Obviously without foundation: Discretion and the identification of clearly abusive asylum applicants

Abstract

This article looks at how asylum examiners in Brazil account for their ability to identify asylum applications they see as clearly abusive. The argument highlights a proclivity in examiners’ talk to shift discursive registers, from law and evidence to psychological and social bias, when talking about identifications they agree and do not agree with. This way of talking, it is argued, allows examiners to treat the refusal of clearly abusive claims as self-explanatory, while also acknowledging discretion as an inescapable feature of asylum decision-making. By relying on this ambiguous style of speaking truth, which portrays the outcome of claims as both dependent on and free of discretion, Brazilian examiners accrue authority to their views. Decisions to deny clearly abusive claims are rendered reasonable, in spite of barely being accounted for. What is worse, register-shifting kills the drive to ask what makes these denials so obvious by making the very question sound absurd.

Keywords

Asylum and abuse Brazil borders decision and discretion critical security studies enactment and refugees

Introduction

Brazil’s Federal Attorney sounded certain that something fishy was going on. Many asylum-seekers in Brazil were not asylum-seekers at all, it seemed. Although they claimed to fear persecution, these claimants were fraudulently using asylum requests to circumvent immigration rules. The press release talked openly about bogus applicants abusing the asylum procedure: ‘Refugee status is often sought in an abusive way, as a subterfuge by foreigners who do not have rights threatened in the country of origin and who, in fact, intended to immigrate to Brazil’ (AGU, 2017, my translation). This article looks into how examiners in Brazil account for their ability to single out these ‘clearly abusive’ and ‘manifestly unfounded’ requests, as the UN High Commissioner for Refugees (1983, 2010) calls them. We shall look at how examiners manage to convince others, as well as themselves, that denying protection to these asylum-seekers is the only option they have.

Decisions like these, to deny asylum, have occupied the attention of security and migration scholars in recent years. Studies now argue that decisions made by street-level border workers are assembled in multiple inconspicuous locations, which might range from desert stopovers to high-tech university labs (Doty, 2007; Bourne et al., 2015). Border decisionmaking seems increasingly enmeshed with dataveillance and probationary techniques, which are at work long before and long after a border is crossed (Amoore and de Goede, 2008: 8; Amoore and Raley, 2017: 4; Moffette, 2014). Decisions made at the border also seem to be affected by the behavior of all textures of matter, from desert dirt to databases (Aradau and Blanke, 2015; Nyers, 2012). Amid this scholarship, there seems to be an overall consensus that more research is needed on how decisions made at the border are authorized within this ‘array of technological, operational, occupational, organizational, social and political factors that may sustain, enhance or constrain discretion’ (Côté-Boucher, 2016: 51; Hall, 2017; Huysmans, 2014).

Taking heed of this call, I propose to dwell in this article on how the role played by discretion in border decisionmaking is being revised in Brazil, as migration law and the asylum procedure in the country undergo a complete overhaul. Adopting a localized perspective, characteristic of the scholarship on border security as practice, I ground the argument on the work of a small group of examiners formally responsible for the assessment of asylum applications.

The snippets of conversations used to build the argument were collected during 24 months of participant observation, interviews, and documentary research in Brazil. During this period, I was engaged in eligibility activities while working as an intern at Cáritas, one of the civil agencies involved in the asylum procedure in the country. I worked as an interpreter, attended asylum interviews, and even drafted legal recommendations myself. Taking part in these activities brought me into contact with examiners from all walks of life, ranging literally from nuns to cops and diplomats. Our conversations were sometimes arranged as formal interviews. Often, it was the things examiners told me over coffee that most puzzled me.

The remarks that follow pivot around what struck me as proclivity in the talk of these examiners. Speaking of cases that they saw as clearly abusive, my informants would explain their outcome by pointing to intrinsic qualities of the case, such as its lack of material support or legal conformity. In contrast, when talking about less clear-cut cases, in particular when they disagreed with the results, examiners would explain them as highly discretionary, referring to extrinsic social and psychological factors to explain the ‘error’ made by their peers. Borrowing a term from sociologists Nigel Gilbert and Michael Mulkay (1984), I use the notion of ‘register-shifting’ to refer to this way of talking, which jumps from law and evidence to social and psychological features when accounting for the identification of clearly abusive claims.

This Janus-faced way of talking has immense political implications for asylum-seekers, I contend. Thanks to register-shifting, examiners are able to convince others that denial is the adequate decision in face of a clearly abusive claim. What is more, register-shifting diverts us from even asking what makes the denial of a clearly abusive claims so obviously justified.

To lend support to my own statements about register-shifting, I shall examine a number of remarks concerning examiners’ ability to identify clearly abusive claims. Before moving on to this, there is just one caveat about the authority of writing that I would like to make. As is well known, the notion that ethnographers are able to represent the way their informants ‘actually behaved’ has long been unmasked as a stylistic accomplishment. Banishing the ‘I’ in writing and sanitizing reports from all things emotional are among the literary techniques shown to sustain the anthropologist’s authority over his subjects (Clifford, 1983: 100; Clifford and Marcus, 2010). Recently, critics of the literary norms of detached and impersonal writing have also become more vocal in defense of aesthetically conscious and emotionally invested forms of storytelling in international relations (Doty, 2004; Inayatullah, 2011; Dauphinee, 2013; Leander, 2015). In contrast, although often quoting this literature as inspiration, some of the works on how border security is practiced still give the impression that their authors were not there.

As will become clear, that is not how I have chosen to proceed. Although I do try to stay close to the words I heard and to the things I saw in Brazil, the stories below are not neutral, and I do not claim they are. I tell these stories with a practical purpose in mind. My hope is that attending to register-shifting will help turn the clearly abusive nature of claims—routinely invoked by examiners as a resource to justify denials—into a topic of study in its own right. The next section sets the scene for the argument discussing the birth of clearly abusive asylum claims as a policy problem in Brazil and other refugee-receiving states. This context will help establish how the concept of register-shifting may contribute to the study of discretion in border decisionmaking.

The birth of ‘clearly abusive claims’ as a policy problem in Brazil

Although rising asylum applications have certainly revived the issue, the concern with ‘clearly abusive claims’ is far from new. Framing the increase in South–North migration as a crisis requiring better filtering follows the same reasoning that allowed clearly abusive claims to first emerge as a problem, more than 30 years ago (Chimni, 2009). As early as 1983, UNHCR (1983: 01) described these cases as a ‘serious problem to both affected countries and genuine applicants.’ Member states were called to ‘deal in an expeditious manner with applications which are considered to be so obviously without foundation as not to merit full examination’ (1983: 02).

UNHCR member states have since broadened the notion to such an extent that there seem no longer to be any limits imposed on which claims can be considered as clearly baseless (UNHCR, 2010: 256). UNHCR defines manifestly unfounded or clearly abusive applications ‘as those which are clearly fraudulent and not related to the criteria’ laid down in the 1951 Geneva Convention (2010: 255). Yet member states rely increasingly on procedural grounds—like missed appointments or deadlines—to frame claims as abusive (2010: 259–260). UNHCR insists that, in principle, procedures used to identify manifest requests ought to prioritize ‘applications which are likely to be well founded’ (2010: 263). In practice, admissibility assessments, detained fast-tracking, and other policies adopted by member states seem to be geared almost exclusively towards the swift denial of clearly abusive claims (UNHCR, 2010: 249; ECRE, 2016: 47).

Against this background, ongoing attempts to modernize Brazil’s asylum procedure start to look somewhat different. In a sense, it is undeniable that status determination needs remodeling. The procedure followed by Brazil’s National Committee for Refugees (CONARE)—an agency comprised by envoys of the Justice, Labor, and other ministries, Federal Police officers, NGOs, and non-voting UNHCR delegates—is severely outdated. Total asylum applications in Brazil remain low in comparison to other destinations in the global South. But the rhythm of growth has been staggering. Submissions increased 35-fold in less than seven years, from 966 in 2010 to 33,865 in 2017 (Arcoverde, Sousa et al., 2018). Established in 1997, when Brazil was the destination of less than 500 asylum-seekers a year, the process has proven unable to keep pace (Gogolak, 2014).

Astonishingly, although requests remain at a record high, CONARE’s operational structure has changed little. CONARE’s eligibility team numbers less than 15 examiners to assess questionnaires, interview asylum-seekers, and prepare recommendations for all requests placed in Brazil’s 26 states and Federal District. Public defenders and lawyers hired by civil society agencies, like Cáritas, do the best they can to fill the void. Resources are so limited, however, that they soon find themselves overwhelmed by work. Meanwhile, CONARE’s Group for Preparatory Studies (GEP) and Plenary Board gather on a once-a-month basis to debate and vote on requests. Unsurprisingly, given these circumstances, CONARE has accumulated a sizeable backlog. At a rough estimate, if CONARE maintains its current assessment rate, it will need more than 40 years to respond to the 55,000 pending requests, as of October 2017 (Arcoverde et al., 2018).

Among the many reasons to take pause, it is at the very least curious to note that some substantial measures adopted to reform Brazil’s asylum procedure seem to be principally concerned with clearly abusive applications. The new legal category of temporary humanitarian migrant is a good example of this. In 2010, CONARE transferred responsibility for assessing thousands of claims placed by Haitian migrants to Brazil’s National Immigration Council (CNIg), alleging that Haitians were moved by economic and environmental drives, and therefore outside the remit of refugee law (Moulin and Thomaz, 2016). Instead of refugee status, CNIg granted Haitians a temporary visa on humanitarian grounds (2016). Similar forms of subsidiary protection were extended to Syrians in 2013 and to Venezuelans in 2017 (CONARE 2013; CNIg 2017). In May 2017, a change in Brazil’s migration law formalized the categories of temporary visa and temporary residency on humanitarian grounds (Law 13445/2017, articles 14 and 30).

Revealingly perhaps, the disjunction of asylum-seeker and humanitarian migrant was almost immediately exploited by Brazil’s Justice Ministry to reduce the asylum backlog. Arguing that the need for protection is satisfied through the award of temporary humanitarian residency, Brazil’s National Justice Secretary (SNJ, 2018) has suspended hundreds of asylum requests, ‘transforming’ cases into visa applications. Cutting the backlog down at the stroke of a pen, CONARE now boast about its ‘remarkable gains’ in efficiency (Arcoverde et al., 2018).

A similar concern about clearly abusive claims seems to pervade a number of measures adopted by CONARE in its search for more efficient asylum procedures. In 2013, as part of an agreement signed with UNHCR, Brazil’s Justice Ministry (2013) agreed to drastically reduce decision time frames, from a ten-month average to less than 90 days. In 2014, flirting with procedural grounds for refusal, CONARE (2014, Art. 6) asserted its right to suspend requests without an analysis of merit, when applicants miss two eligibility hearings or fail to inform the police of their address.

On March 2018, taking flirtation further, CONARE (2018) approved normative resolution 26 and included a new article, 6B, to its resolution 18. These measures now allow CONARE, by a majority vote and without analysis of merit, to suspend applications by migrants legally residing in the country. The change in resolution 18 also led to a subtle inversion in the administrative burden: asylum-seekers whose requests are denied now have 15 days to appeal. Bluntly, CONARE’s General Coordinator justified these measures as necessary to ‘prioritize proper asylum cases’ (Justice Ministry, 2018).

The insidious manner in which the notion of clearly abusive application is colonizing attempts to reform Brazil’s migration law and asylum procedures attests to a warning often repeated by migration scholars. As one commentator puts it, procedural reforms risk constraining asylum-seekers to such an extent that ‘the notion that a refugee applicant has the opportunity to present a claim and to be heard in good faith comes close to existing as a conceit only’ (Vogl, 2016: 140).

Discretion as an ambiguous style of speaking truth

The previous section suggested that, although cloaked in humanitarian clothes, reforms carried out in the name of efficiency risk tipping procedures against asylum-seekers. These reforms risk widening the already sizeable gap in power between examiners and asylum-seekers in the assembly of credible statements (Aradau and Huysmans, forthcoming). On the one hand, these reforms encourage examiners to treat inconsistency, lack of documents, or absence in interviews as evidence that asylum applications are baseless (Griffiths, 2012; Bohmer and Shuman, 2017). On the other, by reducing time frames for decisions, inverting the administrative burden, and allowing decisions on procedural grounds, they make it increasingly hard for asylum-seekers to contest denial (Magalhães, 2016).

As the British social theorist Nicholas Rose (1998) noted in relation to psychiatry, one should not underestimate the extent to which well-indented policy reforms can perpetuate power gaps. Ironically, as Rose notes, early advocates of risk-based psychiatrics justified this change in diagnostic method precisely on humanitarian grounds: ‘objective’ risk assessments would stop doctors from imposing detention and involuntary treatment on patients who did not need it.

Crucial to my argument is Rose’s insight that the discretion of mental health professionals is put at stake by this turn towards risk-thinking. If patients discharged today go on to hurt someone or die of neglect tomorrow, doctors now expect to be held accountable for their poor risk assessment. As a result, assuming that assessments might have to be defended, doctors change the way they account for diagnoses (1998: 189). Making sure their calls would be able to stand up to any attack in a court of law becomes a major concern for them. To achieve this effect of objectivity, Rose claims that doctors adopt a particular ‘style of speaking truth’, at times emphasizing and at others downplaying the discretionary grounds of their assertions (1998: 189). Without dwelling much on its implications for social control, Rose points to this ambivalent take on discretion as crucial to the production of credibility around statements concerning risky individuals.

Studies on how migration and border policies are carried out at street level have more to say about this style of speaking truth to which Rose alludes. Three lines of inquiry through which discretion has been turned into a topic are particularly relevant to this discussion. A first take has consisted in underlying the room for discretion in activities carried out by state officials in border spaces. Pioneer studies found discretion being used in decisions by immigration officer in US airports to refer migrants to secondary inspectors (Gilboy, 1991: 574–576). Analyses have also noted discretion in choices made by officers in the UK to check criminal records when assessing asylum claims (Weber, 2003: 251). Discretion has also been linked to decisions by officers not to inspect migrants crossing the US–Mexico border (Heyman, 2009). These studies seem to agree that discretion stems from legal loopholes created by conflicting instructions and vague expressions in immigration law and policy (Gilboy, 1991: 575). It is exacerbated by the lack of accountability and low standard of proof under which border practitioners operate (Pratt and Sossin, 2009).

Building on these insights, a second line of inquiry insists on the need not to reduce discretion to ‘a formless domain of uncontrolled action’ (Pratt and Sossin, 2009: 301). These studies ask how the void left by vague and ambiguous rules is filled by practical knowledge interactively produced (Côté-Boucher, Infantino et al., 2014: 198–199). A key feature in these studies is their refusal to treat border policies as ‘self-executing’, as if they could be given effect without the continuous need for situated interpretation (2014: 198–199). These studies give credit to the claim that frontline border professionals are ‘policy-makers’, in the sense that their actions become the policies they claim to implement (Lipsky, 2010).

A study on Belgian visa officers in Morocco is illustrative of this line of inquiry. Belgian and European law instructs examiners to assess the likelihood of applicants overstaying their visas. Nevertheless, examiners seem to treat the potential for legal long-term settlement as a more worrying prospect (Infantino and Rea, 2012: 71). In a similar vein, when asked to adjudicate if resident-to-applicant marriages were bogus, German residency examiners seemed to rely less on black-letter law than on tacit indicators, such as the age difference between spouses or couples’ ‘ethnic mixes’ (Eule, 2014: 28). Likewise, asylum officers in Spain and the UK seem to rely on factors like ‘gender’ and ‘level of education’ when assessing the credibility of asylum claims (Jubany, 2017).

The success of these studies in unpacking practical knowledge sets a background against which discretion comes out sharply as an ambivalent style of speaking truth. A study by border scholar Karine Côté-Boucher (2016) captures nicely what is distinctive about this line of research. Côté-Boucher tell us that, as Canada started adjusting to NAFTA, the adoption of electronic declarations and preclearance programs drastically reduced officers’ discretion over customs (2016: 54). According to Côté-Boucher, officers reacted ambivalently to this. They would sometimes insist that these technologies made customs decisions more efficient (2016: 57–58). Other times, officers would repeatedly present themselves as a reserve army, ready to act when technology failed (2016: 61). What made decisions worth taking seriously, they insisted, was the fact that they had been there, exercising discretion to correct computers’ mistakes (2016).

Looking at how border officers select migrants for inspection in Canada, criminologist Anna Pratt furnishes another example of discretion being studied as a style of speaking truth. Pratt (2010: 462–463) notes that statements produced by Canadian officers fall somewhere between ‘hunches’ and ‘scientific knowledge.’ Sometimes, officers justify their actions by placing emphasis on discretion. They make references to ‘craft-knowledge’—intuitions, gut feelings, and professional experience—and present their ability to exercise discretion as necessary to cut through data noise. At other times, when faced with criticisms about racial profiling in particular, officers try to convince others that discretion has nothing to do with their selection of migrants (2010: 472). To downplay discretion, they make their choices sound as scientific as possible, invoking hit rates, compliance/deception probabilities, and other risk prediction techniques (2010: 469). Pratt argues that this ‘discursive gymnastic’, which at times amplifies and at others attenuates the extent to which targeting is discretionary, is key to rendering officers’ suspicions reasonable (2010: 466).

Studying discretion as an ambivalent style of speaking truth strikes me as a particularly useful way of looking at how Brazilian examiners make sense of abusive asylum claims. Not unlike American psychiatrists and Canadian borders officers, Brazilian examiners seem to accrue authority to their views about clearly abusive claims by relying on an ambiguous style of speaking truth, portraying the outcome of these claims as both dependent on and free of discretion. Moving the narrative to Brazil, the next sections introduce the notion of register-shifting to grasp how this ambivalent way of relating to discretion comes across in my informants’ talk.

Register-shifting in asylum talk

The asylum procedure in Brazil is formally described as a series of bureaucratic steps. After arriving in the country, asylum-seekers have to fill out a refugee request form at any Federal Police station. Cáritas and other civil society agencies advise asylum-seekers on how to fill out these questionnaires. Status determination starts officially once the application is submitted. In the months that follow, asylum-seekers undergo eligibility hearings with CONARE officers, public defenders, and civil society lawyers. The recommendations they prepare are debated at CONARE’s Group of Preparatory Studies (GEP, in Portuguese), before being put to a vote at CONARE’s plenary meeting (Jubilut and Apolinário, 2008).

Predictably, the way cases unfold is hardly as smooth as this overview. Far from neutral, this checklist way of describing the procedure impacts the consistency ascribed to claims (Magalhães, 2017). Leaving this issue aside here, let us focus on a first snippet of examiners’ talk. In the scene below, envoys of different CONARE agencies are gathered in the annex building to the Justice Ministry in Brasília, taking part in a GEP meeting. Their task is to propose first-instance rulings for asylum requests, to be later ratified by CONARE’s plenary. GEP members discuss a list of cases during the day. If all members agree that a case is clearly unfounded, the asylum application is rubber-stamped for denial. When disagreement is strong, the case is dispatched for further inquiry. Mr B’s case was one of these.

Case number 33: Mr B, citizen of Colombia, requesting asylum to avoid persecution by the Revolutionary Armed Forces of Colombia (FARC). Cáritas’s lawyer argues that the case is manifestly well-founded and defends acceptance. CONARE’s envoy argues that the case is clearly abusive and recommends denial.

For CONARE, there are signs that someone prepped Mr B’s answers during the interviews: phrases used by him were identical in different hearings. The officer says Mr B also gave inconsistent statements: to Cáritas, he said he was hiding in Cali, while he said to the Federal Police that he was kidnapped. Commenting on the political context in Colombia, CONARE’s officer argues that guerrilla persecution is restricted to the country’s northern region and that the most recent reports on FARC activities contradict Mr B statements. For CONARE, the case clearly lacks foundation.

Cáritas’s lawyer concurs that the political violence in Colombia has decreased, but stresses that persecution against guerrilla adversaries is still fierce. Commenting on an Amnesty International report, she lists pages reporting ongoing human rights violations. In her opinion, whether or not there was direct persecution is not the key issue in Mr B’s case. Mr B’s claim is clearly well-founded given the conflict itself, and the context of widespread human rights violations it breeds.

Result: No agreement is achieved. Mr B’s case is set aside for further debate.

In the controversy surrounding Mr B’s claim, Cáritas’s lawyer and CONARE’s officer hold different views on the case, based on different ways of assessing the evidence. Both examiners refer to the same Amnesty International report as empirical footing but they interpret the report in rather different ways. Should persecution of a social group be given priority, as CONARE’s officer defends, or should we focus on the issue of human rights violations, as sustained by Cáritas? In a case like this, when the interpretation of evidence seems laden with different world-views, how is it decided whose reading is correct?

Disagreements like this might sound minor, but their repercussions are serious. The risk of denying asylum to someone who needs it is very real. And, for some examiners, the chance of granting protection to someone who does not fit the refugee definition is equally vexing. To avoid ‘trivializing the asylum institution’, as some officers are wont to say, examiners tend to be rather watchful of the justifications they give for denial. The ultimate goal, as CONARE’s coordinator once put it, is the ‘optimum functioning’ of status determination.

[A CONARE officer]

No one who deserves asylum should be deprived of it. And no one that doesn’t merit protection should be allowed to wrest it.

Examiners I talked to often stressed that an absolutely bias-free decision is but an ideal aim. In practice, many would say, it is very hard to avoid some degree of bias interfering with the judgment they make.

[A UNHCR officer]

There is a lot of talk about law and evidence and this and that, but, bottom line, people decide based on all sorts of things. Once I was defending this case and CONARE’s chair made a comment that didn’t sit well with me. I made the mistake of answering ironically and he got pissed off. That’s it. He argued for refusal. I didn’t even try to argue anymore.

Although they acknowledge that ‘all sorts of things’ interfere with assessments, few if any examiners, I would speculate, would dispute that the interference of psychological and social factors needs to be carefully controlled.

[CONARE’s coordinator]

We have this image of status determination, but we know, of course, that a lot of things that shouldn’t interfere end up interfering. That is why we have all these stages. We have officers conducting the interviews, we have GEP discussing the case again, and we have the plenary meeting. The idea is to filter these things out.

Examiners seek to filter out bias by listening to different perspectives, comparing different assessments, and repeating interviews again and again. As much as possible, the message is, the case must be allowed to speak for itself. For asylum decisions, I would often hear, ‘cannot be made for no reason.’

[CONARE’s coordinator]

It would be kind of stupid, you know? Everybody agrees the guy is a refugee now in October and then in December CONARE realizes that he isn’t. It wouldn’t look nice. It would look like we don’t have any criteria or that the judgment was made in a rush.

For many examiners to whom I talked, a strong symptom of bias during status determination was the acceptance of claims they saw as clearly abusive. If an asylum request deemed clearly abusive was not refused, this was taken as a symptom that the determination process had been crooked in some way. Conversely, if a decision to deny a clearly abusive application proved straightforward, this was taken as a sign that the judgment had been clean. The reasoning was simple: other examiners agreed with the decision to deny asylum because they also noticed its clear lack of empirical support and legal fit.

[A CONARE officer]

Like this Colombian guy we interviewed in Brasília once. This guy came in and said that he was being persecuted by the FARC and that he would die and all that. But he had nothing to show for it. If he were really being persecuted, he would have received death threats or gone to the police. But all he had was this mimeographed letter, without any stamp or signature, saying he shouldn’t give food to FARCS’s enemies. There wasn’t much to work with … anyone could have printed that.

For this CONARE officer, a mimeographed letter, ‘without stamps or signatures’, does not count as sufficient proof. For him, the case’s lack of empirical support is clear. The fact that such a ‘clearly abusive application’ was denied did not seem to require further explanation. The lack of supporting evidence was deemed enough to explain why the asylum request was refused.

Indeed, this assumption was so entrenched that, when asked to justify the denial of claims, examiners would often receive my request with a puzzled look. I could never say if what their faces expressed was surprise at my question, dismissiveness at its triviality, or genuine concern for my mental health. For how could I not see why the decision on a clearly abusive claim had been negative? Wasn’t it obvious? In cases where the lack of legal fit and empirical support is clear, denial is straightforward because the case’s lack of foundation is plain. What else is to be said?

For examiners, to ask why a clearly abusive claim was dismissed did not seem to make much sense. Now, the reaction was usually very different when I asked my informants to comment on cases in which the decision had not been straightforward. While assent around the denial of clearly abusive cases was explained in passing or not explained at all, dissent around the refusal of claims they saw as clearly abusive raised many questions. The outcome of these requests seemed to demand much more elaborate accounts. Prejudice, political agendas, economic interests, lack of education, and other non-legal and non-empirical factors were quickly identified to explain the ‘error’ made by their peers.

The particular form these replies would take varied considerably according to the examiner to whom I talked. Sometimes, the examiner giving the account would turn into a lay psychologist, shifting the register of discussion from the cornerstones of legal fit and empirical support to some sort of psychological diagnosis.

[A case owner]

These UNHCR people, my god, they are like doctors without patients. Have you heard this expression, ‘doctors without patients’? It’s like they are looking for someone to save.

Another reaction I came across often consisted in the contradicted examiner turning into a sociologist, explaining the wrong decision by referring to social pressures of different kinds—bureaucratic pressures, disputes for funding, a desire to gain authority, and so on.

[A CONARE officer]

I’ll be very honest with you: these guys are looking for attention, for money, you know? I’ll be very honest. They want to bring everybody in to justify their jobs here.

In the next section, we shall look in detail at how examiners develop remarks like these. For now, what I want to highlight in these accounts is this proclivity to ‘shift registers’, as sociologists of science Gilbert and Mulkay (1984) once put it. When a manifestly unfounded application is not treated as such, an explanation is sought in the social and psychological realm. In contrast, the denial of a clearly abusive claim is treated as self-explanatory, given what is deemed to be an obvious lack of legal fit and empirical support. Gilbert and Mulkay coined the expression ‘register-shifting’ to describe this sort of Janus-faced reasoning, in which truthfulness is taken as intrinsic to the claim while only error is to be explored.

It would be misleading to say that Gilbert and Mulkay identify register-shifting. Conversional analysts imprint registers when they want to highlight some quality they see as relevant in their informant’s accounts (Williams, 1999: 151). One might pay attention to a number of lexico-grammatical features when trying to do this: expressions used, ways of building phrases, particular words on which people rely on, and even the volume in which they speak. One might also stress the ‘conversational strategy’ deployed by speakers if they are, for instance, trying to teach someone, trying to make someone laugh, or trying to persuade someone that they are right. When Gilbert and Mulkay write about registers, they are pointing to this cluster of lexico-grammatical features that might strike an analyst as changing in different discursive situations (1999: 179–181).

Gilbert and Mulkay (1984: 51) coined the concept of register-shifting while studying a scientific controversy between natural scientists about energy. They imprinted two registers on the writing and talk of their informants (1984: 53). They argue that, to justify why their conclusions were correct, informants would rely on a two-step ‘empiricist’ register: first, they would present the accumulation of facts and the careful replication of experiments as the only way to produce reliable knowledge. Secondly, they would say their results followed unproblematically from the world as captured by well-executed experiments (1984: 57). In turn, when commenting on conclusions they did not accept, Gilbert and Mulkay’s informants would rely on a ‘contingency register’: first, they would acknowledge that science might be influenced by extrinsic factors, like personal inclinations, funding, and so forth (1984: 70). Finally, they would dismiss opposing views as derived from experiments tainted by these factors (1984: 63).

Gilbert and Mulkay warn that this jumping back and forth across registers limited the kinds of factors considered when explaining ‘correct beliefs.’ When talking about ‘mistakes’, informants would accept that explanations need to go beyond strictly scientific factors and consider the impact of sociological and psychological factors (1984: 168). However, when talking about their own ‘correct’ beliefs, the same informants would ignore the need to consider extrinsic factors. Their accounts were asymmetrical in regards to the kinds of variables analyzed: they were not open to the same ‘kinds of variables’ when accounting for correct and incorrect belief. This Janus-faced way of talking, Gilbert and Mulkay (1984: 166) warn, is a device by means of which scientists make their own views appear unproblematic.

In their talk, examiners I met in Brazil struck me as engaging in a version of this reasoning Gilbert and Mulkay caution against. The same examiners that would discredit their peers’ views as ‘polluted by discretion’ would resign themselves to the realm of law and evidence when talking about the consensual denial of clearly abusive claims. The patterning I chose to imprint over these remarks is meant to highlight this difference: if examiners tied the outcome of a clearly abusive claim to a lack of legal fit or evidential support, I placed this comment within the legal-evidential register. If the examiner tied the outcome to job-seeking or laziness, for instance, I placed this comment within the socio-psychological register.

Of course, my excerpts are far from exhaustive and could be arranged in different ways. I arrange them in order to highlight register-shifting, because I am convinced that this way of talking is both common and of high political consequence. Thanks to register-shifting, I believe, examiners manage to acknowledge discretion as an inescapable feature of border decisionmaking and dismiss the possibility that their own views result from discretionary assessments. The result is that the need to refuse a clearly abusive asylum application emerges as self-explanatory and no more questions are asked.

A la Molière’s opioid, which is said to put people to sleep because it is ‘dormitive’, an account that explains why a case is dismissed by attributing denial to its manifest weakness does not bring us very far. All it does is to move us around in an idle wheel (Latour, 1986: 265). I have looked so far at the basic argumentative strategy I associate with register-shifting. The argument moves on now to a fine-grained analysis of register-shifting in the talk of asylum examiners I met in Brazil.

Accounting for dissent

We are back at Cáritas’s office in Rio, in a crammed office-box built from plastic joints and grimy partitions used as an interview room. Amid the come-and-go of staff and asylum-seekers, Mrs H and I meet for the first time. She is a seasoned UNHCR examiner, having written many legal recommendations upheld by CONARE. After introducing myself, I tell Mrs H that I am curious about how examiners identify clearly abusive requests. I ask her about the main cues I should be looking for. Mrs H answers freely, sharing her views on CONARE and UNHCR.

[Mrs H]

What should matter most is always the objective evidence. Because the subjective judgment, you know… you shouldn’t try to guess whether the person is telling the truth. The law says that there just needs to be a reasonable chance that the person is fleeing persecution; that should be enough to acknowledge the person as a refugee. Now, this issue of subjective elements is very complicated. As we follow the interviews, there are indeed things about the case that UNHCR will know and that Cáritas will know and CONARE simply won’t know; the way the person is telling the story, if the asylum-seeker is crying or sad…

From the start, Mrs H places emphasis on her ability to balance objective evidence and subjective elements to sustain a legal recommendation. In a diplomatic tone, which she will soon drop, Mrs H reminds me that CONARE officers have limited contact with asylum-seekers. A consequence of this, she suggests, is that their judgments are not always so carefully made. Very subtly, Mrs H is already deploying the terms that will later allow her to characterize divergent opinions as mistakes.

[Mrs H]

So, for example, if CONARE rejected the case because, say, a man says he was a soldier but he doesn’t know the kind of weapon he had, for CONARE, this shows that this guy wasn’t really a soldier. But our work is to analyze the case more deeply. Was this man speaking the language correctly? Is it possible that he actually knows the gun’s name, but he doesn’t know how to say it in Portuguese?

In this intermediary step, Mrs H gradually centers the discussion on the examiner’s ability to follow procedural guidelines. If they are to build a proper judgment about an asylum request, examiners ought to comply with procedural good practices, putting in context eventual inconsistencies in an asylum-seeker’s narrative. Mrs H seems convinced that, as UNHCR officers spend more time with asylum-seekers, their understanding of cases is more procedurally adequate in that sense. There are many legitimate reasons for why an asylum-seeker’s narrative might be perceived as inconsistent or lacking detail. Mrs H describes UNHCR officers as being more aware of these possibilities. The interviews they conduct would then be more robust in light of procedural guidelines.

[Mrs H]

For example, in this Colombian guy’s case: he didn’t speak Portuguese well. He gave some information that, unfortunately, was later used against him. He gave the information and then we realized that he probably meant other dates and things like that. It was part of our work to show that, no, he wasn’t lying, that maybe he didn’t say the right dates because he didn’t speak the language well, or because he was suffering a trauma—this kind of thing. These are subjective elements, you see, but which can support the case. Unfortunately, CONARE officers don’t always consider this. They just assumed he was lying. So, this kind of subjective aspect we try to use really only when it clearly supports the case, because, if we leave it to CONARE, it can easily backfire.

Mrs H is speaking her mind in increasingly explicit terms. She charges CONARE officers with jumping to conclusions. They err—she puts it more explicitly now—because they do not attend carefully enough to special circumstances. CONARE examiners are charged for not being attentive enough to the procedural guidelines concerning trauma and language barriers that asylum-seekers might be facing. For that reason, Mrs H defends reference to subjective elements only when it is done in accordance with a tacit criterion of good faith—to consider special circumstances and cultural particularities that will help strengthen the case.

[Mrs H]

Unfortunately, what we see is that, for CONARE, if this guy tells his story and doesn’t cry, then he is automatically lying. But we know this is wrong. We know there are cultural differences, different ways of showing grief. Overall his story was perfectly coherent. The dates he gave were all confirmed by the research, the places where he says there were attacks, we know from humanitarian reports that there really were massacres in that city. But, unfortunately, even after UNHCR said, ‘Yes, he is a refugee’, this was a case that CONARE rejected because they decided based on this completely subjective thing. This happens many times. Even when you research the country of origin and this research proves 100% that everything this person says really happened, they conclude that the person is lying. But this opinion is not based on anything solid. So, you know, many times we see cases being rejected by CONARE because they can’t bother. It just happens. By the law, he should have been recognized. But he wasn’t. This is kind of unfair.

Through a subtle nudge in her argument, Mrs H lets social and psychological factors in. When it comes to sustaining her point of view, Mrs H’s account is two-sided, as ethically desirable as it may sound. Mrs H shifts registers to justify her recommendation and account for CONARE’s mistake. She jumps from law and evidence to society and psychology to account for the outcome she sees as wrong.

Asylum hearings are not light conversations and Mrs H does not pretend they are. Her account is vivid and sad. It speaks of trauma, of sadness, of hope. As a case for a liberal stance towards asylum, Mrs H’s argument was one I found easy to embrace. I would feel very different in my encounters with Mr J—a CONARE case owner and unblushing apologist for strict immigration rules.

Now, looking back at our conversations, I am tempted to describe my reactions as a loose version of what ethnomethodologist Harold Garfinkel (1991: 36) once called a ‘breaching experiment’: without really realizing it, I acted in our conversation in a way that violated Mr J’s expectations of a background understanding, making visible the practical reasoning needed to sustain a common perception. I was annoyed by what shocked me as xenophobic overtones in Mr J’s remarks. So, whenever he made reference to a taken-for-granted understanding of how decisions on clearly abusive applications were established, I made a point to ignore his expectation and asked the very awkward question of what he meant by clearly abusive. In the remark below, Mr J invokes the standard understanding of status determination as the work of assessing the claim’s legal fit and empirical support.

[Mr J]

To decide is basically to check if what the person states—if the things he tells—really happened in his country and whether they fit the law to be an asylum case. So, depending on what the person says, you think what you have to do in the objective research, and then you raise the information.

According to Mr J, to decide if an asylum claim is warranted is to check the statements made by the asylum-seeker against the reality in the country of origin; it is to consider the case against the refugee definition; and it is to evaluate the narrative against itself. In these three steps, it would seem, the examiner can check the legal fit, the empirical footing, and the internal consistency of the narrative.

[Me]

And in a situation like this [when disagreement is strong] what do you do?

[Mr J]

In cases like these, when we discuss a lot and don’t get an agreement, we send the case to the main meeting as an open case. It goes to the plenary. Then you have this issue of argumentation. It’s like I told you: those cases where there is disagreement go to GEP. They go for discussion. We disagree, then, you know, everyone will expose their arguments etc, like that. And it depends on who makes the better argument. When you have doubts about whether the case fits [the legal definition] to be a case of refuge, then we discuss the law. And if the doubt is about the objective evidence, we wait and we do more research.

Mr J speaks of dissent. A listener could be convinced by this quote that controversies among examiners are opportunities for a rational comparison of information and exchange of ideas, where the examiner who builds the best case wins the day. The insistence that divergence can be solved through extra research and cross-argumentation preserves the image of asylum requests as a rational enterprise, properly founded in law and evidence. The recognition that there are cases when examiners do not manage to overcome dissent is immediately followed by the suggestion that not all examiners read law and reality equally well.

[Me]

And this extra-research usually convinces other examiners?

[Mr J]

Sometimes it does. Not always. Like this guy’s case, for instance. He was from Colombia. I interviewed him and he told me that there were training camps in operation at his home city and he was afraid of being hijacked. That was in 2010. You see? This is just unlikely. So, the week this guy arrived I did some research at the US’s State Department website and there was no report about any camp in that city. But, guess what: the case went on to debate and the UNHCR’s guy kept saying that he had checked his internal reports and that these were in line with what the Colombian guy was saying and so forth. So, it was like, me saying the guy wasn’t a refugee and this UNHCR guy saying he was.

In the first comment, we have a politically correct suggestion that divergent opinions can be productively reconciled through rational exchange. This image starts to make way, in the second comment, for hints that the dissident UNHCR examiner might be basing his opinion on dubious internal reports. The ability of GEP members to appreciate which case is supported by trustable evidence and the proper interpretation of law is gradually brought into question. The discussion starts to center on the ranking of sources and their probative value. Which source should be given more weight—UNHCR’s internal reports or the reports by the State Department checked by Mr J? Mr J starts to build his account of why UNHCR examiners disagreed with his view.

[Me]

You disagreed completely, you mean?

[Mr J]

Exactly.

[Me]

And did you try to prove you were right?

[Mr J]

Well, I went on and checked again if the guy’s arguments were believable [note that the ‘guy’ now is UNHCR’s officer]. The case was discussed in GEP and the UNHCR’s guy kept saying that he had seen these reports etc … that the story made sense. So I went on and took a look at his report and, you know what, it was very clear. [Mr J pauses. He lowers his voice] It was very clear, mate. You know what I mean? These UNHCR guys, they really want to bring everybody in …

Mr J questions the credibility of the UNHCR officer, in what marks an important turn in his reasoning. The focus changes from ‘the Colombian guy’ to ‘the UNHCR guy.’ Mr J’s suspicion is not restricted to the existence of FARC camps anymore. The argument is refocused on the issue of how trustable the UNHCR officer is who is making use of the evidence. Mr J invokes honesty. He lowers his voice. He is about to say something that, although apparently sincere for him, also risks coming across as too blunt. His talk is moving beyond the limit of the legal and evidential realm. We are entering the psycho-social register.

[Me]

UNHCR … ?

[Mr J]

Yeah. I’ve seen it, mate, a UNHCR agent going to GEP and saying a lot of bullshit. Because people think, ‘Oww, it’s the United Nations, they can’t make mistakes’, but I saw it man. There was this report prepared by UNHCR’s office in Colombia saying like [pointing to a map of Colombia on a paper sheet], ‘These are the main regions where we have problems with the guerrilla’, and it said, like, ‘The Pacific coast, the frontier with Ecuador and the frontier with Venezuela.’ The report was saying this, very generally, in the first paragraph. Now [prolonging the word “now” as he speaks], in the third paragraph, it went on and said where exactly in these regions the problem was. Like, ‘It’s going on in cities X, Y, Z.’ And the city the guy was coming from wasn’t mentioned.

But you see what happened? The UNHCR officer brought this map to GEP, showing that the city was in one of these hot regions. The city was highlighted in the map and you could see that it really was in there. But what he didn’t say is that the report didn’t mention the city. He just quoted the first paragraph saying that Colombia’s Pacific coast was problematic. He didn’t mention the third paragraph. He stood there saying, like, ‘Yes, what this asylum-seeker is saying is confirmed by what we know about Colombia’, but you see? That isn’t truth! You see? He misled the committee. The decision was based on that information. But it was a rather generous reading, to say the least. Now, I don’t know. I don’t know if the guy lied or was just, you know … though, I must say, this happened more than once. All I’m saying is that this is very problematic. ‘Cause our circle is very small; we have very few people working in this area. We need to trust each other.

Mr J seems convinced that the UNHCR’s officer misled CONARE through an overgenerous reading of the report. Whether this was a mistake or a blatant lie, Mr J is careful not to say. At least not until his next remark when, plunging further into the psycho-social register, Mr J speaks his mind.

[Mr J]

The truth is, if really you want to know, that these guys, [lowering his voice again] well, these guys study like hell. They are like the military in Brazil. They prepare their whole lives to go to war but they never go to war. These guys are like this. Everyone has Master’s degrees, PhDs etc. But what we see is that there are very few refugees who need them. Take the Syrians, for example. Lots of Syrians are coming to Brazil. Why? Because they have a support network here already. These guys don’t need UNHCR. The Syrians in Turkey, the Syrians who are in Jordan, in Lebanon need UNHCR, you see? But the Syrians in Brazil? You see? There is why they vote for asylum. The more refugees the merrier, you see, so they will have something to do.

The change of registers in Mr J’s discourse is noticeable. Mr J’s account started with a politically correct praise of disagreement as an opportunity for a healthy exchange of ideas. It closes with a blunt assessment of what he sees as the real source of dissent. As if in a more intimate tone, Mr J shares his suspicions about the UNHCR examiner. He points to social pressures and personal self-interest that would have biased this examiner’s call. ‘Do you really want to know why he didn’t opt for the right ruling?’, Mr J essentially asks. ‘Then you have to look at this pathological altruism, to his desire to put his Master’s and PhD degrees to good use, to the very down-to-earth matter of guaranteeing his job and justifying the existence of UNHCR’s office in Brazil.’ We start in the realm of law and evidence and we end up in the much more somber arena of bias and manipulation.

Mrs H defends compliance with procedural guidelines as a way of guaranteeing a more inclusive assessment of asylum requests. Mr J argues for the diametrically opposed posture, defending strict compliance with the refugee definition to avoid ‘trivializing the asylum institution.’ Despite these differences, I suggest that both Mrs H and Mr J’s accounts have in common this two-sided way of talking: they justify decisions they agree with as a natural outcome of the case’s legal and empirical footing and they speak of what they see as a deviation from right decisions as a consequence of social and psychological bias. Both accounts shift register. While accepting the decisions on manifest applications as self-explanatory, they divert our gaze from how a decision to deny a manifestly unfounded claim is established as necessary.

Conclusion: Asking absurd questions

As asylum applications hit historic levels, Brazil and other Western democracies are turning to the identification of clearly abusive requests as something of a silver bullet to make asylum procedures more efficient. This article dwelled on how examiners in Brazil make sense of discretion when accounting for their ability to identify such claims.

As the scholarship on border security as practice has suggested, the understanding of discretion as ‘individual judgement constrained (but also enabled) by legal or policy rules’ is being revised, not only by scholars who research it, but also by professionals who resort to it (Hall, 2017: 500). Among authors who study border security as practice, an agreement now seems to exist that not enough attention has yet been paid to the impact of organizational reform and technological change for the work life of border officers, ‘especially in regards to their use, experience and perception of discretion’ (Côté-Boucher, 2016: 64).

Grounded in the work of a small group of examiners formally responsible for asylum decisions in Brazil, this article has drawn attention to one way in which asylum examiners seem to be mobilizing discretion in response to profound changes going on in migration law and the asylum procedure in the country. The argument sought to highlight what struck me as a proclivity in the way these examiners would account for their ability to single out clearly abusive claims. I described comments as unfolding within a legal-evidential register whenever examiners would tie the outcome of claims to factors such as lack of legal fit or evidential support. In turn, if examiners tied the outcome of claims to issues such as job-seeking or laziness, I described these comments as unfolding within a socio-psychological register. Register-shifting refers to the movement of jumping back and forth across these two registers when talking about the outcome of claims examiners agreed and did not agree were clearly abusive.

If apparently trivial, this small shift in talk proves highly significant for those men and women who have their calls for international protection applications wiped out as clearly abusive. This way of talking allows examiners to acknowledge discretion as an inescapable feature of asylum decision-making, while insisting that decisions to deny clearly abusive applications are self-explanatory to any unbiased bystander. As a result of this turn in register, the identification of clearly abusive applications is rendered reasonable, despite barely being accounted for. By treating the denial of clearly abusive applications as self-explanatory, register-shifting erases the work that goes into sustaining this strong sense of necessity around the decision to reject asylum-seekers. What is worse, it kills the drive to ask what makes the decision to deny asylum so obvious by making the very question sound absurd.

Footnotes

Acknowledgements

I would like to thank the men and women from Colombia, DRC, Haiti, Senegal, Syria, Venezuela, and way too many other sites, whose insistence on asking ‘absurd questions’ underlie my every line. Many thanks as well to the Brazilian eligibility officers who agreed to talk to me. For their helpful comments, I am most grateful to Elspeth Guild, Jef Huysmans, Carolina Moulin, and Isabel R Siqueira, as well as Security Dialogue’s peer-review and editorial team. For support, I thank the International Relations Institute (IRI) at PUC-Rio, CAPES, and the Centre for Citizenship, Identities and Governance (CCIG) at the Open University. This research was approved according to The Open University’s Ethics Committee Protocol HREC/#1038.

Funding

This research received funding from the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), award number 0620-10-6.

ORCID iD

Bruno Magalhães

Bruno Magalhães is a postdoctoral fellow at the International Relations Institute (IRI), PUC-Rio, Brazil, with a PhD in Politics and International Studies from the Open University, UK. He is co-coordinator of the IRI Methods Laboratory. His research connects the study of bordering practices with an interest in methodologies that attend to contingent, decentered and heterogeneous manners in which welcome and unwelcome subjects are set apart. Email: brunoepbm@gmail.com

References

Advocacia Geral da União (AGU) [Brazil’s Federal Attorney Office] (2017) Estrangeiros que solicitaram refúgio devem observar regras de retorno ao Brasil [Foreigners applying for asylum must follow rules when returning to Brazil]. 19 July. Available at: http://www.agu.gov.br/page/content/detail/id_conteudo/578604 (accessed 27 May 2018).

Amoore

de Goede

(eds) (2008) Risk and the War on Terror. London and New York: Routledge.

Amoore

Raley

(2017) Securing with algorithms: Knowledge, decision, sovereignty. Security Dialogue 48(1): 3–10.

Aradau

Blanke

(2015) The (Big) Data-security assemblage: Knowledge and critique. Big Data & Society 2(2): 1–12.

Aradau

Huysmans

(forthcoming). Assembling credibility: (In)security, critical knowledge and the politics of method.

Arcoverde

Sousa

Araújo

(2018) Brasil registra número recorde de solicitações de refúgio em 2017 [Brazil registers record number of refugee requests in 2017]. O Globo (Rio de Janeiro), 10 January. Available at: g1.globo.com/mundo/noticia/brasil-registra-numero-recorde-de-solicitacoes-de-refugio-em-2017.ghtml (accessed 18 May 2018)

Bohmer

Shuman

(2017) Political Asylum Deceptions: The Culture of Suspicion. London: Palgrave Macmillan.

Bourne

Johnson

Lisle

(2015) Laboratizing the border: The production, translation and anticipation of security technologies. Security Dialogue 46(4): 307–325.

Chimni

(2009) The birth of a ‘discipline’: From refugee to forced migration studies. Journal of Refugee Studies 22(1): 11–29.

10.

Clifford

(1983) On ethnographic authority. Representations 2(2): 118–146.

11.

Clifford

Marcus

(2010) Writing Culture: The Poetics and Politics of Ethnography. Berkeley: University of California Press.

12.

Conselho Nacional de Imigração (CNIg) (2017) Resolução Normativa No. 126 de 02 de março de 2017 (CNIg) [Normative Resolution no. 126, 2 March 2017]. Diário Oficial da União (Brasília), 3 March. Available at: http://www.acnur.org/fileadmin/scripts/doc.php?file=fileadmin/Documentos/BDL/2017/11016 (accessed 27 May 2018).

13.

Comitê Nacional para os Refugiados (CONARE) (2013) Resolução Normativa No. 17, de 20 de setembro de 2013 [Normative Resolution no. 17, 20 September 2013]. Coordenação-geral de assuntos de refugiados. Diário Oficial da União (Brasília), 21 September. Available at: http://www.refworld.org/cgi-bin/texis/vtx/rwmain/opendocpdf.pdf?reldoc=y&docid=54e72bb94 (accessed 27 May 2018).

14.

Comitê Nacional para os Refugiados (CONARE) (2014) Resolução Normativa No. 18, de 30 de abril de 2014 [Normative Resolution no. 18, 30 April 2014]. Coordenação-geral de assuntos de refugiados. Diário Oficial da União (Brasília), 30 April. Available at: http://www.justica.gov.br/seus-direitos/migracoes/refugio/anexos/resolucao-18-dou-pdf.pdf (accessed 27 May 2018).

15.

Comitê Nacional para os Refugiados (CONARE) (2018) Resolução Normativa No. 26, de 29 de março de 2018 [Normative Resolution no. 26, 29 March 2018]. National Justice Secretariat. Diário Oficial da União (Brasília), 10 April. Available at: http://www.justica.gov.br/central-de-atendimento/estrangeiros/refugio-1/resolucao-normativa-n-26-conare-1.pdf (accessed 27 May 2018).

16.

Côté-Boucher

(2016) The paradox of discretion: Customs and the changing occupational identity of Canadian border officers. British Journal Of Criminology 56(1): 49–67.

17.

Côté-Boucher

Infantino

Salter

(2014) Border security as practice: An agenda for research. Security Dialogue 45(3): 195–208.

18.

Dauphinee

(2013) The Politics of Exile. Abingdon, Oxon and New York: Routledge.

19.

Doty

(2004) Maladies of our souls: Identity and voice in the writing of academic international relations. Cambridge Review of International Affairs 17(2): 377–392.

20.

Doty

(2007) States of exception on the Mexico–US border: Security, ‘decisions,’ and civilian border patrols. International Political Sociology 1(2): 113–137.

21.

European Council on Refugees and Exiles (ECRE) (2016) The length of asylum procedures in Europe. Report, Asylum Information Database, European Council on Refugees and Exiles. Available at: https://www.ecre.org/wp-content/uploads/2016/10/AIDA-Brief-DurationProcedures.pdf (accessed 18 May 2018).

22.

Eule

(2014) Inside Immigration Law: Migration Management and Policy Application in Germany. Farnham, Surrey: Ashgate Publishing Limited.

23.

Garfinkel

(1991) Studies in Ethnomethodology. New York: Wiley.

24.

Gilbert

Mulkay

(1984) Opening Pandora’s Box: A Sociological Analysis of Scientists’ Discourse. Cambridge: Cambridge University Press.

25.

Gilboy

(1991) Deciding who gets in: Decisionmaking by immigration inspectors. Law & Society Review 25(3): 571–599.

26.

Gogolak

(2014) Haitian migrants turn towards Brazil. The New Yorker, 20 August.

27.

Griffiths

(2012) ‘Vile liars and truth distorters’; Truth, trust and the asylum system. Anthropology Today 28(5): 8–12.

28.

Hall

(2017) Decisions at the data border: Discretion, discernment and security. Security Dialogue 48(6): 488–504.

29.

Heyman

(2009) Trust, privilege, and discretion in the governance of the US borderlands with Mexico. Canadian Journal of Law and Society 24(3): 367–390.

30.

Huysmans

(2014) Security Unbound: Enacting Democratic Limits. London and New York: Routledge.

31.

Inayatullah

(ed.) (2011) Autobiographical international Relations: I, IR. London and New York: Routledge.

32.

Infantino

Rea

(2012) La mobilisation d’un savoir pratique local: Attribution des visas Schengen au Consulat Général de Belgique à Casablanca [Mobilising practical local knowledge: The granting of Schengen visas at the Belgian Consulate General in Casablanca]. Sociologies Pratiques 24(1): 67–78.

33.

Jubany

(2017) Screening Asylum in a Culture of Disbelief : Truths, Denials and Skeptical Borders. London and Cham, Switzerland: Palgrave Macmillan.

34.

Jubilut

Apolinário

(2008) Refugee status determination in Brazil: A tripartite enterprise. Refuge: Canada’s periodical on refugees 25(2): 29–40.

35.

Justice Ministry [Brazil] (2013) Acordo entre MJ e ONU irá aprimorar reconhecimento de refugiados no Brasil [Agreement between MJ and the UN to improve the recognition of refugees in Brazil].

36.

Justice Ministry [Brazil] (2018) Avisos Importantes: alteração no artigo 6º da Resolução Normativa Nº18 de 30 de abril de 2014 [Important Notices: change in Article 6 of the Normative Resolution no. 18 of 30 April 2014].

37.

Latour

(1986) The power of associations. In: Law

(ed.) Power, Action and Belief: A New Sociology of Knowledge? Sociological Review Monograph. London: Routledge & Kegan Paul, 264–280.

38.

Leander

(2015) Ethnographic contributions to method development: ‘Strong objectivity’ in security studies. International Studies Perspectives 17(4): 1–14.

39.

Lipsky

(2010) Street-Level Bureaucracy: Dilemmas of the Individual in Public Services. New York: Russell Sage Foundation.

40.

Magalhães

(2016) The politics of credibility: Assembling decisions on asylum applications in Brazil. International Political Sociology 10(2): 133–149.

41.

Magalhães

(2017) One thing left on the checklist: Ontological coordination and the assessment of consistency in asylum requests. In: Leese

Wittendorp

(eds) Security/Mobility. Manchester: Manchester University Press, 170–188. Available at: http://manchester.universitypressscholarship.com/view/10.7228/manchester/9781526107459.001.0001/upso-9781526107459-chapter-010 (accessed 27 May 2018).

42.

Moffette

(2014) Governing immigration through probation: The displacement of borderwork and the assessment of desirability in Spain. Security Dialogue 45(3): 262–278.

43.

Moulin

Thomaz

(2016) The tactical politics of ‘humanitarian’ immigration: Negotiating stasis, enacting mobility. Citizenship Studies 20(5): 595–609.

44.

Nyers

(2012) Moving borders: The politics of dirt. Radical Philosophy 174(2). Available at: https://www.radicalphilosophy.com/commentary/moving-borders (accessed 27 May 2018).

45.

Pratt

(2010) Between a hunch and a hard place: Making suspicion reasonable at the Canadian border. Social & Legal Studies 19(4): 461–480.

46.

Pratt

Sossin

(2009) A brief introduction of the puzzle of discretion. Canadian Journal of Law and Society 24(3): 301–312.

47.

Rose

(1998) Governing risky individuals: The role of psychiatry in new regimes of control. Psychiatry, Psychology and Law 5(2): 177–195.

48.

Secretaria Nacional de Justiça (SNJ) (2018) Despachos do chefe, de 26 de Janeiro de 2018. Departamento de Migrações. Divisão de Processos Migratórios [Dispatches from the chair: Migrations Department. Migratory Processes Division]. Diário Oficial da União (Brasília), 29, section I, 26 January. Available at: https://www.jusbrasil.com.br/diarios/175184211/dou-secao-1-26-01-2018-pg-29 (accessed 27 May 2018).

49.

UN High Commissioner for Refugees (UNHCR) (1983) The Problem of Manifestly Unfounded or Abusive Applications for Refugee Status or Asylum. ExComm Resolution No. 30 (XXXIV). Available at: http://www.unhcr.org/excom/exconc/3ae68c6118/problem-manifestly-unfounded-abusive-applications-refugee-status-asylum.html (accessed 27 May 2018).

50.

UN High Commissioner for Refugees (UNHCR) (2010) Improving Asylum Procedures: Comparative Analysis and Recommendations for Law and Practice. Available at: http://www.refworld.org/docid/4bab55752.html (accessed 18 May 2018).

51.

Vogl

(2016) A matter of time: Enacting the exclusion of onshore refugee applicants through the reform and acceleration of refugee determination processes. Oñati Socio-Legal Series 6(1): 137–162.

52.

Weber

(2003) Down that wrong road: Discretion in decisions to detain asylum seekers arriving at UK ports. The Howard Journal of Criminal Justice 42(3): 248–262.

53.

Williams

(1999) French Discourse Analysis: The Method of Post-Structuralism. London and New York: Routledge.