextualise their accounts of persecution. Similarly, in another hearing that we observed, during which the presiding judge repeatedly interjected to assuage her concerns that the interpreter was not translating everything that the applicant was saying, the interpreter explained to the judge at the end of the hearing that Farsi women typically repeat the same point a number of times. Thus, in translating, she had taken it upon herself to streamline the repetition, notwithstanding the fact that there could have been considerable significance in the different ways through which a point was expressed during its various iterations.

Disclosure is inevitably a dialogic and interactive process – particularly where an interpreter is involved – but this generates several challenges. Ensuring the ability of the listener to ‘hear’ disclosures and to respond appropriately to the gaps or silences embedded within narratives is every bit as important as creating the formal opportunity for women to speak out in the first place. Likewise, factoring in a sensitivity to divergent cultural modes of narration is crucial in decision-makers’ search for ‘coherence’ and ‘relevance’, and it has to be acknowledged that resulting accounts of past experiences are always in a sense ‘mutually constructed’, such that not all seeming inconsistencies can be attributed solely or unreflectively to the applicant (Eisenmann et al., 2000: p. 108).

Engagement with, and Understanding of, the Asylum Process

This quote exemplifies the view, expressed by many of our respondents, that asylum claimants generally find the process difficult to understand and – contrary to the expectations of UKBA personnel – often fail to appreciate the importance of full and early disclosure. It is true, and well-reflected in the discussion above, that asylum structures and procedures are neither always transparent nor easily navigated. This can and does have very real consequences in terms of an applicant’s ability to disclose. At the same time, however, it is also important to bear in mind that disclosure is not the only choice open to women, and that a failure to do so – although likely to undermine the strength and/or perceived credibility of her asylum claim – may be the result, not of ignorance or misunderstanding, but of personal agency. Full or partial silence can reflect a survival strategy consciously chosen in order to (re)exert ‘control’ or, as one respondent put it, ‘to avoid being completely overwhelmed’. It can also reflect a deliberate calculation as to the likelihood of securing any benefit from the disclosure vis-à-vis the difficulties that its recounting may generate: as one NGO worker who supports trafficked women put it, ‘there’s the purpose of it as in “what am I going to say, what’s the point in saying this”’.

This goes to the heart of the resistant potential of disclosure, or indeed of withholding disclosure. In the context of asylum claims on the basis of sexual orientation, Johnson re-formulates (some) claimants’ silence not as a by-product of trauma or an indicator of disengagement but as a small but significant act that resists the demands of a potentially traumatising process (Johnson, 2011). In a system where women are often the subjects of forcible attempts to elicit their own life stories, perhaps the last power a woman retains is the choice to remain silent. One claimant in a case we observed, for example, when asked at her UKBA interview to disclose whether she had been raped, had stated simply ‘I don’t want to speak about that or tell it to anyone’. Her resistance was then specifically mentioned in an indignant UKBA refusal letter: ‘You … implied you were raped but refused to discuss this further despite efforts to explain why divulging this information would be relevant’. Although her failure to provide further detail clearly disadvantaged her application, she maintained her silence during her appeal hearing. A number of factors no doubt contributed to this response, but it can be seen that her silence does not necessarily denote incredibility nor indeed confirm her powerlessness, but may demonstrate an attempt to assert at least some control over her own claim. To the extent that such resistant potential and scope for agency exists, even in the midst of traumatic experiences, procedural constraints and the prospect of potentially life-threatening return ‘home’, it often lacked recognition by decision-makers and professionals. This in turn promoted too quick an attribution of silence to other factors (culture, trauma, lack of understanding or incredibility), which enabled more comfortable confinement of female asylum-seekers within conventional ‘victim’ or ‘liar’ modes.