What’s in a Category? The Politics of Not Being a Refugee

Introduction

Being labelled or classified as a refugee is no trivial technical matter. The act of classifying people according to particular bureaucratic categories is politically relevant in multiple interrelated senses. First, these categories directly impact on migrants’ rights and, in extreme but not rare cases, on their chances of survival. Second, classification, as Nyers (2006: 7) reminds us, ‘the process of assigning characteristics or an identity to an entity – gives order to the world by marking off limits, assigning positions and policing boundaries’. Each migration category thus assumes specific subjectivities, shaping perceptions about political agency and needs in a particular time and space. Not only that, the categorization of people as ‘refugees’, ‘illegal migrants’, ‘economic migrants’, ‘(bogus) asylum seekers’, ‘humanitarian immigrants’ and so on does not correspond to an actual description of the different motivations and experiences of mobility at stake. Rather, they represent a political intervention which creates those categories according to particular and contingent assumptions of what constitutes ‘normal’, ‘desirable’ and ‘lawful’ forms of human mobility. Finally, the existence of different categories is inextricably tied to the state’s attempt to monopolize the ‘legitimate means of movement’ (Torpey, 1998). Categories of migration, and especially the divide between ‘legal’ and ‘illegal’ migration, reassert the state’s capacity to decide on issues of inclusion and exclusion from its bounded territory, reproducing its sovereignty and allowing it to ‘know’ or, to use Torpey’s term, ‘to embrace’, its population.

In sum, categories are relevant because they might determine whether people live or not, and under what conditions, because they ascribe identities and appropriate policies, and because they reproduce the authority of the categorizer (in the case of migration politics, the most conspicuous categorizer being the state). Zetter (1991, 2007) and Scheel and Squire (2014) use the terms ‘labelling’ and ‘figures of migration’, respectively, to refer to this eminently political character of classifying, governing and representing migrants according to historically contingent categories. Their work highlights the relationship between migration and attempts to control it as well as how each ‘figure’ or label is ‘related to particular framings of agency of people on the move’ (Scheel and Squire, 2014: 189).

In this article, I explore some of these key aspects about categorizing human mobility. I particularly focus on the refugee category and how it has undergone drastic changes since its formal emergence in the post–World War II period, in terms of both the perception of who is a refugee and how refugees are to be governed. The refugee category was created for the protection of Europeans who were fleeing totalitarian regimes. The figure of the refugee was originally built on a heroic image of mostly male and White individuals persecuted by oppressive regimes. Especially since the end of the Cold War, the figure of the refugee has changed to that of deprived, racialized and faceless masses mostly from the South fleeing extreme poverty and/or endemic conflicts. This new representation, so starkly distinct from the Eurocentric original formulation, is deeply related to the fact that access to asylum has been made ever more difficult (Chimni, 1998; Johnson, 2011, 2014). In sync with this rationality, I will show that in Brazil, Haitian asylum seekers have been represented through a depoliticized lens as deprived and racialized people affected by a natural disaster and that this depiction has justified their exclusion from the refugee category. In a context where states have been adopting a restrictive stance towards asylum seekers, Fassin (2013: 50) argues that they do so according to two main logics: repression (through interdiction, prevention, deportation, detention, dataveillance, etc.) and compassion (through the proliferation of precarious temporary ‘humanitarian’ statuses).

I argue that these two stances can be understood by considering not only how the ‘figure’ or the ‘label’ of the refugee has been associated with deprived and/or potentially dangerous migrants from the South but also with how states’ borders, boundaries and limits (Walker, 2010, 2016) have been reconfigured in the context of globalization. I explore how both states’ geographical borders and legal limits have been reconfigured in an attempt to control unauthorized migration and therefore reproduce their sovereignty. Particularly, I focus on the emergence of new legal limits that restrict access to asylum in the form of temporary humanitarian protection statuses. While the literature that critically analyses the exclusionary character of contemporary asylum politics tends to focus on destination states in the global North, I briefly explore some implications of this restrictive stance as it plays out in the context of the Haitian migration to Brazil and the emergence of ‘humanitarian visas’ to govern Haitians’ mobility to and presence in Brazil. I contend that Haitians’ categorization as ‘humanitarian immigrants’ is in sync with a broader, perhaps global trend in governing asylum seekers through new and precarious legal limits, while it also had important particular, even if unintentional, ‘side effects’ in terms of neglecting their welfare in the country while promoting a benevolent image of Brazil as a global South leader.

From Heroic Dissidents to Depoliticized Masses

The advent of the legal category of the refugee took place in the immediate post–Second World War context, through the establishment of the 1951 Convention Relating to the Status of Refugees (henceforth the 1951 Convention). The definition established in the Convention corresponded to the particular European context of the time, seeking to address the plight of those displaced by the recently ended world conflict and those who escaped the communist regimes of central and eastern Europe. The refugee phenomenon was regarded as transitory and mostly caused by totalitarian regimes that targeted and persecuted specific people. Hence, it defined as a refugee a person with a ‘well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion’ (United Nations High Commissioners for Refugees, 1951: 3) whose displacement was caused by events that occurred before 1 January 1951. According to the Convention, the referred events could be interpreted as those that took place exclusively in Europe or both in Europe and elsewhere, so that the signing parties could decide on the geographical extension of refugee eligibility they would take into account.

Importantly, the politics of the Cold War that was surfacing at the time influenced the formation of the refugee regime in the 1950s and how international protection was granted. As Loescher (1993, 2001) highlights, the fact that political persecution was the key element of the refugee definition served the interests of western states in their attempt to demonstrate the superiority of their democratic regimes, stigmatizing the eastern countries of origin of the refugees as oppressive. Therefore, dissidents from communist states were generally welcomed and offered protection in the West as heroic individuals, whereas other possible drivers for forced displacement, such as extreme poverty, natural disasters, armed conflicts or foreign invasions, were not considered as legitimate grounds for claiming refugee status.

In these initial decades of the refugee regime (1950s and 1960s), the refugee figure was, as Chimni (1998: 357) originally argued, represented as ‘White, male and anti-communist’. Johnson (2011) corroborates this argument by analysing the photo archives of the United Nations High Commissioner for Refugees (UNHCR), the UN agency in charge of overseeing the implementation of the 1951 Convention and of finding ‘durable solutions’ for refugees. Her analysis shows how refugees were originally conceived as heroic, powerful political individuals who resisted tyranny. The archival images of that time depict mostly men, sometimes accompanied by their nuclear families, the photos portraying their dignity and specific life trajectories. Particular representations or imaginations of groups of migrants (the labels or figures invoked) cannot be separated from the particular policies that are designed to govern them. The ‘preferred solution’ to the situation of these heroic individuals was either local integration or resettlement, what Chimni (1998, 2009) terms the ‘exilic bias’ of the refugee regime at that historical time. Evidently, the third possible solution delineated by the UNHCR, voluntary repatriation, was unimaginable given the particular characterization of the refugees as brave dissidents and their countries of origin as tyrannical.

However, the politics of mobility – which Squire (2011) defines as the struggles that characterize migration and the attempts to control it – changed in the following decades as the global context of forced migration went through some important transformations. The 1960s witnessed the outbreak of the struggles for national liberation in Africa and Asia, defying the geographical and temporal limitations of the 1951 Convention. Those were eventually lifted with the 1967 Protocol, allowing refugees from beyond Europe to be recognized as such. Chimini (1998) contends that in this post-1960 context, a ‘myth of difference’ was forged characterizing the refugees from the then Third World as markedly distinct from their European counterparts. They were considered to be unacceptably numerous and their reasons for flight were not seen as related to political individualized persecution as established by the 1951 Convention but to endemic national contexts of violence, war, ethnic clashes and/or extreme poverty.

Nevertheless, it was only with the end of the Cold War that the ‘myth of difference’ was invoked to justify institutionalized and wide-ranging restrictive asylum policies, consolidating a new paradigmatic figure of the refugee in the western public imagination as a depoliticized faceless mass of racialized people mostly from the global South whose right to protection in the North was always subject to suspicion (Chimni, 1998, 2009; Johnson, 2011). As refugees lost their ‘moral–political value’ with the end of the Cold War and more refugees from the South began seeking protection beyond their regions of origin, the label or the figure of the refugee was transformed. The myth was based on the assumption that the increasingly numerous flows of asylum seekers were fundamentally different from those of the anti-communist European refugees from before. The normalized heroic White male anti-communist individuals with a strong political voice were replaced by depoliticized, speechless and racialized and often feminized ‘flows’ and ‘floods’ from the global South. The contingent figure of the refugee or the refugee label went through a radical change: this category of migrants would now be met by an increasingly indiscriminate and exclusionary politics. Instead of local integration or resettlement, the dominant ‘preferred solution’ advanced by the UNHCR became voluntary, but sometimes forced, repatriation (Long, 2013; Lui, 2002). If successful in reaching northern territories, since they are portrayed as faceless masses from poverty-stricken and war-torn parts of the world, their claims to protection are subject to suspicion and greater scrutiny in eligibility assessments. ¹ As Griffiths (2012) shows in the case of the UK, assessments made by border agents are often informed by an explicit ‘culture of misbelief’ in which the asylum seekers’ statements are held to an unrealistic and arbitrary standard of ‘truthfulness’.

Agier (2005: 33) describes this shift in the governance of asylum seekers as the construction of a ‘great planetary segregation’. As the author maintains, this segregation is built on racist or ethnicist assumptions that construct an image of asylum seekers as impure, disorderly, and dirty. This image, an extreme and violent way of conceiving difference, justifies the construction of a symbolic wall to keep the ‘undesirables’ away. This wall is not just symbolic, as it is also concretely articulated in the myriad and increasingly sophisticated mechanisms that governments – especially of the global North – deploy to try to hinder their mobility, and in the legal manoeuvers they create to refuse granting refugee status. The discourse of security, in which asylum seekers are seen as threats, is combined with the discourse of humanitarianism, in which they are both undesirables and victims. Either as villains or as victims, their access to protection under the refugee regime is conspicuously circumscribed.

The Reconfiguration of States’ Boundaries in a Globalizing World

This new representation attached to the category of refugees as abject masses was accompanied by, and helped justify, a new more restrictive approach to the governing of refugee flows by host states predominantly, but not exclusively, from the global North. As will be discussed further below, different authors have referred to this new approach to governing refugees and highlighted particular aspects of it, but the common thread in these analyses is that effective access to asylum has become increasingly difficult for those who try to cross borders without previous authorization. Hathaway and Gammeltoft-Hansesn (2014), for example, describe this exclusionary stance towards asylum as constituting a ‘politics of nonentrée’. As they explain, states from the global North have, in the past three decades, developed a series of policies designed to impede most asylum seekers’ access to their jurisdictions, often co-opting states of origin and transit to achieve that goal, for example, by offering them financial incentives or providing equipment and training. Watson (2009: 1) lists some examples of policies enacted under this exclusionary logic, which includes ‘carrier sanctions, international safe havens, visa requirements, safe third country agreements, offshore processing, “nonarrival” zones, mandatory detention, temporary protection and the withdrawal of socio-economic benefits’.

However, in order to better understand this increasingly restrictive stance towards asylum, we also need to take into account the changing relations of state sovereignty and how new boundaries have been erected so as to reproduce state control amid globalizing flows – again, most notably in the global North. To put it differently, the contemporary ‘exclusionary politics of asylum’ (Squire, 2009) and the way it has been enacted by many states is best understood if we consider the changes undergone both in the figure of the refugee and in state sovereignty.

Since the late 1980s and early 1990s, researchers have been debating the accelerations, disjunctions and structural transformations affecting contemporary political life around the world. These dynamic processes, often subsumed under the term ‘globalization’, are related to a greater global interconnectedness through myriad and growing, but still select, flows of people, information, capital, ideas and goods across international borders. Globalization, as Scholte (2005) defined it, implies a shift in social space beyond mere territoriality, in the sense that the borders of states are increasingly less capable of containing social relations. This current process poses a challenge to state sovereignty and its key assumption of containing political community within a bounded and homogeneous territorial and legal space (Held and McGrew, 1993; Sassen, 2007). In the early 2000s, Lapid (2001: 17) highlighted that academic analyses on the shifting grounds of political life seemed stuck in a polarization pitting those who denied the existence of transformative changes (the partisans of ‘modal invariance’) to the ‘advocates of flux’, whose analyses only contemplated flows and mobility.

Moving beyond the dichotomy of the obsolescence or obstinacy of state sovereignty, a critical and interdisciplinary body of scholarship (sometimes referred to as ‘critical border studies’) has developed in the past two decades. This literature stresses how borders increasingly permeate the interstices of societies and how they are not mere static geographical lines but contingent practices (e.g. Anderson et al., 2009; Balibar, 2002; Mezzadra and Nielson, 2013; Parker and Vaughan-Williams, 2012; Rajaram and Grundy-Warr, 2007; Vaughan-Williams, 2009, 2015; Walker, 2010). Walker’s (2010) broad theoretical elaborations are particularly helpful in presenting some of the main assumptions of this field. The author claims that we need to problematize the regulative ideal of modern political boundaries as neutral lines that simply separate a pre-existing inside from an outside and that converge around the borders and limits of states. Walker proposes a reinterpretation of these demarcations as productive of what they seem to be separating. In this sense, the boundaries of state sovereignty do not merely separate one state from another, citizens from noncitizens or the domestic order from international anarchy, but they produce these very demarcations of ‘belonging and nonbelonging and authorize a distinction between norm and exception’ (Rajaram and Grungy-Warr, 2007: ix). In the process of constituting those fields, the authority of state sovereignty is reinscribed.

Additionally, Walker recommends a focus on how these demarcations are currently becoming disaggregated and proliferating beyond the conventional boundaries of the international system of states. Balibar (2002) offers a similar analysis when positing the contemporary ‘ubiquity’ of borders, that is, the fact that they can be found in many other places beyond the formal geographical lines between states. Mezzadra and Nielson (2013) likewise identify a proliferation of boundaries in contemporary political life. As they state, ‘The multiple (legal, cultural, social and economic) components of the concept and institution of the border tend to tear apart from the magnetic line corresponding to the geopolitical line of separation between nation-states’ (Mezzadra and Nielson, 2013: 2–3).

Hence, Walker (2010) maintains that in times of uncertainty and change, our gaze should turn towards what is happening at and to the boundaries articulated by state sovereignty, closely following what sort of politicizations and depoliticizations they engender. The author refers to boundaries as a broad term for the lines drawn by sovereign practices, more specifically, to borders as geographical demarcations and to limits as enacted in law and assigning identities:

Modern forms of sovereignty express and reproduce very specific ways of drawing the line, both literally and metaphorically. They do so both through claims to physical territory and through institutional expressions of (legal) principle. Boundaries are articulated as both borders and limits, as lines upon physical terrain and lines inscribed as limits of principle, jurisdiction and identity. (Walker, 2010: 101)

Hence, we can conceive migrant categories as bordering practices as they constitute ‘attempts by diverse actors to identify and control the mobility of certain people, services and goods’ (emphasis in the original, Vaughan-Williams, 2015: 6). Categories work as limits (in terms of law and subjectivity) imposed by states that do not work to objectively distinguish and filter pre-existing groups of people but to produce the very groups at stake. Boundaries, after all, enact ‘distinctions, discriminations and classifications’ (Walker, 2016: 15). Therefore, our analyses of boundaries should go beyond dichotomist distinctions between two sides and clear-cut dynamics of either inclusion or exclusion and instead be able to see how they ‘affirm much richer patterns of connection, complementarity, mutual production, recognition, transgression, classification, negotiation, antagonism, aporia, exception and violence’ (Walker, 2016: 3).

The Proliferations of Boundaries Around Asylum Through New Precarious Categories

When considering how access to asylum has been made increasingly more difficult, we can identify two main boundary strategies. One of them is through the enactment of new borders that, by being produced both within and beyond the official contours of states, work to deter, interdict and detain asylum seekers who dare to enact their right in international refugee law and cross international borders without previous authorization. This sort of response to their mobility, or attempt at mobility, evokes two main villainized images of the asylum seekers at stake: either as opportunists who ‘abuse’ the asylum system in order to obtain economic gain (the figure of the ‘bogus asylum seeker’) or as potential threats to national security (since 9/11, a threat associated with terrorism). ²

The other boundary practice aimed at constraining access to asylum is enacted through the creation of new limits, novel legal categories that allow for entrance and permanence but under provisional terms. Especially since the 1990s, temporary protection statuses have emerged and been granted to those who are not seen as legitimate refugees but who still, owing to their ‘humanitarian’ plight, should not be completely excluded. These precarious temporary protection statuses are given to asylum seekers who are perceived as victims, people whose manifest suffering demands some sort of provisional formal inclusion instead of outright exclusion. Temporary protection is a key element of the ‘repatriation culture’ that developed in the post–Cold War period (Long, 2013). A core premise of temporary protection is that the only possible ‘durable solution’ to displacement is the return back to the country of origin and that this return can be achieved in a near future, barring the possibilities of local integration or resettlement. In Europe, this mechanism was largely employed for the first time in the 1990s during the Balkan crisis (Durieux, 2014; Fitzpatrick, 2000). A justification for its widespread use, which was supported by the UNHCR, was the perceived magnitude of the flows and the understanding that it was ‘generalized violence’ the main driver for displacement instead of individualized persecution as stipulated in the 1951 Convention (Long, 2013: 117).

Comparably less discussed by the critical migration studies literature, the emergence of temporary protection statuses has also been an important mechanism through which forced migrants have been governed and state sovereignty has been reaffirmed. In order to explore this dimension of the contemporary politics of asylum, Zetter’s work proves to be helpful. Zetter (2007) argued that the increasingly restrictive migration stance adopted by states from the global North is also articulated through the fragmentation of the refugee label. This fragmentation is reflected in the current proliferation of labels for ‘temporary protection’ on humanitarian grounds. These labels are commonly applied whenever an asylum seeker’s case is considered not to fit the definition established by the 1951 Convention but, at the same time, for humanitarian reasons, it is understood that he or she should not be returned immediately to the country of origin, even though repatriation is kept as the necessary eventual outcome. As the author highlighted:

An increasing number of states are codifying generic forms of ‘subsidiary protection’ and ‘complementary protection’, in some cases called ‘humanitarian protection’ and ‘temporary protected status’ (TPS). These forms of protection are essentially the response of countries in the global north to the rising demand for asylum seen in the past two decades or so. The positive view is that these governments recognize that highly vulnerable people need protection even when refugee status has been or is likely to be denied. On the other hand, it could be argued that these supplementary forms of protection allow countries to reduce the volume of people receiving refugee status and the obligations this imposes, as none of these provisions afford the same level of protection as the 1951 Convention. (Zetter, 2015: 18)

Therefore, the current proliferation of temporary protection statuses might be seen not as a necessary answer to the current ‘globalized process and patterns of migration’ (Zetter, 2007: 174) but as a new limit erected by states in response to ‘flows’ of people perceived as undesirable (‘too numerous, too different, too dangerous’) and to the contemporary globalized reality that challenges states’ control over their territory and population. Even when asylum seekers are not perceived as threatening but understood through the marks of suffering and victimhood, they are still often times denied refugee status since they do not correspond to the normalized Eurocentric image of the political, individually persecuted refugee. Once they are seen through a humanitarian lens, as abject victims instead of as fully political beings, a precarious protection on the grounds of compassion is able to trump protection as a political right. Their formal inclusion is temporary, guaranteeing greater state control over their presence and fewer responsibilities in terms of assuring their protection and well-being.

As Bailey, Miyares, Mountz and Wright (2002) contend in the case of El Salvadorans granted TPS in the United States, this policy might work to extend the experience of displacement (it is a provisional and revocable instrument), as it encompasses a more limited set of rights than refugee status, and it frames migrants’ identities as characterized by a certain level of demonstrated ‘hardship’. By conceding a provisional documentation to asylum seekers, possible anxieties about their presence are therefore attenuated; they are, after all, helpless victims to whom one should show compassion and the burden they impose is temporary and relatively lighter. State sovereignty is thus (re)founded (Nyers, 2003) by determining their limited inclusion.

In order to explore this particular dimension of the contemporary politics of asylum, the next section briefly explores the Haitian migration to Brazil since 2010, when Haiti was hit by a high-magnitude earthquake. Moving beyond global North destinations, I explore how these asylum seekers have been categorized as ‘humanitarian immigrants’ and the implications of this framing.

The Productivity of the ‘Humanitarian Immigrant’ Category

The year of 2012 began setting a somewhat unusual tone for Brazilian debates on international migration. On January 1st, the front-page headline of Brazil’s main newspaper alerted that Acre, an Amazonian state bordering Peru and Bolivia, was ‘suffering’ a ‘Haitian invasion’ (Carvalho, 2012). According to the article, approximately 500 Haitian migrants had crossed into the country ‘illegally’ in the last days of 2011 pre-empting rumours that the new year would bring attempts to close the border and stop their mobility. The porous Amazonian border has historically invoked sovereign concerns in Brazil given the region’s rich biodiversity and sparse population (Becker, 2005), and these concerns were now mobilized against the unauthorized arrival of Haitian migrants. Between 2010 and 2015, Haitian asylum seekers entered Brazil in growing numbers. Their mobility to and within Brazil, the decision made about their legal status in the country and their precarious integration in society turned international migration from a dormant issue into a heated topic in Brazilian politics.

International migration has been an integral part of the Haitian reality since the revolutionary birth of the country in 1804 and today represents a key element of the country’s cultural and socioeconomic landscape. The 20th century witnessed the consolidation of international human mobility as a strategy employed by Haitian citizens to improve their standards of living and to guarantee their safety and survival in the face of internal conflicts, stark poverty and/or foreign interventions. The almost three decades of the Duvalier dictatorship (1957–1986) were particularly important in that regard. The regime was responsible for violent political repression and the aggravation of socioeconomic hardships in the country. In response, a transnational Haitian universe was established, connecting the ‘homeland’ of Haiti to Miami, Montréal, Paris, Guadaloupe, French Guiana, New York, the Bahamas and many other locations (Dubois, 2012).

Brazil was not among the main destinations for Haitian migrants throughout the 20th century. Haitian nationals did migrate in small numbers to Brazil and other South American countries following the outbreak of a severe political crisis in 2004, when President Jean-Bertrand Aristide was ousted from power and the Mission des Nations Unies pour la Stabilisation en Haïti (MINUSTAH) was established. Nevertheless, Brazil was gradually integrated into the transnational Haitian network after the earthquake of January 2010. The arrival and presence of Haitian migrants in Brazil was a notorious phenomenon in a country that has a relatively small number of migrants and refugees. Between 2006 and 2016, the number of new migrants being registered in Brazil per year increased by 160%, and this increase is mostly attributed to the recent arrival of Haitian migrants (Velasco and Mantovani, 2016). Their numbers were initially small, totalling less than 200 people in 2010, but in 2012, they reached a total of approximately 20,000 people (Fernandes and De Castro, 2014) and recent estimates indicate the presence of more than 80,000 Haitian migrants in Brazil (Portal Brasil, 2016). As will be discussed below, they are not considered refugees in Brazil, and their presence is approximately ten times more significant than the number of legally recognized refugees in the country, estimated at 8,863 people in April 2016 (Alto Comissariado das Nações Unidas para Refugiados, 2016). The Brazilian refugee population is predominantly urban and most commonly travels directly to the South and Southeast regions of the country, a situation much less visible than the hundreds of Haitian migrants crossing a seemingly vulnerable geographically and socially marginalized borderland in the Amazonian region.

The increasing Haitian migration to Brazil that began in 2010 represented a new reality for Brazilian authorities because these migrants were claiming asylum upon arrival. This represented a diplomatically sensitive case to Brazilian authorities because since 2004, the country had been actively engaged in peacebuilding and humanitarian efforts in Haiti, through which the government sought to promote an image of itself as a relevant ‘global player’ capable of taking leadership in South–South cooperation endeavours (Vigevani and Cepaluni, 2007). In fact, some authors attribute the fact that Brazil became a destination country for Haitian migrants to the increased political proximity between the two countries since 2004, when Brazil took on the command of MINUSTAH’s military component (Da Silva, 2013; Patarra, 2012). This engagement was accompanied by governmental cooperation projects as well as an increased presence of Brazilian NGOs in Haiti. According to this interpretation Brazilian military–diplomatic commitments and social projects constructed a positive image of Brazil among Haitians. The country was thus inserted in the emigration imaginary of those seeking alternatives once the 2010 earthquake struck.

The response given by Brazilian authorities to the asylum claims made by the incoming Haitian migrants was to grant them so-called ‘humanitarian visas’, an ad hoc and novel legal category in the country. There is no clear evidence that by giving Haitian migrants provisional humanitarian visas the Brazilian government was directly replicating previous policies adopted in Europe or in the United States through the TPS model, even though the rationale is the same in allowing for temporary stay in the country given humanitarian imperatives that are interpreted as going beyond the refugee definition. The National Committee on Refugees (CONARE), the governmental body in charge of asylum decisions in Brazil instituted by the national refugee law, rejected Haitian migrants’ asylum claims and referred their cases to the National Immigration Council (CNIG), a body regulated by the country’s national migration law established in 1980 during the military dictatorship and which is informed by a national security perspective (De Godoy, 2011). ³ However, a 2007 CNIG resolution had instituted the possibility of issuing humanitarian visas in cases that were not contemplated by the 1980 law on migration. The Haitian case became the first one in which this legal mechanism was employed to a whole national group, and this was a highly controversial decision. ⁴

The controversy revolved around the fact that Brazilian refugee law is considered to be progressive and broad by encompassing not only the 1951 Definition in terms of individualized persecution but also the expanded conception of the 1984 Cartagena Declaration, thus recognizing as refugees those who flee ‘severe and widespread violations of human rights’ (Lei 9474/1997: section III). The decision to frame them as ‘humanitarian immigrants’, instead of refugees, was celebrated by some actors such as the UNHCR (2011) as a positive example that allowed Haitians to remain outside their country during the turbulent post-earthquake context. Nonetheless, local NGOs such as Conectas challenged this decision, arguing that the situation in Haiti could be legally interpreted as one of ‘severe and widespread violation of human rights’ as stipulated in Brazilian refugee law and that each case should be evaluated individually (Roman, 2012).

The verdict on Haitian asylum claims was justified by Brazilian authorities on the grounds that their mobility was driven by ‘extreme poverty’ and ‘environmental causes’ (Leão, 2011: 88). CNIG’s Resolution 97, passed in January 2012, determined that Haitian migrants were to receive humanitarian visas owing to the ‘deterioration of living conditions of the Haitian population caused by the earthquake of January 12, 2010’ (CNIG, 2012: 1). The Resolution has been renewed every 2 years since 2012 and the humanitarian visa is valid for 5 years, being extendable as long as Haitian migrants are able to prove regular working status in Brazil. This shows one of the ambiguities of the ‘humanitarian immigrant’ category, since they are depicted as victims of a natural disaster in need of assistance but are expected to work in order to be allowed to stay for more than 5 years in the country.

As stated previously, bureaucratic categories, as boundary practices, have direct impacts on the population being classified while reinforcing the authority of the categorizer. However, before explaining such impacts of categorization in this case, it is important to clarify that I do not assume these to be necessarily intentional on the part of Brazilian authorities deciding on the humanitarian status of Haitian asylum seekers. Instead, following Ferguson (2006), I understand these effects as being part of the very logic or intelligibility that surrounds the humanitarian discourse deployed to govern Haitian migrants’ presence in Brazil rather than a direct consequence of government officials’ intentions. As the discourse of development analysed by Ferguson, the problematic of humanitarianism invoked by the Brazilian government works as an ‘antipolitics machine’, causing the ‘suspension of politics from even the most sensitive political operations’ (Ferguson, 2006: 273).

Hence, one important impact of the humanitarian category was to depoliticize Haitian migrants’ presence in Brazil and their claims to protection. During the Partido dos Trabalhadores (PT) governments, 2003–2016, especially the mandates of Luiz Inácio Lula da Silva, Brazil’s military and diplomatic engagement in Haiti was promoted as a demonstration of the country’s potential as a southern leader, and as supporting evidence of Brazil’s historical claims to a permanent seat in the UN’s Security Council. To recognize Haitian asylum seekers as refugees would have implied an acknowledgement of the existence of ‘severe and widespread violation of human rights’ in Haiti, which could put into question the efficacy of Brazil’s cooperation efforts in the country. By framing their mobility in terms of the devastating impact of the 2010 earthquake, Haitian asylum seekers are represented by a humanitarian discourse as victims of vicissitudes instead of as political agents who resist a specific oppressive reality. ⁵ The denial of their asylum claims is also related to the image through which refugees from the South are currently portrayed, as depoliticized, speechless and racialized flows. As such, they are not understood as meriting the political status of refugees, and the reasons for their mobility are attributed to internal and naturalized/depoliticized factors.

Additionally, Haitians’ categorization as ‘humanitarian immigrants’ keeps the number of Brazilian refugees low (as we saw if Haitian migrants were recognized as such it would lead to a tenfold increase in the number of refugees in the country). Thus, this boundary practice or legal limit works, as Zetter (2015: 18) highlights, ‘to reduce the volume of people receiving refugee status and the obligations this imposes’. It reasserts Brazilian sovereignty over its borders and decisions on inclusion/exclusion by normalizing the Haitian mobility as exceptional and temporary. Additionally, it helps promote a positive, compassionate image of Brazil as a global South leader, since the visas allow for Haitian migrants’ presence in the country, even under precarious conditions, as will be discussed. Instead of describing Haitian migrants’ motivations and experiences, this legal limit produces them as ‘humanitarian immigrants’ and depoliticizes their unexpected arrival in Brazil and Brazil’s own involvement in Haiti’s political context and stability. This new legal limit reinscribes a relatively numerous and diplomatically sensitive migration under an authorized but temporary category that provides the state with more control but also less responsibility over these migrants.

Another important effect of the humanitarian category has been Haitian migrants’ precarious integration in the country through the labour market, that is, their transformation into a vulnerable and temporary labour force in Brazil. Even though refugees in Brazil also lack a proper governmental structure able to assist them as they settle in the country (Moreira, 2014), ‘humanitarian immigrants’, as an ad hoc and ambiguous category, have no governmental, civil society or international institution dedicated to assisting them in Brazil, thus constituting an even more vulnerable group despite their victimized portrayal. Between 2010 and 2015, Haitians’ arrival in Brazil through the northern border led to the creation of a ‘humanitarian crisis’ in poor border towns in Acre unequipped to receive and shelter them (Conectas, 2013). The contradictions of exposing recognized ‘humanitarian immigrants’ to a humanitarian crisis through a politics of neglect are stark. Most migrants were only able to move from Acre to other states in the South and Southeast of Brazil once they were hired by Brazilian companies, mostly from the construction sector and meatpacking industry, which would send buses to Acre to pick them up. In the past 6 years, numerous cases of Haitian migrants being exploited and subjected to conditions tantamount to slavery have been reported by the Brazilian media (see, for example, G1, 2014; Puff, 2015).

These consequences in terms of inhospitable conditions at the border and exploitation in the labour market have been resisted and contested by Haitian migrants and civil society organizations that have been mobilizing to promote their rights in Brazil and to generate a broader debate on Brazil’s migration policies or lack thereof (Moulin and Thomaz, 2016). More recently, especially since the beginning of 2016, given the current aggravation of Brazil’s political and economic crisis, thousands of Haitian migrants have lost their jobs and decided to migrate to other Latin American countries as well as the United States (Semple, 2016).

Therefore, the categorization of Haitian asylum seekers in Brazil as ‘humanitarian immigrants’ had many crucial ‘side effects’: the depoliticization of the Haitian migration to Brazil considering the delicate geopolitical relations between the two countries, the creation of a humanitarian crisis at the border and Haitian migrants’ vulnerable insertion in the Brazilian labour market. As seen above, I do not assume that these effects were intentional or purposefully designed by Brazilian authorities when they decided on Haitian asylum seekers’ claims. The humanitarian discourse mobilized in this case speaks to a broader, perhaps global trend on how unauthorized migrants seen as ‘undesirable’ have been governed. This trend evokes a logic of compassion that denies these migrants the political right to asylum, while also reproducing state sovereignty through the creation of new legal limits.

Conclusion

As we reach historical records on the figures of forced migrants worldwide, access to refugee status has been restricted, not only in the global North, through ever more sophisticated boundary practices. As we saw, since the end of the Cold War, asylum seekers are perceived as apolitical, racialized, and deprived masses, and they are faced with increased borders and limits that reaffirm state sovereignty in a world marked by growing globalized flows. These borders and limits constitute a ‘great planetary segregation’ (Agier, 2005) or a nonentrée regime (Hathaway and Gammeltoft-Hansesn, 2014) that keeps those considered to be too numerous, too different or too dangerous either physically outside or provisionally and precariously inside destination states. Temporary humanitarian statuses can be interpreted as part of this contemporary regime. They respond to a historically constructed figure of the refugee who is understood to have no legitimate claim on political protection but who might be admitted under temporary and compassionate grounds. This status can also be understood as a boundary practice, a legal limit, that reinscribes the states’ power to decide on matters of inclusion and exclusion when anxieties about the possibility of drawing sovereign lines are raised.

Following Scheel and Squire’s (2014: 196) call to ‘neither simply adopt bureaucratic labels as analytical categories nor fully distance oneself from them’, I conceive Haitian migrants’ categorization as ‘humanitarian immigrants’ as a boundary practice that produces multiple but not necessarily intentional effects. The emergence of this category in Brazil might be seen as operating in chime with a rationality identified in the global North to treat unauthorized arrivals as generally undeserving of a political right to protection as refugees and to establish new legal limits to govern their presence on provisional and precarious grounds. As I argued, Haitian migrants’ framing in Brazil through a humanitarian category might be understood following this broader trend in governing asylum seekers, while it also had particular ‘side effects’.

Considering Brazil’s prominent role in MINUSTAH and other cooperation efforts in Haiti, to recognize Haitian asylum seekers as refugees would compromise the successfulness of the country’s engagement and aspirations as a southern peacekeeping leader. Another central effect of such framing was the neglect of Haitian migrants’ well-being through their precarious integration in society. Hence, in this case, the humanitarian category worked to govern an unexpected, relatively numerous and politically sensitive human mobility, producing the representation of Haitian migrants as depoliticized victims of a natural disaster towards whose welfare the Brazilian state had few responsibilities. As no particular governmental authority is accountable for assisting this ad hoc and provisional category of migrants, Haitians have been struggling for their rights and a dignified living in Brazil and, more recently, even beyond Brazil, by migrating to other countries that might offer better prospects.