‘Disturbing the Work of the Office’: The Limits of Refugee Collective Action on ‘UNHCR Territory’ in Beirut

Abstract

No refugee or asylum seeker will be prevented from having direct access to UNHCR offices and staff members (UNHCR 2009, para. 47).

***

They didn’t let me pass from the fence. I said, ‘I am a refugee’.

‘Even if you are a refugee you are not allowed to pass, you can come here on Thursday, resettlement is only on Thursday.’

I said, ‘I have a right, anytime I can come. I’m a refugee already, and if I’m not a refugee you can tell me by letter [in writing], that I cannot come here anymore’ (Kareem, Beirut, 18 February 2016).

***

1. Introduction

There is a seemingly inherent tension in the policy and practice of refugee presence in and around the offices of the United Nations High Commissioner for Refugees (UNHCR), on what those interviewed for this research often spoke of as ‘UNHCR territory’. The core principle of direct accessibility to UNHCR offices and staff is continuously balanced against other competing approaches and concerns (UNHCR 2009: para 47). But what, in policy and practice, are these competing concerns? And at what point does the presence of (a) protection seeker(s)¹ become illegitimate from the perspective of UNHCR? Through a case study of collective action by Sudanese protection seekers at UNHCR’s office in Beirut, in this article I explore these pertinent questions, showing how once a presence is considered to ‘disturb the work of the Office’, a more restrictive policy involving host state security is triggered.

Lebanon is one of the most urbanized and securitized² places for refugees and asylum seekers. Here, they often live in a limbo-like situation with few formal human rights protections, including the denial of legal residency rights and work rights (Janmyr 2016, 2022a; UNHCR 2012). Refugees and others seeking asylum have often been framed as an existential threat to national security, and the securitization of migration has long been a key feature of Lebanese society, particularly in the capital Beirut (Fawaz et al. 2012; Mazzola 2023). As Sanyal (2018) has observed, this often includes the policing and containment of refugees by an assemblage of different legal authorities and local communities.

In this environment, protection seekers rely on direct access to UNHCR for numerous reasons—whether it is to apply for refugee status, report on protection problems or request help in the form of assistance and solutions. Yet, refugee access is not straightforward; not only has UNHCR been notoriously reluctant to respond to the complexities of refugee provision in urban settings (Hoffstaedter 2015), but refugees in urban areas have also often been positioned as the ‘problematic’ counterpart of encamped refugee populations (Hyndman and Giles 2011). In Beirut, like elsewhere, this is further complicated by a growing securitization of humanitarianism, involving the increased physical separation of humanitarian workers from asylum seekers and refugees including through gated architectures (Pascucci 2017).

The difficulties of balancing refugee access manifestly come to a fore in the context of collective action, that is, protests and demonstrations involving, in the words of UNHCR (2011: 25), ‘people gathering together to object to an issue, make their position or grievances known, or attract attention’. While such action can be conceived as a legitimate form of rights claiming, as this article will demonstrate, UNHCR approaches it predominantly through the prism of security. Indeed, UNHCR’s Urban Policy (UNHCR 2009: para 47) acknowledges that access ‘will, of course, have to be carefully regulated and supervised and take full account of security considerations’. For UNHCR offices worldwide, it is often challenging to ‘balance being open, accessible and responsive with the need to maintain secure premises and firm and consistent policies’ (UNHCR 2012: 32. See also Sandvik 2011). We have seen this most notably in the case of the 2005 protests in Cairo, where some 3,000 Sudanese refugees opposed UNHCR’s protection and assistance (FMRS 2006; Häkli et al. 2017; Moulin and Nyers 2007). That sit-in was violently dismantled when Egyptian security actors forcibly removed the protestors, resulting in at least 28 deaths, many injuries, and the arrest and detention of hundreds of protestors. As the Egyptian example shows, refugee access to UNHCR is not unlimited.

In this article I explore how, in Beirut, the spatial practices of UNHCR influence its relationship with Sudanese protection seekers, and how these relations intersect with broader dynamics of securitization that includes the framing of protection seekers as simultaneously at risk and a risk. As such, I argue, UNHCR’s access policies are not disconnected from host state approaches but are rather to be seen as intertwined with the state’s securitized responses to collective action, on the one hand, and refugees, on the other, practices which at once are both gendered and racialized. Recognizing that refugee protest in these settings is inherently provocative, adversarial or confrontational (Clochard 2016; Edkins and Pin-Fat 2005), I identify ‘UNHCR territory’ as an important site of contestation between non-institutional (protection seekers) and institutional actors (UNHCR and the Lebanese police).

My article draws inspiration from Aradau’s (2004) characterization of humanitarian action as sliding registers of care and control, amid which protection seekers are seen as both at risk and a risk, and as individuals who require both paternalistic management and security responses. It also employs Pascucci’s (2017) notion of a ‘spiralling pattern of securitization and contestation’, in which the protection seekers’ frustration and repeated requests for more and better assistance are perceived as a threat by UNHCR, and met with enhanced security and restrictions in access, which in turn leads to further grievances and protests. These approaches combined help us to understand under what circumstances a person or an activity become a security concern that justifies a restriction on access and presence. I argue that while UNHCR’s actions can be criticized for consisting almost exclusively of securitization, once a presence is considered to ‘disturb the work of the Office’, a more restrictive policy is triggered, including the involvement of security personnel. I interrogate what constitutes a disturbance in this context and spotlight the blocking of UNHCR entrance doors in Beirut as a pivotal example of when collective action is considered by UNHCR to be so disturbing that host state security intervention is warranted. Protection seekers are then forcibly removed from ‘UNHCR territory’.

These findings are particularly valuable in their contribution to the discussions concerning collective action on ‘UNHCR territory’ elsewhere, in the Middle East region but also globally. In unpacking the notion of ‘disturbing the Office’, I seek to contribute to the burgeoning scholarship on refugee and migrant agency (Atac et al. 2016; Nyers and Rygiel 2012; Steinhilper 2021; Stierl 2017). I also seek to further the growing body of cross-disciplinary work that explores the spatial arrangements of contemporary humanitarianism, and in particular its increasing tendency towards securitization (Duffield 2010; Hyndman 2001; Pascucci 2017; Smirl 2008, 2015).

My research is based on fieldwork in Beirut over a period of 18 months between 2014 and 2017, with follow-up interviews in 2020. During this time, I engaged in participant observation at a protest camp outside UNHCR (for a lengthier discussion of positionality, see Janmyr 2022c). The material from this fieldwork was gathered through the ethnographic methodological approach of deep ‘hanging out’, that is, the method of physical, informal, and prolonged immersion within an environment (Geertz 1998). The material consists of field notes and photos, in addition to video recordings that have been uploaded online by solidarity activists. I conducted semi-structured and unstructured interviews with 25 Sudanese protection seekers, as well as with 12 UNHCR staff at various levels of the organization. Supplementary interviews with solidarity activists and Lebanese government officials were also done. Given the contentious legal status of most of the participants in this study, consent was taken from all respondents, and I use pseudonyms throughout this article.

In sections 2 and 3 I discusse, in light of a growing securitization of both humanitarianism and the city of Beirut, the situation for Sudanese protection seekers in Lebanon and provide an overview of UNHCR’s representation in Beirut. Drawing on this, in section 4 I explore humanitarian regimes of care and control, highlighting how protection seekers are at once considered by UNHCR to be both a risk and at risk. An understanding of this is key for us to comprehend during which circumstances refugee collective action is seen as legitimate—or illegitimate—by UNHCR, and subsequently how UNHCR responds to refugee presence. Finally, in section 5 I unpack what it means to ‘disturb the work of the Office’, an activity that is often used to justify restrictions on access to ‘UNHCR territory’. Drawing on three examples—from 2012, 2015 and 2019—I identify the blocking of UNHCR entrances, seen by the protection seekers as a political act but by UNHCR as a criminal one, as the paramount example of when such a disturbance warrants the interference of host state security and the removal of protection seekers from UNHCR.

2. Sudanese protection seekers in Lebanon

Lebanon has long been a destination for persecuted Sudanese, who arrive in the country to seek asylum and access UNHCR protection, aid, and resettlement services. While the country is not a signatory to the main refugee protection instruments, the 1951 Refugee Convention and its 1967 Protocol, the Memorandum of Understanding (MoU) that it signed with UNHCR in 2003 delegates refugee status determination and assistance operations to that organization. Through this handover, UNHCR has been considered to function as a de facto ‘surrogate state’ (Kagan 2012). Importantly, however, through the MoU UNHCR also advances important elements of Lebanon’s securitized approach to migration. Rather than being negotiated with the Ministry of Foreign Affairs, the MoU was agreed with Lebanon’s General Security Office and I’ve elsewhere criticized it for mirroring Lebanon’s framing of refugees as threats to national security (Janmyr 2018).

While Lebanon is the country in the world hosting the highest number of refugees in proportion to its population size, Sudanese refugees constitute only a small percentage of all persons of concern to UNHCR Lebanon. My previous scholarship has pointed to the existence of hierarchies of protection, amid which the situation of a few thousand Sudanese refugees has been continuously overshadowed by larger humanitarian responses targeting majority refugee groups, for example from Syria. This has led to substantial differences in treatment in terms of refugee recognition, resettlement, and overall protection (Janmyr 2022a). The 2018 Vulnerability Assessment of Refugees of Other Nationalities in Lebanon (VARON) finds, for example, that Sudanese refugees are among those that are ‘systematically worse off, and at times significantly so, for virtually all indicators’ (UNHCR 2019: 2, 4).

In Lebanon, humanitarianism as the ideology of hegemonic states (Achiume 2021; Chimni 2000), with geopolitical dynamics applying a framework of preference to specific groups of refugees as a primary feature, are further intertwined with Arab nationalism, colonial histories and race. For Sudanese protection seekers, racial discrimination remains a major issue, and there have been numerous accounts of racially charged physical and verbal abuse also coming from official security actors such as the Lebanese police (Alami 2010; Malik 2010; Sidahmed 2015). Indeed, as I will discuss in a later section, when the Lebanese Internal Security Force (ISF) violently dismantled the protest outside UNHCR in 2015, security staff outright and repeatedly referred to the protection seekers as ‘animals’. Racialization is in Beirut a key feature of the spatial, social and legal dynamics of refugee governance.

For reasons such as these, Sudanese protection seekers in Beirut have sought to influence UNHCR practice through a range of measures, including by establishing representative refugee committees and by demonstrating outside the UNHCR office (Janmyr 2022a, 2022b; Janmyr and Al-Saadi 2023). They have objected to the considerable delays and backlog of refugee status determination cases and expressed concern about procedural safeguards and lack of resettlement opportunities (Kagan 2007: 18; Özkul 2023: 43). Protesters have also highlighted discriminatory treatment and racism.

In Beirut, the protests were led by relatively small groups of people—mostly men—who would spend the night camping on the fringes of UNHCR’s Beirut office. During the day, they would be joined by different categories of people, including women and children. The protest camp functioned as a stage for communicating both individual and collective contention, mainly through slogans and placards, but also through occasional hunger-strikes. The protesters were aware of the risks of partaking in such forms of collective action in Lebanon, where the government and other non-state actors regularly thwart protests through repressive and violent means (Fawaz et al. 2012; Geha 2019). Nonetheless, for many protection seekers, such action was believed to be a successful strategy, catalysing the opening of closed files as well as speeding up refugee recognition and resettlement processes (Janmyr 2022a).

To illustrate my overall argument, three instances of collective action are highlighted. First, in the summer of 2012, 21 Sudanese men—the vast majority UNHCR-recognized refugees—held a 55-day hunger strike demanding speedier resettlement processes (Al-Akhbar 2012). In frustration at not being heard, they decided to block one of the two entrances to UNHCR, whereupon Lebanese police intervened at the request of UNHCR (Daily Star 2012a). Thirteen refugees were arrested and detained without charge for over a month (Dockery 2012; Slemrod 2012b). Upon UNHCR intervention, 12 of the thirteen refugees were resettled straight from detention to third countries.

Second, in July 2015, when the number of registered Syrian refugees in Lebanon had long since surpassed one million, a small group of Sudanese protection seekers began a new sit-in outside UNHCR that would last, albeit in various forms, until the spring of 2017. They protested the daily marginalization and discrimination and called for the re-opening of files previously closed, and resettlement to third countries (Janmyr 2022a). This group also escalated their protest by blocking one of UNHCR’s two entrances, and Lebanese security forces were yet again called upon to disrupt the protest. Third, in the autumn of 2019, at a time when Lebanon was experiencing a major uprising (Karam and Majed 2022), a new sit-in was staged calling for UNHCR to re-open closed files, for speedier resettlement processes, increased protection and respect towards African refugees (Al-Saadi 2020). After two months on the street, the sit-in was escalated to block the entrance to the UNHCR building. Once again, Lebanese security forces were called to the scene and arrested several of the protesters, amid allegations of excessive force.

3. UNHCR’s representation in Beirut

Having been present in Lebanon since the early 1960s, UNHCR has had offices in many different locations around the capital city, Beirut. Today, we find the main office in Jnah, a high-end residential neighbourhood in the south-west of the Mousaitbeh district. The neighbourhood is located along the southern border of Beirut, extending eastward from the sea. Despite UNHCR’s view that it is ‘essential’ to ‘establish reception arrangements that are both accessible and convenient for refugees and which facilitate the work of the organization’s staff’ (UNHCR 2009: para 46), in many respects Jnah is an inconvenient location. For UNHCR, it is at a distance from Beirut’s many other humanitarian organizations, who generally have their headquarters across the city in Achrafieh, downtown or Hamra. For protection seekers, the Jnah location is relatively inaccessible.

Many of the Sudanese interviewees were fearful about moving around the city and of getting to and from Jnah. Most could not afford taking private or shared taxis (service) and were left with the option of walking. Beirut is a highly securitized city, with temporary or more permanent military checkpoints and roadblocks everywhere to be seen (Fawaz et al. 2012). The city’s securitization coalesces with wealth inequality and sectarianism, and, as Monroe (2011) has described, disproportionately affects those who are already disenfranchised. The ability to pass through checkpoints often rests on presentations of class, wealth, and membership of particular social groups. Those without valid residency permits are obviously at particular risk, and, in the case of Sudanese protection seekers, this risk is further both gendered and racialized.

While locating UNHCR’s offices in affluent and more isolated neighbourhoods can be seen, as pointed to by Sandvik (2011: 17), as a ‘symbol of disengagement’, in Beirut, the location of the office was influenced by the need to comply with UN field security requirements, the UN MOSS (Minimum Operating Security Standards). Areas designated as safe under MOSS are ‘mostly those situated far away from city centres, less populated and characterized by the highest concentration of wealth’ (Pascucci 2017). In important ways, this securitization arguably also shaped UNHCR’s interactions with the protection seekers. For example, the residential nature of Jnah was often used to justify UNHCR’s responses to refugee presence. During my research, UNHCR attempted to persuade refugees to disband their protest by pointing to a need for the organization to maintain friendly relations with local residents. I was often told that neighbours complained about the disorderly refugee presence around the office, and, on some occasions when local security forces became involved, UNHCR pointed to local residents as the ones ‘making that call’ (UNHCR staff, Beirut, 12 May 2016). I will return to this dynamic in a later section of this article.

Recent scholarship on the architectures of humanitarianism provide insights that are similar to the experiences of the gated compound of UNHCR Beirut (Duffield 2010; Hyndman 2001; Smirl 2008, 2015). In her study of UNHCR Cairo, for example, Pascucci (2017) describes how ‘[r]emoved from the street … the front of the building is isolated by mobile concrete barriers … all contact between UNHCR staff and asylum seekers was strictly limited and regulated’. In Beirut, as elsewhere, UNHCR’s office was required to have two physically separate entrances—one for asylum-seekers and other persons of concern, and another for staff and visitors (UNHCR 2020: 42). Access was regulated by the presence of security personnel from the Protectron private security company. At the time of my research, the sit-in was set up only a metre or two from the security post, and the protesters and Protectron staff had regular interaction. I observed both friendly relations and courteous small talk, but over time the relations grew increasingly strained. By 2017, my interviewees were complaining regularly about harassment (Abdo, Beirut, 6 February 2017; Peter, Beirut, 20 February 2017).

To limit the presence of unwanted groups of people and activities in the vicinity of the office, UNHCR also actively influenced the physical space outside of its compound (for visual representations of this space, see Janmyr and Al-Saadi 2023). Such re-structuring is permitted under MOSS with the approval of the local authorities. The extension of UNHCR’s physical authority in this way may certainly give an impression of there not only being a UNHCR office, but also a sizable ‘UNHCR territory’ beyond the walled compound. For example, the street upon which UNHCR’s Jnah office is located is closed off at one end, transforming it into a cul-de-sac with limited traffic. Most protests took place in this cul-de-sac. In 2012, in direct response to the sit-in and allegedly to prevent new sit-ins, heavy cement flowerpots were placed just outside the office (Dockery 2012). In 2016, UNHCR erected high gates blocking immediate access to the office—one staff member claimed that the gates were an attempt to clear the UNHCR reception area ‘so in part not to have protesters blocking [the UNHCR staff] to take care of people [sic]’ (UNHCR staff, Beirut, 12 May 2016). This resonates with observations elsewhere by scholars like Duffield (2010) who have argued that the fear experienced by UN staff was not simply about an external environment of generalized insecurity, but was also specifically related to the dynamics of conflict between the UN and its clients. Central to these dynamics, as I will discuss in the next section, is a perception of refugees at once at risk and a risk.

4. Humanitarian regimes of care and control

4.1 Refugees at risk and a risk

An enigmatic encounter on a warm Friday afternoon in September 2015 exposes UNHCR’s perception of the Sudanese protesters as both at risk and a risk, and as individuals who require both paternalistic management and security responses. This perception influences the way UNHCR responds to protection seekers when it comes to, on the one hand, permitting, and on the other, limiting, access and presence.

It was getting late in the day when UNHCR staff, accompanied by Protectron security, met with a group of protesters—men, women and children—outside the office in Jnah. UNHCR wanted to let the protesters know that from that point on they would not have access to the UNHCR reception area. The meeting was captured by the cell phones of the protesters, and later circulated on social media. Translated from English to Arabic by a local member of staff, a senior staff member told them:

We have taken the decision, we UNHCR, to close access to the reception area during the weekend. I understand that in the past sometimes the reception area was open for you to go to the toilet or to get water, and we are finishing that because we cannot afford to open the reception centre during the weekend. I wanted our colleague from Protection and from Security to be here because it is not their decision. We have taken that decision so you should not hold that against them.

But I wanted also to let you know that as of Monday, the reception area is for people who are having an interview appointment, a counselling appointment, who are coming to register with UNHCR. We have to prioritize those people who come through the regular channels to ask UNHCR to carry out certain activities. I’m sure you understand that we have to prioritize people who are inside waiting for an appointment over the people who are out in the street demonstrating.

I also want to tell you that I’ve asked Protectron colleagues that if people get out of the space you are not allowed to re-enter the space in front of the Office except if you have a valid reason to come and to meet with a UNHCR staff.

The UNHCR staff member continued by saying that they wanted to deliver this message ‘face to face’ so that both sides could ‘continue again this trust relationship’, but also for UNHCR to ‘be able to resume [its] normal activity’. They clarified:

For those who are registered with us, you can go through the regular channels—on Thursday for the regular counselling appointment or resettlement or at any given time—for those who are registered with us. For those who are not registered, you really have to understand that you can stay here the longest time—it will not change the position of UNHCR. So you are not helping yourself by staying here.

At this point, something caught the attention of the UNHCR staff, and they interrupted themselves and pointed beyond the group. ‘I just take this example’, the protesters were told:

You have seen the little girl running over there. OK? An accident can happen at any given time. There is a car coming. The vespa could have fallen over her and injured her. So when I was saying that I am concerned for your security and the security of your children, this is exactly what has happened now. And we cannot accept that to happen. We want also to take care of your own security.

The staff summarized the message once again:

Now it’s going to be the weekend. The reception centre will be closed. You will not get access neither to toilet nor to shower nor to potable water. You have to decide for yourself what is the best way … You know that I’ve been very clear with you—the next person who disturbs the work of the Office … my Protectron Security colleague will call the Police, and there is no discussion with you whatsoever.

Now if you want to demonstrate peacefully and not be a problem for the Office, and for the other refugees, and not use the reception centre, you can stay. But again, those who are not registered with UNHCR, we cannot do anything for you. And I can assure you that in five months’ time, in five years’ time, it will be the same answer that you will get from UNHCR. Because we only apply rules which have been decided by states. Is it well understood?

This encounter raises a number of issues worthy of closer discussion. Perhaps most prominent is the framing of refugees and other protection seekers as at once at risk and a risk. Claudia Aradau (2004) was first to observe how humanitarian action is characterized by sliding registers of care and control in which two apparently incompatible discourse regimes are entwined and even feed upon each other. She noted how the humanitarian effort to salvage becomes appropriated within a securitizing discourse. As argued elsewhere, refugees and protection seekers are both ‘casualties of care’ (Ticktin 2011) and casualties of ‘excessive governance’ (Stierl 2017). In her work on social protests against UNHCR in Ghana, Elizabeth Holzer (2015) has similarly described this distinctive form of rule that accompanies humanitarian intervention as ‘compassionate authoritarianism’.

In this sense, the Sudanese protection seekers in Beirut are integrated in a continuum of danger. On the one hand, they are constructed as ‘speechless emissaries’ (Malkki 1996), whose ability to judge is limited due to the experiences leading to their flight. As they are not capable of representing their own interests, it is ultimately their advocate—UNHCR—that decides what is best for them. This imputed incompetency places refugees in an eternal ‘sick role’ (Pupavac 2006). In conversations with UNHCR staff about the Sudanese protection seekers, I would continuously hear that they, because of their background and current position, do not necessarily know what is in their best interest. One UNHCR staff commented that ‘Many of the Sudanese are not educated, and I think that it’s difficult for them to understand how the processes work … There’s a lot of misunderstanding’ (UNHCR staff, Beirut, 13 January 2017).

The protection seekers were not only at risk in the sense described above, but their actions were seen as constituting a risk to others around them. The effect the protests had on the protesters’ children was an issue that came up repeatedly. The September encounter suggests that UNHCR did not trust the protesters to provide sufficient care for their children in the context of the protests. In conversations about the removal of the protesters in October 2015—an incident that I discuss in a later section of this article—the welfare of the children was again highlighted in a similar way. On that occasion, a senior staff member commented:

The fact that I was increasingly worried that there were a lot of children and a lot of cars passing right, left and centre, and that two or three times we were that close from an accident. And so I had told the women, ‘Ok, you are free to demonstrate. But don’t camp here with your children right, left and centre because we have vehicles coming in and out all of the time. You are a perfectly good mother but when you have four children to look out for, one can easily escape your attention and then maybe an accident.

And in some instances, I was concerned about the welfare of the children. Because some of the children had been used to put pressure on us—a mother holding a child up and pushing the child against the army. Again, I cannot understand other than that she was desperate and so on, but that’s not right for a child.

It is clear that UNHCR staff perceived the presence of children at the sit-in as a protection concern. For the refugee protesters, however, the welfare and the future of the children was often the very reason why they were protesting. Indeed, participating in protest is not something anyone here took lightly—it was a conscious action, and as observed by Walgrave et al. (2011), one that ‘entails costs and implies deliberation and choice’. As similarly argued by Holzer (2012: 273) in relation to refugee protests in Ghana, ‘protests in humanitarian settings are high-risk, arduous endeavours. People do not choose protest when they can see other more effective and safe channels of communication’. My interviewees were certainly aware of the risks associated with protesting; more than once they commented that if it hadn’t been for their children, they would not have bothered protesting outside UNHCR but rather opted for the riskier journey by sea to Europe. While few were willing to expose their children to that risk, bringing them along to ‘UNHCR territory’ was a careful calculation.

4.2 Legitimate and illegitimate collective action

The encounter also raises an important question about the legitimacy of the protesters’ claims and methods. From UNHCR’s perspective, only those with ‘valid’ reasons for coming to UNHCR would be let in, and individuals coming through ‘regular channels’ were to be prioritized over those ‘who are out in the street demonstrating’. But who has the authority to decide what constitutes a ‘valid reason’'? And on what basis is this prioritization done—other than for the purposes of perhaps disciplining protection seekers into a defined bureaucracy of care? Indeed, as Scheel and Ratfisch (2014: 936) have argued, ‘the refugee protection discourse is full of obligations and norms prescribing the appropriate behaviour of a “good refugee”’. While UNHCR essentially divided the protection seekers’ concerns into those that were legitimate and those that were not, the protesters were through their actions challenging UNHCR’s authority in deciding upon such issues.

Within the UNHCR bureaucracy of care, as observed by Moulin and Nyers (2007: 358), refugees are not to emerge as ‘political subjects in a public space of their own making’. In contrast, then, to those perceived to be patiently waiting their turn through regular channels, the protesters ‘broke the rules of international order, literally speaking out of place’ (Moulin and Nyers 2007: 358). Indeed, in 2012, when many of the hunger strikers were held in arbitrary detention for over a month, one of the released protesters told local media that as a condition of his release he had been forced by local authorities to sign a paper saying he would not protest at the UNHCR, nor would he return to the agency without an appointment (Slemrod 2012b). By enacting themselves as independently acting political subjects, then, refugee protesters ‘directly challenge the legitimacy of UNHCR’s proxy policies as well as its authority as their advocate’ (Scheel and Ratfisch 2014: 936). In response, UNHCR’s actions consisted almost exclusively of securitization.

The division between those whose concerns were perceived by UNHCR as legitimate and those whose claims were not, was most obvious when it came to UNHCR’s disengagement with individuals whose files were closed, that is, those who had had their claims to asylum rejected. UNHCR made clear on repeated occasions that it had no responsibility to care for this group of persons because they were not ‘of concern’ to the Office. While logical from the perspective of UNHCR’s mandate, the fact that a significant number of closed files were re-opened during the course of the protests—and refugee status subsequently recognized—certainly complicated this rigid position (Özkul 2023: 43). Refugee status and the opening and closing of files were often rightly perceived by protesters as fluid and not fixed. As such, the protesters argued, those with closed files should also be of concern to UNHCR.

UNHCR continuously encouraged the protesters to disband their protest and rather channel their concerns through a more formal (and from UNHCR’s perspective, more legitimate) representative committee—a committee that I have elsewhere shown was primarily seen by UNHCR as a good in and of itself rather than as an opportunity to actively involve refugees in decision-making processes (Janmyr 2022b). In fact, this representative committee was used by UNHCR to prompt the protesters to abandon their sit-in. As one committee member turned protester commented:

The Lebanese [UNHCR staff], how can I say, they refused to negotiate with the protesters, saying it’s illegal. An illegal sit-in. They requested two from the committee and Sudanese outside of the committee. They asked them not in order to solve the problem but to persuade those sleeping at the sit-in to leave from there. Supposedly they should ask the committee to negotiate with them, ‘What do we want? Why did they have the sit-in?’ They should ask us, and we negotiate with them. But they don’t want that, they say that the sit-in is prohibited, protesting is prohibited (Kamal, Beirut, 10 March 2016. Similar claims were made by Amna, Beirut, 19 January 2016 and Mohammed, Beirut, 19 January 2016).

Apart from the perceived illegitimacy of the protesters’ claims and methods, as indicated in the above quote, there was also a widespread perception among UNHCR staff that such demonstrations were illegal. Similar framing has been used by UNHCR in the context of refugee protest elsewhere; in Cairo, for example, UNHCR issued a press release following the violent crackdown by Egyptian police stating that the demonstration was illegal (FMRS 2006: 22). Commenting on the arrest of many protesters in Beirut, one senior UNHCR staff told me:

They didn’t have a right to protest outside [UNHCR]. They didn’t have a right to do that. And you know, it was a residential neighbourhood, so there were many people living there that called the Police. That complained to UNHCR. And then there’s the rumour that it was UNHCR that called the Police, and that it was UNHCR that had them arrested. It wasn’t like that. Not to my knowledge. Those working in my Department, they went outside [to the protest camp] every day after we found out that the Police would come and get them, and told them that ‘you cannot be here, we know the Police will arrive. And if they come, you will get arrested’. So that they would leave (UNHCR staff, Beirut, 13 January 2017).

In a different conversation, another senior staff said:

As you know, in Lebanon … protest—you cannot protest in public, you cannot stay there, the police can do the surveillance and [ask] ‘Why are you here?’. Because we [UNHCR] are responsible for this building, but the sidewalk is not ours to control, we cannot say ‘Okay the police cannot ask’ so the police would ask ‘What are you doing here? Why are staying here?’ So this would cause them problems, so it was clear, this situation will complicate their safety and protection (UNHCR staff, Beirut, 12 May 2016).

Not everyone in UNHCR shared this view, however, and several of the more senior staff I interviewed recognized in principle a right to peaceful demonstration. With one important caveat, however—that the protesters ‘not be a problem for the Office’. This is examined further in the following sections.

5. Disruptions and disturbances

5.1 ‘Disturbing the work of the Office’

Restrictions on access and presence are triggered when an individual or an activity is considered to ‘disturb the work of the office’. It is notable, as quoted above, how, during the September 2015 encounter, the UNHCR staff told the protesters directly:

You know that I’ve been very clear with you—the next person who disturbs the work of the Office … my Protectron Security colleague will call the Police, and there is no discussion with you whatsoever.

But what constitutes a disturbance, really? And on what basis would a disturbance warrant such a harsh reaction—a reaction that arguably places a protection seeker at increased risk of arrest and even deportation?

In its official policies, UNHCR does not readily define what would be a disturbance in the context of its operations. Its Manual on Security of Persons of Concern, where ‘protests, demonstrations and group disturbances’ are noticeably categorized under ‘threats’, nonetheless provides some indication:

If a disturbance involves acts of violence, causing immediate danger to those involved or others, or results in damage or injury to the property or person of other individuals or organizations, it becomes civil disorder or disturbance (UNHCR 2011: 25).

Ultimately, UNHCR has a monopoly over the definition of what constitutes a disturbance of such gravity that it warrants outside intervention. In this vein, Scheel and Ratfisch (2014: 936) have argued, UNHCR’s decision to respond to refugee protests with police operations ‘uncovers the violent character of the asymmetric power relation inscribed in the refugee protection discourse’.

Indeed, once labelled as a disturbance of a certain calibre, protests are often depicted by UNHCR as criminal acts warranting police intervention. When ‘self-styled leaders’ disrupted programming and dialogue with refugees in the Buduburum camp in Ghana, for example, UNHCR condemned the disturbances and labelled them a criminal act (Holzer 2012: 269). In that context, Holzer (2012) observed how the transformation of the social protest into ‘criminal acts’ subject to police action constituted a political breakdown. Recently, Ghanaian police evicted ‘nuisance causing’ Sudanese refugees from UNHCR premises for protesting delays in the issuance of their refugee documents (GH Headlines 2019).

In Beirut, disturbances warranting the intervention of Lebanese security forces included protests that endangered the lives and safety of UNHCR staff, protection seekers or others. As one UNHCR staff commented, ‘it becomes an issue for the police to address’ when the protesting is ‘a hazard or children [are] at risk or someone is hunger striking and perhaps someone needs to intervene’ (UNHCR staff, Beirut, 12 May 2016). However, as UNHCR’s initial approach to protection seekers is increasingly securitized, its threshold for categorizing actions as ‘security challenges’ is arguably fairly low. The perception of refugees at risk and as risk within sliding registers of care and control meant, moreover, that UNHCR did not appear to trust the protection seekers to provide their children with sufficient care and protection during the (however peaceful) protests.

This securitized approach is also evident in UNHCR’s evaluation of the implementation of its 2009 Urban Policy. Over a third of the offices surveyed ‘noted security challenges to UNHCR staff and premises’, including ‘angry individuals refusing to leave the premises, unruly refugees, small protests and in Sudan, a sit-in demonstration of refugees lasting over half a year’ (UNHCR 2012: 31). These examples are arguably relatively benign activities, albeit perhaps a nuisance to UNHCR staff. As I discuss below, in the context of UNHCR Beirut, a disturbance warranting the intervention of Lebanese security actors may therefore also occur when UNHCR’s operations are disrupted or hindered, for example through the blocking of UNHCR entrance doors.

5.2 Blocking UNHCR entrances

Through three vignettes from 2012, 2015 and 2019, I discuss how the blocking of UNHCR entrances by protesters was considered by UNHCR to hinder and disrupt—to disturb—its operations to such an extent that access was restricted and intervention by Lebanese security forces warranted.

5.2.1 Summer 2012

In July 2012, after 50 days of peaceful protest outside UNHCR’s office in Jnah, and frustrated over being ‘ignored’ by UNHCR, a group of Sudanese hunger strikers moved the strike to the staff entrance of UNHCR, effectively blocking one of the two entrances with ‘cardboard, blankets, a wire bed frame, and their bodies’ (Slemrod 2012a). The majority of the people involved in the sit-in were recognized refugees whose paperwork had been submitted for resettlement (Dockery 2012).

In response to the door blockade, UNHCR issued a public statement calling the move an ‘escalation’ that prohibited ‘staff from entering and exiting the building as needed’ and thus constituted a ‘serious safety concern’ (Slemrod 2012a). In an interview with local media, a UNHCR spokesperson also said that UNHCR was trying its best to ‘keep the channels of communication open’ with the protesters, ‘but this [the closure] cannot continue, it is a matter of security and protection for the refugees who approach the office, for asylum seekers and employees’ (Slemrod 2012a).

Hunger strikers interviewed by local media said that two UNHCR staff ‘came to speak to them after they blocked the entrance, asking about their demands and telling them that the closure was unacceptable’ (Slemrod 2012a). They were urged to leave the sit-in. The protesters were reportedly insulted by UNHCR’s questions about their demands, taking them to mean that the organization had ignored their previous complaints. As one of them told local media, ‘We have given them thousands of papers stipulating our demands and after 50 days they asked what our demands are. This is clowning around’ (Slemrod 2012a).

Less than one week later, on a hot August day, the Lebanese Internal Security Forces (ISF) re-opened the UNHCR entrance and ended the protest, telling the refugees that their protest was ‘a violation of the law’ (Daily Star 2012b). A press release issued by UNHCR reportedly stated that UNHCR asked police to intervene in the protest, which UNHCR no longer saw as legitimate because it impeded aid work (ibid.). It also said that UNHCR had taken steps to resolve the Sudanese refugees’ concerns but that the protest had become a safety concern:

UNHCR considers that the protest has lost its legitimacy, and its escalation has become a serious protection and safety concern to other asylum-seekers and refugees, staff, visitors and neighbors who have voiced numerous complaints to the office.

The press release also noted that ‘the protest has also started to severely impede UNHCR’s operation in the country, especially with the number of persons of concern having quadrupled since last year’. In connection with this operation, over a dozen refugees were arrested and detained in what UNHCR told reporters was an ‘unintended consequence of the need to break up the disruptive protest’ (Dockery 2012). Later, many of these were resettled abroad straight from the detention centre.

5.2.2 Summer and autumn 2015

During my fieldwork in 2015, protesters blocked UNHCR’s entrance doors more than once. As in 2012, the blockades were seen as a matter of last resort; those interviewed did not take the escalation lightly, being fully aware of the potential repercussions of the blockades. In a conversation with Mohammed, he stated that they had decided among themselves that they would not close off the doors and ‘obstruct the other refugees’. ‘We agreed on that’, he said firmly (Mohammed, Beirut, 19 January 2016). Despite this, when a group of women and children joined the protest, this is precisely what happened. Mohammed said they tried to:

talk with those who block the doors, we say ‘this is not what we do, we came here to have a sit-in and we want protection and the rights of our children, it’s not our right to close the door’(Mohammed, Beirut, 19 January 201).

In the summer of 2015, the door blockade nonetheless culminated in a violent and distressing scene with the arrival once again of the ISF, who cleared the blocked entrance and moved the protesters away from the immediate UNHCR premises. A video of the operation has been uploaded to social media (Abdo, Beirut, 25 March 2016), and the refugees interviewed for this research tell similar stories from that day, of ISF using force to move them out. Amna recalls how: ‘The security beat them, they were sitting inside and now they’re outside. We were all inside and they kicked us out’ (Amna, Beirut, 19 January 2016). Jamal similarly remembers, ‘They started beating them and forcing them out’ (Jamal, Beirut, 18 January 2016).

Many of those who blocked the doors this time were women with children. Mohammed explained the circumstances of the women’s arrests:

There were pregnant women, they attacked them in an unbelievable way, they pushed them inside the car. I was personally at the bathroom and when I came the incident was going on, I talked with the soldier and told him ‘consider her your sister’ and ‘she’s pregnant, almost six months pregnant, how can you do that? And assault her and put her by force in the car?’ He would tell her [imitating the voice of the soldier] ‘get inside the Jeep, get those animals inside’. ‘Get those animals inside the Jeep’ (Mohammed, Beirut, 19 January 2016).

Following this incident, UNHCR erected metal gates to keep the sit-in as far as possible from the UNHCR entrances. The sit-in then moved to the sidewalk across the small street, but in October, UNHCR once again considered it to be too close to its entrances, blocking access. In conversations with UNHCR staff, one told me how,

The sit-in moved from the sidewalk to in front of the gate of UNHCR. The employees couldn’t enter the building, except through the entrance that is for the refugees and asylum seekers. Therefore, our work with the other refugees has been hindered (UNHCR staff, Beirut, 12 May 2016).

Once again, the ISF was called in to forcibly move the protesters away from UNHCR entrances.

5.2.3 Winter 2019

In the autumn of 2019, after two months on the street, a new sit-in escalated to block the entrance to the UNHCR building. Once again, Lebanese security forces were called to the scene—not by UNHCR, UNHCR claims, but by ‘diplomatic police stationed at the gates calling for back-up’ (Al-Saadi 2020). Several of the protesters were arrested amid allegations of excessive force on the part of ISF. When some were still held in detention three weeks later, a civil society organization drafted an open letter to UNHCR calling for the immediate release of those detained and for accountability by UNHCR (Al-Saadi 2020). On 27 December 2019, UNHCR responded with a statement on its Facebook page, referring to the allegations as ‘misleading and inaccurate reports circulating on social media’. The statement continued:

Since October, a group of Sudanese and Ethiopian nationals started protesting outside UNHCR’s office demanding more assistance in Lebanon and third country resettlement. UNHCR has been actively engaging with the protestors and has respected their right to peaceful demonstration, while reminding them of the obligation to respect national laws and regulations. UNHCR staff have at no time used verbal or physical abuse against the protestors, nor triggered their arrest.

[…]

A video filmed on 5 December shows the guards and police officers regularly stationed outside UNHCR’s office moving the belongings of protestors after they had blocked the main entrance to the building. UNHCR is closely following up with the authorities on the situation of the protestors who remain in detention after being arrested by the police and is continuously advising refugees and asylum-seekers in Lebanon to seek legal residency to avoid putting themselves at risk of arrest and potential deportation.

While UNHCR now claimed to recognize a right to peaceful protest, the statement simultaneously reminded the protesters to respect national laws and regulations—implicit in this an obligation not to protest without the consent of the Lebanese authorities. It also acknowledged that the blocking of the main entrance was the reason for police intervention.

At the same time, however, UNHCR sought to undermine the legitimacy of the protesters and their demands by making the accusation that ‘false and misleading information’ had been circulated. The statement placed blame for the arrests on the individual protesters themselves, stating that they should have sought legal residency ‘to avoid putting themselves at risk of arrest and potential deportation’. The impossibility for most protection seekers to obtain legal residency in Lebanon is well known to UNHCR, whose own vulnerability assessment in 2018 recognized that the financial, bureaucratic and practical barriers to obtaining such residency were so grave that two thirds of refugees above the age of 15 did not have legal stay (UNHCR 2019). It was no secret that securing protection from among other things arrest, detention and deportation was precisely one of the protesters’ main demands.

5.3 A spiralling pattern of securitization and contestation

The three Beirut vignettes as described above empirically demonstrate what it means to ‘disturb the work of the Office’ in the context of UNHCR’s operations in Lebanon. The key features of this disturbance were the spatial disruption or hindrance of UNHCR’s work coupled with a spiralling pattern of securitization of contestation (Pascucci 2017). The frustration and requests of the protection seekers were either ignored or met with enhanced security measures and restrictions in access, which in turn led to an escalation of protest tactics through the disruption of UNHCR activities by, in all three instances, the blocking of the entrance doors. The disruption was a strategy taken only after serious consideration and as a matter of last resort, when the protesters felt that other tactics had been exhausted. Indeed, we know from studies elsewhere that when other means are absent, articulating political claims using disruptive means are often what is available as a tool of protest (Clochard 2016; Edkins and Pin-Fat 2005). This spiral of securitization and contestation then culminated in the forcible removal of protesters from UNHCR ‘territory’.

Central in this spiralling pattern was a perception by UNHCR that the blocking of its entrances was a disruptive, even criminal act, following a largely illegitimate protest. For the protesters, however, it was clearly a political act. Their motivations stemmed from a desire to improve the protection and assistance of the Sudanese protection seekers—indeed, asylum seekers and refugees in Lebanon could turn to no other agency but UNHCR. As the months passed without recourse, the door blockade had the aim of pressuring UNHCR into improving their conditions. UNHCR, however, refused to acknowledge the political nature of the acts, approaching them as disturbing ‘noise’ rather than legitimate ‘voice’ (Moulin and Nyers 2007; Nicholls 2013). To treat them as unambiguously criminal acts rather than to engage with the debate that underlay the protest and the door blockade, was thus also a distinct policy choice by UNHCR.

6. Conclusions

In this article I have the tension in the policy and practice of refugee presence in and around the offices of UNHCR through a case study of Sudanese protection seekers in Beirut. I show how the spatial practices of UNHCR influence its relationship with protection seekers, and how these relations intersect with broader dynamics of securitization, including a sliding register of care and control (Aradou 2004). Protection seekers are framed as simultaneously at risk and a risk, and this significantly influences UNHCR’s access and presence policies. Importantly, these refugee-UNHCR interactions demonstrate how the humanitarian effort to salvage becomes appropriated within a securitized approach that largely mirrors the securitization that defines Beirut as a city. My article highlights how the difficulties of balancing refugee access manifestly come to a fore in the context of collective action, including protests and demonstrations.

I find that restrictions on access and presence are most notably triggered when an individual or an activity is considered to ‘disturb the work of the Office’. I discuss ‘UNHCR territory’ as a contested space of control—it is, as observed by Moulin and Nyers (2007: 359) an ambiguous site where ‘power relations are enacted, taken, and retaken by various actors’. Most notably, in Beirut, it is a space where UNHCR seeks to retain its monopoly over the definitions of which type of presence is ‘legitimate’. Drawing on three examples—from 2012, 2015 and 2019—I identify the blocking of UNHCR entrances, seen by the protection seekers as a political act but by UNHCR as a criminal one, as the paramount example of when such a disturbance is defined as illegitimate and warrants the interference of host state security and the removal of protection seekers from UNHCR. The vignettes empirically identify the spatial disruption or hindrance of UNHCR’s work coupled with a spiralling pattern of securitization of contestation (Pascucci 2017) as key features of this process.

By unpacking how the spatial practices of UNHCR are an important element in explaining refugee protest, I have showed how provocative actions that disturb and disrupt UNHCR’s work may often be the only way for protection seekers to voice claims for rights and recognition. As a group, these protection seekers were structurally disempowered and highly vulnerable, and were not expected to be politically vocal during their exile. But, as Nyers and Rygiel (2012: 8) have observed, spaces of control are also sites of political action where ‘mobilisations occur, subjectivities are formed, and contestations of the regimes governing mobility are enacted’. My article exemplifies how, through radical action, protection seekers are able to leave their attributed place at the margins of society and voice claims for their rights. As such, the securitized approach by humanitarian organizations and state actors does not go uncontested.

Conflict of Interest

None declared.

Funding

This work was supported by the Research Council of Norway under Grant number 286745.

Footnotes

References

Author notes