Introduction

The present article analyses the participation of the Brazilian Army in the reception of Venezuelan immigrants and refugee seekers who arrived in Brazil via the state of Roraima. These Venezuelans have different statuses in Brazil, as they access different forms of legalisation. Thus, a referral to ‘Venezuelans’ covers all these legal conditions—undocumented immigrants have no access to shelters administered by the army. The main objective of this article is to expose the process of securitisation of the Brazilian migration policy via militarised humanitarianism.

The background is the economic and political crisis in Venezuela, which led to the emigration of citizens to bordering countries. There are more than 1,800,000 Venezuelan immigrants in Colombia (Palma-Gutierrez, 2021), more than 750,000 in Peru (Castro Quispe, 2019), and approximately 168,000 in Brazil (UNHCR, 2019). People come to Brazil, specifically Pacaraima and Boa Vista, lacking health, food, and housing, and are generally viewed by the government as a migratory crisis and social problem. Venezuelan sociologists Iván De la Vega and Vargas (2014) argue that the Venezuelan migratory pattern has been changing since 2006. Previously, there was a highly qualified flow to Europe, North America, and Latin American countries such as Chile and Argentina. Political instability coupled with the persistent economic crisis has recently led to the exodus of popular classes and indigenous peoples across border regions.

Paéz and Vivas (2017) divide recent Venezuelan migration into three phases. First, the former higher class (around 2000); a second in 2012, marked by the raw materials crisis; and with a very different profile migration (people from all social strata). The destinations of these migrations were mainly Europe and the USA, but the movement to Colombia, Panama, etc. had already started. From 2015, with the worsening of social conditions in Venezuela, the third phase, which he calls ‘migration of despair’, began, and the destinations were several countries in South and Central America.

The land route through the Brazil/Venezuela border, delimited between the municipalities of Santa Elena de Uairén (VE) and Pacaraima (BR), has become one of the most viable in the current stage of Venezuelan displacement, titled by Paez and Vivas (2017) as the ‘migration of despair’. The escape from Venezuela is an alternative to overcome the wounds caused by the stress of the humanitarian crisis, marked by a lack of food, medicines, high inflation rates, violence, and political repression. According to Paez and Vivas (2017) and Subero (2017), these migrants make up the poorest and have the lowest levels of education in relation to other layers of Venezuelan society that previously emigrated.Footnote 1 In this context, the so-called ‘Operation Acolhida’ was initiated, with three main goals: border control, reception, and relocation of the immigrants.Footnote 2 There is a double commitment from the armed forces: on the one hand, with the humanitarian reception of refugees and, on the other, the guarantee of law and order for Brazilian citizens. This second point reveals an international tendency towards securitisation of migration policies, reproducing the view that migrants are not only a social problem but, above all, a threat to national security (Stephen, 2018a, 2018b).

Fassin and Pandolfi (2010) argue that military interventions are increasingly occurring in terms of humanitarian moralism. In this article, we seek to look at the same principle, which organises actions within the state, in this case the Brazilian state and the new practice of militarising the humanitarian reception of the displaced Venezuelans. We will pay more attention to comparisons with what happens in other examples in Brazil and Colombia as a way of exposing the advance of militarised humanitarian logic in South America. According to Eduardo Domenech (2015: 27), what happens nowadays is not international mobility, but the establishment of new criteria of exclusion and discrimination, along with the reconfiguration of old jargon that legitimises migration control, such as the relation ‘between migration and the notion of security’ and the use of ‘humanitarian discourse’. We see precisely the replication of this discourse in the case of Venezuelans in Roraima, advancing in a conception of migrants’ management as public security and as control of ‘uncontrolled’ immigration. In the context of Operation Acolhida, we see a certain policy of sanitising public spaces by removing Venezuelans. Previously living in streets, avenues, empty lots, and abandoned buildings in September 2018, most Venezuelans were concentrated in eleven shelters scattered in the capital of Roraima and two in Pacaraima (about 214 km from the capital), on the border with Venezuela.

From a methodological point of view, the article is elaborated based on visits to shelters in Boa Vista and Pacaraima, and interviews with the Brazilian military, sheltered Venezuelans, and agents of non-governmental organisations, especially religious ones. During the second half of 2018, seven shelters were visited, in which seven interviews were conducted with military personnel of different ranks. Interviews were also conducted with residents and ex-residents of the shelters (a total of 10 interviews) and five leaders of non-governmental entities. In addition, a focal group made up of local internees was organised in one of the shelters. This work was additionally supported by ethnographic fieldwork among Venezuelans in Boa Vista between January 2018 and June 2019.

The article is divided into three parts: (i) an analysis of the issue of securitisation, (ii) the ethnography of shelters controlled by the armed forces, and (iii) a statement on the role of the Brazilian armed forces and the military in this process. If there are criticisms regarding the lack of competence of the military institution to fulfil the aspirations of a humanitarian reception, it is evident that the armed forces collaborated greatly with the logistical organisation of the host policies in Roraima. The personal relations established in the context of the operation lay bare a human aspect usually invisible by the patents and the uniform.

Securitisation

In the context of South America, Acosta and Freier (2015) state that migration policies and migration management policies in recent years have been organised by a ‘liberal populist discourse’. From this perspective, there would be a progressive discourse on migration, but for a security and discriminatory practice concerning immigrants from the global south who are part of this South American scenario. However, we demonstrate that immigration management practices in contemporary Brazil are openly based on the perspective of securitisation. The development of securitarian perceptions in Latin America has been recurring recently, as stated by Araujo and Eguiguren (2009), Magliano and Clavijo (2011), Stang (2016), Blouin (2021), Clavijo et al. (2019), Penchaszadeh (2018), Machado (2020), and Domenech (2017). These same perspectives have been described in national contexts on different continents, such as Huysmans (2000), Fassin (2005), Arifianto (2009), Ilgit and Klotz (2014), Menjívar (2014), and Stephen (2018a, 2018b). This global securitarian perspective leads to restrictive migration policies, produces a connection between displacement and crime, and considers migration as a threat to nations. These securitarian practices have also been described as contemporary forms of racism (Ibrahim, 2005). In this study, we understand a facet of this security perspective in military migration management policies in Brazil.

In the last decades, actions taken as humanitarian aid have been the justifying elements for the advancement of securitisation policies. Feldman-Bianco (2015), Piscitelli and Lovenkron (2015), and Dias (2014) indicate how certain humanitarian injunctions, such as combating human trafficking, are used to control the displacement of ‘undesirable’ subjects, while asserting a moralising anti-prostitution and humanitarian discourse. Another contemporary aspect of humanitarian action as a mechanism of social hygiene is refugee camps wordwide. Places of varied exceptions, isolation, and containment of foreigners are places that try to prevent the flow of people in different ways. Several authors claim that refugee camps are controlled by humanitarian discourses that serve to produce exclusion and containment of otherness. Fassin (2007), Agier (2008), and Agamben (1998) are some of the authors who reflect on refugee camps from different viewpoints. From a perspective that depoliticises them (Agamben, 1998) to a properly political anthropology of refugees (Malkki, 1995), as far as the refugee camp itself is concerned, there is an agreement about its separateness from the ‘normal’ world. It is a place where the rules and laws are different. There is a constant depoliticisation of refugees (Agier, 2010) built from humanitarian intervention agencies, but there are ambiguities present in these spaces, as Rancière (2004) emphasises, dehumanising biopower versus the capacity of resistance to this dehumanisation. Thus, we understand that securitisation processes imply the development of specific devices and practices (Bigo, 2002), in addition to the rhetoric of humanitarianism. In the case of this article, the practical result of these processes is the constitution of what we call ‘hybrid refugee camps’.

In Brazil, both the new migration law (law 13,445 of 2017) and the policies for regulating refugees follow the same logic of securitisation and hygiene (Machado, 2020). We will see here how Brazilian policies advance even more towards the constitution of foreigners as a place of distrust, the need for security, hygiene, and health control. The management of the refugee assistance in Brazil was, in general, outsourced to non-governmental entities, with financing from the Brazilian government or international entities such as the United Nations High Commissioner for Refugees (UNHCR). The refugee camp was refused as a model and a policy for outsourced management of refugees.

However, the militarisation experience in managing the refugees in Roraima presents a change in the Brazilian refugee policy, indicating an approach to the model of refugee camps applied worldwide. Although they are not stricto sensu refugee camps, many approximations can be made with the examples narrated by the above authors, indicating the constitution of a hybrid model of refugee camps under the figure of the shelters administered by the military in Roraima. We call this model a hybrid because it does not have all the usual characteristics of a refugee camp (Turner, 2016). However, there are structural similarities: in Roraima, shelters also produce a complete exclusion of immigrants/refugees/asylum seekers from the urban scene (Turner, 2005). There is entry and exit control and attempts to prevent shelter users from spending long durations of time outside; there is military control of the shelters; those in shelters are excluded (Agier & Lecadet, 2014) from Roraima society and shelters are places of exception (Agamben, 1998). The shelters also have differences: they are small in contrast to the traditional refugee camps, as there is a strategy to dilute them around the periphery of the city; there is not such a clear separation from the city (there are no fences); these are places imagined as being of rapid transition since those who live in shelters expect to be ‘interiorised’.

What defines the hybridism of this model of the refugee camp is not admitting itself as such (they are only shelters, from a military point of view) and to be structurally integrated into the interiorisation programme and, finally, the eventual movement of immigrants between different shelters. Thus, the ambiguity surrounding refugee camps (Turner, 2016) is more pronounced in the hybrid refugee camp. It is more difficult to distinguish them from the city; the movement of people is more fluid than in traditional refugee camps, which recognises a situation of exception but does not name it, and formally, they are part of a social programme of interiorisation, but many of the shelters are assuming the usual perenniality of refugee camps.

This has obvious implications for strengthening the perspective of foreign management through a securitisation policy under the aegis of fear, threat, and mistrust. The peace process in Colombia, Jimeno (2017) highlights how the fear and revulsion to the Venezuelan regime, for example, spurred a negative reaction to the agreement by identifying the FARC with the Maduro regime and fearing a ‘Venezuelisation’ of Colombia. The Brazilian scenario articulates similar feelings when dealing with the Venezuelan ‘refuge’, because the discourse of the fear of a Venezuelisation was fundamental in instrumentalising the ideological apparatus of the far right in the last Brazilian presidential elections. The uses were very similar to the ‘no’ campaign in the 2016 Colombian plebiscite (Gruner, 2017), instigating all kinds of fears. The sentiments mobilised about the Roraima/Venezuela border are of a similar order: fear and aversion, somewhat transformed into political action by militarising the reception of Venezuelans fleeing the regime and economic chaos in Venezuela. The militarised management of this population displacement puts Venezuelans under suspicion and deepens the militarisation in the region, a process that has already advanced in Colombia and Venezuela.

Violence against Venezuelans was seen in Roraima, with national prominence, even justifying the new model of militarised population management. Cases of xenophobia have been reported in many media sources (Oliveira Filho & Hilgemberg, 2020). While it is true that the military presence inhibits the kind of violence seen in Roraima, the very form of migratory flow organisation can be understood as state violence as it reduces the possibility of Venezuelan mobility and produces a policy aimed at public space sanitisation (a process in which the city needs to be ‘clean’ of Venezuelans, who become the object of policies of spatial exclusion, justified by a humanitarian discourse). The association of Venezuelans with the idea of plague is a process of criminalisation of immigration/refugees and is essentially the result of securitisation policies: producing security and, for that, producing the threat.

Ramírez, in an analysis of the militarisation of Colombian peripheral areas, states the following:

‘a confluence of illegality, counterinsurgency, and regional marginality led to the militarisation of social life, national politics, policy making, and state practices in peripheral areas and that this militarisation is now challenged under the peace agreement, which provides an opportunity to reverse processes of militarisation in marginal territories, making them and their populations not only visible but central to achieving peace. Thus, the implementation of the Havana Agreement provides a context for seeing the inhabitants of areas in conflict as citizens rather than guerrilla auxiliaries.’ (Ramírez, 2019, S135).

Here, the opposite of militarisation is the visibility of the subjects. Savell (2016: 60) points out that one side of humanitarian military action is precisely how it makes some groups invisible. The same can be said of Venezuelans in Roraima, who are rendered invisible under the rules of the hybrid refugee camps. The objective of this management is the invisibility of Venezuelans, either in relation to the cities of Boa Vista and Pacaraima—where they are properly sanitised—or in relation to the general movement of Venezuelan migration, which is expected to be dissolved in the Brazilian territory, avoiding concentration anywhere.

The idea that peripheral zone control is one of the facets of the migration securitisation process is evident in military intervention in humanitarian reception in Roraima. The model of conquering territories and military occupation is what we have seen happen as policies to combat organised crime in Rio de Janeiro’s favelas over the last decade. As Eliana Silva (2017) demonstrates, there is an explicit idea of ‘pacification’ behind the occupation of these places, extending to all residents the contrary idea of non-peaceful people. Although the target is crime, the whole community is the target of ‘pacification’ and therefore, criminalised as a whole.

Savell (2016) also indicates how the notion of humanitarianism and militarisation intertwine in the pacification experience of Rio's favelas. As in Colombia, the struggle for ‘control’ is justified by the fight against drugs and their cartels. As Harig (2019, 141) and Savell (2016, 64) point out, the military itself establishes a connection between the experience of the Brazilian armed forces as the main actor in the United Nations’ peacekeeping forces in Haiti (2004–2017) and the exercise of militarised humanitarianism in Rio’s favelas, which we claim to be the same model taken to Roraima: securitisation through militarised humanitarianism. Greenburg (2013, 96), in his analysis of the peacekeepers in Haiti, indicates that a general idea of ‘disability’ was linked to Haitians, in contrast to the self-attribution of modernity of soldiers from various nations (but mainly from Brazil). Pacification, therefore, is also a project of producing stereotypes of disability, victimisation, and vulnerability, as opposed to the ever-modern and strong definition of the self by the military.

Management and Military Organisation of the Shelters for Venezuelan Migrants in Boa Vista, Roraima

In September 2018, Boa Vista hosted a total of 11 facilities to house Venezuelans. Of this total, ten units were counted with the active participation of the Brazilian armed forces and only one unit was exclusively managed by a religious fraternity (although a military team was present at the site). Those with military services sought to follow the guidelines of the UNHCR and were called ‘shelters’. The other, under the responsibility of the ‘Fraternity Without Borders’ entity, was treated as a ‘reception centre’, demonstrating a clear distance between the two methods of reception. Of the 10 ‘militarised’ shelters, only two were managed exclusively by the armed forces; the others were managed in partnership with religious and laic non-governmental entities—cooperation that is mediated by UNHCR.

Each shelter works with a team of 10 military personnel. This team is divided into two subgroups that we were able to identify: (i) those who came from outside, consisting of two officers and two graduates who are permanently in the shelters and (ii) ‘local’ soldiers, that is, people already serving in the state of Roraima. Those from outside were assigned the coordination functions, carried out by sergeants and officers (lieutenants, captains, majors, and lieutenants-colonels). The “roraimenses”, in turn, mostly soldiers, were responsible for identifying exit and entry into the shelters, as well as operational tasks, such as food distribution and storage of internees’ belongings in the shelter.

The military team coordinating the shelter had air conditioning and reserved quarters, including a workroom, bedroom, and bathroom. The day-to-day administration work was carried out by the military in cooperation with an NGO selected by the UNHCR as the implementing agency, while UNHCR personnel moved between shelters, collected information, and performed sporadic actions. The different shelters were classified into segments: shelters for single men; for families with children; for couples without children, women, and LGBTQ population; and for indigenous people. The spaces reserved for the dormitory were distinguished according to the shelter category. In the shelter intended for single men, there were hundreds of individual camping tents lined up and dozens of bunk beds located next to the walls. In family shelters, larger tents were set up to house an average of two families. The indigenous shelters were equipped with a structure to hang several hammocks. There were no private toilets for the shelters. Usually, there were two shared bathrooms for use by hundreds of people.

Food supply was the responsibility of the armed forces. The exception to this rule was the shelter for the indigenous people, in which a structure was built for each family to prepare food in wood burners. Not all shelters had areas for dining, and people dined in their own tents. The daily life of shelters was regulated. Breakfast was offered at 7:00. Those who worked went to the streets, while others stayed on the premises. Teams of volunteer workers made periodic visits: oral health care, Portuguese classes, activities for children, artistic presentations, and so on. At noon, lunch. Dinner was served at 18:00. Entry was allowed until 22:00 (except for proven working situations).

All shelters were surrounded by walls and controlled by cameras. Some had electric and concertina fences. The shelter entrance was under the surveillance of two soldiers, 24 h a day. The entry and exit of people were controlled by an identification card that must be presented by every Venezuelan living in the place. During entry, bags and backpacks were searched for. Piercing and cutting materials, such as gardening and construction tools, were temporarily seized and returned the next day for migrants to work. The only forms of gathering allowed within shelters were religious gatherings and ‘cultural’ presentations. The occupants of the shelter could not gather freely in relaxed conversations.

The safety of the shelters was the responsibility of the army police who regularly patrolled around the eleven facilities. We could testify that the army police sought to ensure hygiene around the shelters to meet the requirements of the neighbourhood. Sheltered people were ‘oriented’ not to hang around the vicinity of the shelters, and the army police passed from time to time, having the personnel circulate or enter the shelter, besides carrying out random searches. It can be said that there was a certain disproportional use of the special police force, armed as if prepared for a confrontation. Military control in the outskirts of the shelters was reinforced in cooperation with the ‘National Force’ and the Battalion of Special Police Operations of the Military Police of the State of Roraima. Venezuelans, in turn, feared the truculence of control officers on the streets, bothered about surveillance cameras and clothing restrictions in the shelters (in Boa Vista’s hot weather, men are not allowed to walk shirtless, and women cannot wear short clothing), and resented the prohibition to remain at the doors of the shelters, situations that gave them the idea of being in prison. The controls over bodies, which directly affected the lives of Venezuelans, resembled a prison perspective.

When asked by the local population about alleged benefits offered to Venezuelans (shelter, food, and medicine), the military insisted on clarifying a somewhat delicate point of this welcoming action. It is not a matter of the Brazilian military providing welfare to the Venezuelans, but rather guaranteeing the welfare of the Brazilian population of Roraima. They do this by taking Venezuelans out of public spaces such as streets, sidewalks, avenues, vacant lots, and abandoned buildings, and housing all of this street population indoors, with time restrictions to go out and come back.

The use of force was legitimised by presidential decree n° 9.483, named ‘Guarantee of Law and Order’ (GLO), which confers police power to the armed forces. Originally published on 29 August 2018, it was limited to the border town of Pacaraima and was motivated by the violent assault of Brazilians against an improvised camping of Venezuelans by the roadside. They burnt tents, destroyed immigrants’ belongings, and forced their return to the country of origin. The case was widely reported by the national and international press as the response of the local population to a crime allegedly committed by Venezuelans (so far, unproven). The GLO was reissued on 12 September 2018 through Presidential Decree No. 9.501, expanding the military’s scope of action to protect host facilities in the capital, Boa Vista. This time, the justification was that two deaths occurred in the vicinity of a shelter. A new decree was published at the end of October 2018 (9.543, 10/29/2018), renewing the deadline until the end of the year. A legal apparatus was created to justify the day-to-day management of the shelters by the military as if they were common police, making it clear that the Venezuelan case was a matter for the military and not for the Roraima society itself, from the perspective that resulted in a planned disconnection between the civil society of Roraima and the Venezuelan migration.

Within the shelters, we observed a certain hierarchy involving the Venezuelans as well. There were ‘delegates’ or ‘collaborators’ among the sheltered people. They were entitled to mediate the relationships between shelters and managers, ensure compliance with the rules, and organise cleaning tasks and queues for food distribution. These people were, to a certain extent, empowered by the act of performing these mediations. They unequally distributed food and donations according to personal convenience. There were many complaints from Venezuelans regarding the authoritarian exercise of the leaders. The very stay in the shelter could be put at risk if the person disagreed with one of these ‘delegates’ that occupied a privileged place in the internal organisation of the shelters.

Although Operation Acolhida is called a ‘humanitarian aid’ by the military, their presence in Boa Vista divides opinions. The Sister of Charity in charge of the Caritas Arquidiocesana in Roraima warns that Operation Acolhida is a short-term mission that does not intend to leave any permanent structure for the state of Roraima.Footnote 3 This is an aspect that deteriorates the image of the military towards its civil partners in the host services, and a criticism of the lack of commitment on the part of local authorities to the situation of the Venezuelans. From the point of view of both religious and lay entities, when the armed forces leave, they take with them all the material support that guarantees a minimum welcome for immigrants and refugee applicants. However, it is clear that the provisional character of the shelters associated with the immigrant/asylum seekers ‘relocation’ policy corresponds to the desire to eliminate the problem in Roraima, as if it were possible to cleanse the city and the state from the presence of refugees.

Strangeness and Empathy Between the Military and Migrants

Compared with operations in Haiti, more experienced military personnel have observed that in managing the shelters, they establish greater personal contact with the ‘target audience’. In Haiti, the focus was on the defence and protection of international agencies’ officials—there was no room for interaction with Haitians. In Boa Vista, military personnel are in direct contact with the Venezuelans inside the shelters, developing bonds of solidarity and empathy for the hardships of the Venezuelans. Personal relationships also lead to the transgression of some regulations of the facilities such as ‘not taking food out of the shelter’, ‘not coming in after hours’, ‘not marketing products’, among other rules that are ‘reinvented’ according to the context. The story of Lieutenant A. is illustrative:

With his eyes watering, Lieutenant A. pointed to a photo of a Venezuelan family on his cell phone. He says that he ‘fell in love with them and did everything to help’. This was Mr. M.’s family, one of the sheltered people whom he had made friends in the first month he arrived in Boa Vista. Cohabitation in the day-to-day activities of the shelter allowed Lieutenant A. to know Mr. M.’s life trajectory, suffering, and obstacles encountered for family maintenance at a distance. It had been established that, for structural reasons, the shelter coordinated by Lieutenant A. could not receive children, only men, and single women at that time. One day, Mr. M. showed up with his wife and two daughters in front of the shelter. One girl had a fever, and her arm was plastered. Sensitised and contrary to the orders of the superiors, the lieutenant welcomed all the family members, but not without causing unexpected situations. Other families appeared for shelter, a fact that exposed the attitude of the lieutenant before his superiors.

A new decision was made with a broader impact after this incident was communicated to his immediate boss. The colonel sent the family of M. and all other applicant families to another shelter intended to receive families. The number of vacancies in the shelters is restricted, so any decision to relocate people implies the spread of new adjustments and relocations. It started because Lieutenant A. decided to help his Venezuelan friend. Lieutenant mediated a kind of ‘anticipatory relocation’, not content to arrange a temporary shelter for these people. He bought a ticket with his own money and arranged a job for Mr. M in the countryside of São Paulo State. While his friend settled in a new job, Lieutenant A. tried to send his friend’s family through the federal government-sponsored relocation programme.

One segment of the Venezuelan Relocation Program is managed by the military (there is also a UN-managed programme). Thus, relocation was managed according to personal relationships established within the shelters. Responsible for drawing up lists of who goes and where they are going to, the military does not hide the fact that they prefer their friends to go to destinations close to their own homes, indicating a desire to extend relations beyond institutional welcoming. Flexibility of regulations in the name of personal relationships is not limited to financial aid and support for travel. Lieutenant A. said that Venezuelans came and went from shelters with the mission of carrying food to relatives in Venezuela. Maintaining relations with the shelter as a point of support manifests the desire not to leave Roraima on the part of some migrants. To leave the state means to stay even more distant from the family in Venezuela, preventing periodic visits and the personal sending of remittances.

With the relocation process, the shelter became a migratory tactic to speed up the exit of the state of Roraima. That was clear when during the visits to the shelters, we found Mr. R. interlocutor who we have been following since 2016. We were surprised to find him in the shelter for single men, as we knew that Mr. R. had brought the two children, the daughters-in-law, and grandchildren to live with. Later, Mr. R. confided that he had daily contact with a son and daughter-in-law, but decided to go to sleep in the shelter in the hope of getting to the south of the country.

However, the relationships established within the shelters are limited by the maximum length of stay of the military, which is temporary. The personnel sent to Roraima to participate in the ‘Operation Acolhida’ follow a kind of rotation system. Every 90 days, the team is changed. Every 40 consecutive days, they have 10 days off to visit their families. These military personnel were recruited voluntarily from different parts of the country. At the time of the fieldwork, those in Roraima were predominantly from the southern region. They were preceded by military personnel in the southeastern region. This rotation was motivated by the previous experience of the Brazilian armed forces with the peacekeeping mission in Haiti (Harig, 2019). On the occasion, it was verified that the extension of time served in coexistence with people in situations of vulnerability implies a certain emotional destabilisation of the military. The interviews with the most experienced officers highlighted some concern with PINO, an acronym in Portuguese that represents different emotional stages developed by the individual in the prolonged contact with the vulnerability of others, namely: Pity, Indifference, Disgust and Hate (Pena, Indiferença, Nojo e Ódio).

Lieutenant A. says, ‘Ninety days is good. More than 90 days, a person will get very attached to the mission. Sometimes the person starts to get tired and, if there is no escape area, they begin to get angry’. Captain Q., in turn, emphasises that this is a guideline of the institution based on academic criteria, ‘There are studies demonstrating that the military, after a longer period than this, begins to acquire post-traumatic stress due to the situation of impotence’. The Lieutenant and Captain’s statements emphasise concern about the psychological aspect of the military in this type of ‘civil action’ and/or ‘humanitarian aid’.

Evaluating the position of the Venezuelans over their shelters, we understand that reciprocity exists in the relationship of friendship. The first impression of men and women regarding the Brazilian military is quite positive, considering the comparison they make with the national guard of their country. It is important to understand that Venezuela is undergoing a process of militarisation, and the reference of the armed forces by the Venezuelans is so bad that they are impressed by the treatment offered by the Brazilian military. We have noticed gratitude on the part of the Venezuelans in relation to the managers of the shelters. Even complaining about repetitive food and the truculent treatment received outside, Venezuelans emphasise their friendly relationships with the soldiers they live with daily. It is important to emphasise that this is not a generalisation, but a specific case of interpersonal relationships.

The Perspective of Civil Society

The immediate management of the shelters is conducted by the army. However, the military follows the guidelines offered by the international cooperation institutions, more experienced when it comes to the reception of immigrants/asylum seekers. Before the arrival of the armed forces, the reception of the Venezuelan migrants in Roraima was carried out predominantly by the so-called local organised civil society (OCS), a broad category that brings together non-governmental organisations, churches, unions, and other collectives. The state government provided logistic support. The OCS was mostly represented by religious entities, in partnership with professors and students of the Federal University of Roraima (UFRR), and international agencies performed welcome actions. During this first moment, spaces of articulation, mobilisation, and debates on the theme were created. The Committee on Migrants and Refugees (COMIR) represented one of these spaces. COMIR was initially composed of 40 entities. COMIR meetings were held periodically and subdivided into working groups organised on specific themes coordinated by international agencies and social movements, such as labour, indigenous peoples, women, and gender. In March 2018, the start of Operation Acolhida was marked by a lack of dialogue with pre-existing actions, according to the religious representatives of Cáritas in Roraima:

‘When the army came here in March, they did not count on us to think together and they started to set up the structure and do things from top to bottom... with the Operation Acolhida, there was a concentration of information and demobilisation of the host network of organised civil society.’

In the search for dialogue in May of the same year, mediated by the UNHCR, OCS began to participate bi-weekly in the meetings of the operation. However, these organisations did not find space for the debate and reflection of actions within the scope of the operation. For the coordinator of the Institute of Migration and Human Rights, the agenda is always the same: shelter, relocation, and documentation. The federal government’s appeal to the Brazilian armed forces has also modified the relationship between the UNHCR and international agencies with the local OCS. Previously crucial to the execution of actions in the shelters by collaborating in the capture of volunteer labour, they have come to a less preponderant role in the spaces of interlocution and decision making. The management of the shelters is the privileged locus for observing this transformation: access to the facilities was bureaucratised, presence monitored, and contact with the people mediated by the managers of the shelters. Foreign organisations were invited to take on services previously offered by local groups, and these, perceiving a certain closing of doors, have modified their focus of action on prioritising Venezuelans who were not sheltered.

In the treatment of immigrants and asylum seekers by the armed forces and international organisations is possible to see a version of the ‘guardianship’ model—as practiced by Brazil in relation to the Indians before the 1988 Constitution (Silva et al., 2018) —in which the autonomy of the refugees is withdrawn in the name of an alleged protection of the Brazilian society. Another local entity, the Centre for Migration and Human Rights (CMDH), also linked to the Catholic Church, has a critical position regarding the militarisation of the reception service. The sister who coordinates the CMDH points out that it is a return to previous forms of dealing with migration, with an emphasis on security and defence issues, ‘The new migrant law makes it clear that the migrant is not a threat to the state. Why would the Ministry of Defence be triggered? There is no threat to the state”. She also highlights the lack of articulation between the different power spheres, suggesting that a dispute is taking place not by the offer of the reception itself, but by the power to manage federal resources and a prominent position in the spotlight.

Aspects of the Securitisation Policy

The example mentioned in this paper presents the border as a test place: a test for displacement control. It is also a continuation of a practice that begins at the centre: the stratagem of calling on the army as a population management police. The case is obviously derived from the military interventions in Rio de Janeiro (Silva, 2017; Savell, 2016), but is even more directly related to the intervention of the Brazilian army in Haiti (–2004/2017), where soldiers’ stress management experiences were developed and applied to the case of Venezuelan refugees (Harig, 2019). Hirata (2015) points out that border security policies in Brazilian borders have been militarised with the presence of armed forces in combating drug trafficking and the emergence of various ‘illegal markets’. What we have seen in this article is that militarising the flow of asylum seekers and immigrants can be understood as an extension of this movement, incorporating the displacement of people in the same logic that justifies the presence of the army in the fight against illegal markets. Violence, illegalities, epidemic dangers, and all sorts of criminalisation were linked to the displacement of Venezuelans in Roraima, authorising the intervention of the armed forces in the management and containment of this movement through the security logic.

The hybrid refugee camp is a humanitarian technology (Gilman, 2012, 174) developed in Roraima to activate militarised humanitarian aid, with all the effects of securitisation on migration management processes: criminalisation of the movement, on the one hand, and political gains for the armed forces, seen as ‘humanitarian’, on the other. The subjects trapped in this web of securitisation try to circumvent the practical effects of immobilisation and helplessness with individual strategies, as we have seen, seeking better conditions within the shelters, better living alternatives with the help of friendly military, searching for relocation and also the use of camps to maintain a stable place in Boa Vista or Pacaraima and avoid relocation. Operation Acolhida empowers the armed forces (both politically and financially) and the criminalisation of Venezuelans. Another important issue is to highlight the fact that Operation Acolhida is also an operation of identification and registration of all immigrants that pass through the shelters, conforming to a vigilant humanitarianism, a control device.

As a background/image-obviated game (Wagner, 1978), it is possible to say that the militarisation and securitisation of humanitarian assistance is only possible with the obviated creation of its object, which in turn legitimises and calls for assistance. Militarisation is the process that creates the immigrant as vulnerable, miserable, and dangerous, as contagion and pest to be controlled. As a sign of this vulnerable/dangerous dichotomy, it is enough to resort to the Brazilian media. One of the main television channels produced a programme about Venezuelan ‘refugees’ in which the tone was to highlight the vulnerability of those Venezuelans living in hybrid refugee camps and, to simultaneously exalt the humanitarian role of the Brazilian army and of the Operation Acolhida (Kaysar Dadour, n.d.). Another major information portal the uol.br website reported that the presence of Venezuelans in Roraima is responsible for bringing one of the Venezuelan drug trafficking mafias to Brazilian soil, highlighting the fact that 5% of prisoners in state prisons are Venezuelans. The image that the text gives is, of course, the danger posed by Venezuelans (Ramalho, 2019).

The creation of this migrant’s image requires a militarised response, properly organised, and, in this case, the hybrid refugee camp is the practical result of the creation of both the victim/dangerous migrant and the military protector/oppressor. This is the same leitmotif we see in the Colombian peace process, where the army needs the ‘dangerous’ civilian (supposedly allied with the FARC) to justify its humanitarian intervention and the territory control—we can say that controlling space is defining subjects as dangerous. We also see the same situation in Rio de Janeiro’s favelas, where pacification operations result in the criminalisation of all ‘pacified’ communities. Residents, confused with the surrounding trafficking, play the same role as Colombian civilians and Venezuelan migrants: a game in which one defines the other. Other examples could be cited in Latin America, such as the triple frontier (Uruguay, Argentina, and Paraguay), where the same security and militarising discourses also flourished, as stated by Jusionyte (2015). What we see in Roraima is an extension of the same logic, now applied to migration management.

The mechanism for constituting these generalisations, in which populations are confused with crime or guerrilla warfare, is precisely a matter of making people invisible in their own territory, which is one of the mechanisms of constitution of the security discourse. In the case of Venezuelans in Roraima, this process was hampered by the initial dispersal of migrants through the cities. Thus, the creation of the hybrid refugee camp is precisely the territorialisation of the Venezuelan Otherness, which is spatially contained to be ‘territorialised’ and immediately stereotyped in the vulnerability/danger binomial. We have the constitution of a territoriality that allows the production of a securitisation discourse, which results in the criminalisation of migration and, at the same time, the ‘humanitarisation’ of the armed forces. However, if one figure turns into another, the obviations are not the same: there is an effective power play, and there are all sorts of capital interests on the side of militarised humanitarian forces.

The containment carried out by the military is also related to the relocation of the Venezuelans. The policy, promoted by local governments and international agencies, aims to dilute the concentration of Venezuelans from Roraima and ‘dissolve’ their presence throughout the country. Even the OCS, which constituted support networks for Venezuelans, is now relieved in the process of constituting military shelters. The tension lies precisely between the wish of local governments to get rid of the Venezuelans and the wish of the people in displacement. Research indicates that some of them want to be ‘relocated’. However, many do not want that because they are able to trade with Venezuela and operate as backpackers of basic goods (food, hygiene, medicine), and thus help and support families in Venezuela. That is, there is some trade flow in this movement, which indicates the non-fixation of the Venezuelans in Brazil, precisely because they are in a movement of comings and goings.

According to a recent UN/IOM report, 52% of the Venezuelans in Brazil are identified as ‘flowing’, with no intention of establishing themselves in Brazil (UN Brazil, 2018). It seems that this ‘flowing’ population that is not so much interested in the army-managed shelters as they clearly limit the mobility of their ‘internees’. We could say that the relocation policy is related to the armed forces as a form of containment and mitigation of the supposed ‘danger’ generated by this movement, while the willingness to move and trade of part of that population appears in the key of the illegal trade (even considering how much this market moves the local economy). The management of this population by the army is precisely a mechanism to combat cross-border trade, seen in the key of illegality.

Final Considerations

The military shelters/refugee camps in Roraima appear; therefore, even in their hybrid form, as an advance of securitisation related to population control, an advance narrated as a humanitarian action. Thus, the scenario of this experience can be considered under the same auspices that preside a series of political movements in contemporary Brazil as the increasing experimentation of the use of the armed forces as an instrument of containment of dangers represented by illegal trade (whether in a large city such as Rio or in a small capital such as Boa Vista), the military administration of population displacement (whether in the slums of Rio, in the neighbourhoods of Port-au-Prince or on the outskirts of Boa Vista); and, finally, the advance of prejudiced perceptions against foreigners in Brazil. However, it is necessary to consider the challenges in the Venezuelan migration in Brazil are against a tiny flow, compared to what is happening in South America. The Brazilian share in this Venezuelan drama is approximately 0.5% (UNHCR, 2019).

The idea of ‘humanitarian aid’ is a mechanism that makes it possible to use it as a form of segregation. For conservative politicians, the federal police and armed forces in the context of Roraima, this means the segregation and exclusion of the city. The political solution articulated by this ‘humanitarian’ ambiance, in which international organisations have their place and local civil society are gradually excluded, is simply a stiffening of migration policy. The experience of military shelter management in Roraima creates a hybrid refugee camp system: a middle ground between traditional Brazilian policy towards refugees and European policies. This hybrid field imposes a mimicry of the military order among refugees, leading to a specific military sociability, from which some Venezuelans take advantage of improving their living conditions, but which is inevitably an alternative permeated by casual and sporadic approaches. However, these sporadic approaches serve as a paradigm for a form of reception seen as typical of the Brazilian military, as opposed to the tension these refugees faced in Venezuela (as well as the bureaucracies of international institutions).

The hybrid refugee camp eventually results in an attempt to sanitise, in the form of denying the visibility of the difference expressed by the Venezuelans, in a way to manage both this erasure in relation to the city and the relocation of these people, which we can also see as a project of dissolution of the otherness resorting in the dimension and size of the Brazilian territory and population. The armed forces, in agreement with international organisations, operate even after excluding the voluntary civil society of Boa Vista itself as an agent for hiding the difference throughout Brazilian territory. The issue then becomes a shelter, relocation, and documentation. Small individual alternatives are built in this clash between visibility and erasure that the new policies offer, but always as residual in relation to the range of this collective of Venezuelans in Roraima.