Urbanizing Refuge: Interrogating Spaces of Displacement


  • This research was funded by a Social Science Research Council International Dissertation Research Fellowship, the Al-Falah Fellowship from the Centre for Middle Eastern Studies, UC Berkeley and the Spiro Kostof Fellowship from the Department of Architecture, UC Berkeley. I would like to thank Nezar AlSayyad, Steve Graham, Steve Pile, Martin Coward and three IJURR referees for their insightful comments on earlier drafts of this article.
  • Correction Note: This article was first published online on the 23rd of May 2013, under a subscription publication licence. The article has since been made OnlineOpen, and the copyright line and licence statement was therefore updated in June 2014.


Refugee spaces are emerging as quintessential geographies of the modern, yet their intimate and everyday spatialities remain under-explored. Rendered largely through geopolitical discourses, they are seen as biopolitical spaces where the sovereign can reduce the subject to bare life. In conceptualizing refugee spaces some scholars have argued that, although many camps grow and develop over time, they evolve their own unique form of urbanism that is still un-urban. This article challenges this idea of the camp as space of pure biopolitics and explores the politics of space in the refugee camp using urban debates. Using case studies from the Middle East and South Asia, it looks at how the refugee spaces developed and became informalized, and how people recovered their agency through ‘producing spaces’ both physically and politically. In doing so, it draws connections between refugee camps and other spaces of urban marginality, and suggests that refugee spaces can be seen as important sites for articulating new politics.

Refugees, and other ‘persons of concern’ to the Office of the United Nations High Commissioner for Refugees (UNHCR), now number over 30 million — the population of a sizeable country. Yet the focus on them is highly specialized within both policy and academic work, as they fall ‘outside’ the debates on ‘normal’ societies. There is a prominent silence in urban studies on the question of refugees, despite the fact that increasing numbers of displaced persons continue to flock to cities, and conditions for them are little better than those of the urban poor. In policy circles, discussions on refugees are largely confined to humanitarian relief, living conditions in camps and the role of UNHCR, the United Nations organization responsible for the welfare of the world's displaced/refugees (except Palestinian refugees, who are under the mandate of the United Nations Relief and Works Agency in the Near East — UNRWA). While there appears to be an emerging consensus that refugee spaces are increasingly becoming ‘slumlike’ or mimicking cities, the lens of critical urban studies that carefully interrogates the relationships between ‘slums’ and urbanity is tragically missing in this case. Perhaps this is because of the evolving political conditions of refuge itself in the late twentieth and early twenty-first century. This article interrogates the urbanization and informality of building in refugee camps to understand how such spatial productions are an important means by which refugees articulate their politics.

Theorizing refuge

Theorizing refuge, be it the camp or settlement or any other spatial configuration, has become a difficult task, both because their stateless condition makes such places an anomaly to the taken-for-granted state sovereignty of the twenty-first century (Malkki, 1995a) and because their spatio-temporal condition mimics that of slums around the world when perhaps it shouldn't. Unlike the ‘ideal’ cold war refugee,1 most contemporary refugees flee in large numbers from low-intensity conflicts and civil wars in countries of the global South. Usually under these circumstances, host nations negotiate with humanitarian agencies to create refugee camps that are meant to be temporary. The UN, for example, registers refugees and conducts health screenings, or calculates the approximate population through aerial photography and GPS. Registration is important for ascertaining numbers of refugees and developing a ‘cost-effective’ method of delivering aid. It also enables the UN to separate genuine refugees from bogus ones and thwarts attempts by refugees to acquire multiple registration documents. Finally, registration allows individuals and families to receive ration and health cards, which are important in obtaining regular supplies of food, water, tents and healthcare, and for asserting refugee status.2 The UN, along with various aid agencies such as the Red Cross, provides technical expertise on camp planning to refugees for the duration of their displacement. Such practices of care have evolved from the post-second world war era, and are ultimately devised to control and manage populations in an orderly fashion (Malkki, 2002). Humanitarian intervention provides a convenient ‘carrot’ for host governments to allow refugees to remain in the country, as aid flows not only to the refugees but also to the governments themselves to help them carry the burden of rehabilitation. In these camps, refugees are often compelled to languish for years, if not generations, while awaiting the right conditions to ‘return’. Prolonged exile has become the norm, compelling aid agencies to ‘rethink’ refugee spaces as ‘transitional settlements’ that require more careful negotiations and planning around issues including education and employment (Corsellis and Vitale, 2005). Hannah Arendt's (1966) theorization on the rise of totalitarianism and the condition of statelessness between the two world wars is appropriate here: the figure of the refugee is the body that marks the growing ethno-nationalism and racism of modern nation-states legitimated by international consensus. Politics, policy and legality of contemporary refugee protection continue to legitimate the violence of nation-states and the marginalization of refugees in wider geopolitics.3

Refugee camps are complex places that challenge socio-spatial imaginations of practitioners and academics alike. As spaces that fall within the remit of humanitarian protection and aid, and outside the national order of things, they are simultaneously within and outside the law. Georgio Agamben's (1998: 175) theorizations on the camp and ‘bare life’ have been useful in thinking through the ways humanitarian agencies ‘manage’ displaced populations (Elden, 2006; Hanafi and Long, 2008). Agamben argues that ‘bare life’ (zoe) is one stripped of political life (bios), rendering ‘humans as animals’ (Owens, 2009: 568). This legal abandonment is an active relational process, in that one is included through exclusion (Agamben, 1998; Pratt, 2005). The refugee camp for Agamben is the quintessential zone of indistinction, where refugees can be reduced to ‘bare life’ and be subjected to various forms of violence without legal consequences (Owens, 2009). He goes on to argue that the phenomenon of the camp has now become widespread and manifests itself in different forms within various spaces of the city (Agamben, 1998; Pratt, 2005). Some scholars have taken this further, suggesting that the city and modern society cannot be understood without the camp (Diken and Laustsen, 2005). What is particularly troubling about such Agambenian formulations of ‘bare life’ and ‘camps’ is that there is no possibility for the re-articulation of politics (Owens, 2009) and no recognition that the legal and moral terrain of the world has become increasingly complex, requiring us to step outside such rigid binaries (Ong, 2006). For example, while Agamben and others would see acts of lip-sewing by refugees in Australia's Woomera detention camp as an embrace of the reality of ‘bare life’ and the loss of politics, Owens (2009) argues that in fact such acts can form the basis of a new politics. Such acts not only expose the violence refugees are subjected to, but also give voice to others — to demand they speak for them.

Likewise, Agamben's focus on the management of life and death of refugee populations is useful in interrogating the biopolitics of camps. International humanitarian organizations intervene from beyond, taking on the role of the state while preserving its territorial integrity. Establishing solidarity with the host government, humanitarian organizations nevertheless become sovereigns acting on these spaces (Elden, 2009). However, there are issues with this particular analysis of sovereignty within camps. Stuart Elden points out that Agamben overemphasizes the ‘exceptional’ nature of the camp, as similar models have existed since colonial times. The concentration camp — which is the model on which Agamben rests his analyses — was used as an administrative tool to deal with domestic opponents and others. A crucial aspect was that it denied the right to mobility (ibid.: 57). Contemporary refugee camps, particularly those located in the global South, do not conform to such neat and bounded geographies or to such one-way relations of power (Abourahme and Hilal, 2009). Rather, the transgression between the space of the camp and the space of the host territory is messy, creating political ‘gray spaces’ (Yiftachel, 2009). The term ‘exception’ also does not capture the complex violent and non-violent power relations penetrating the intimate existence of refugees in their everyday lives (Harker, 2011). In particular, the growing trend of ‘urban refugees’ complicates this understanding of refugee camps, calling into question its categorization as spaces of ‘bare life’ and biopolitics. It also disregards the ways in which the ‘exceptional’ category of being a refugee can be used by displaced populations to advance their own goals as the case studies in this article will show. Other examples, some going back to the production of the ‘modern’ refugees, also demonstrate this. As Daniel Cohen (2006) points out, in the postwar period Jewish refugees were treated as ideal types of refugees and as such received preferential treatment. They were therefore also able to take advantage of their preferred status amongst various groups of postwar displaced persons to advance migration, political and nationalist agendas. Theorizing refugeeness through ‘bare life’ fails to account for the ways in which different groups of people struggle over the definition of life itself and ‘the ways that it is mediated, interpreted, abstracted, patented’ (Comaroff, 2007: 209).

Urban camps

The urban question of camps is particularly perplexing and can follow at least two different threads — discussing the urbanity of camps and discussing urban camps. Each generates important empirical and theoretical concerns. Refugees have often preferred to go to cities: work is more readily available, plus cities afford a degree of anonymity allowing them to escape the apparatus of humanitarian assistance and the stigma of refugeeness, and also promise easier assimilation into the host population (Malkki,1995b). The movement of refugees into urban areas further complicates the mission of humanitarian agencies attempting to ‘care’ for them. Locating refugees in slums and squatter settlements (where they often go) becomes quite difficult, as do attempts to ‘care’ for them since international standards are often much higher than those maintained in settlements of the urban poor in the developing world (Corsellis and Vitale, 2005).

‘Camps’ themselves can come to resemble slums of thousands of people, especially when they have existed for many years. Are these just forms of ‘emergency urbanism’4 used to house displaced populations, or are they new and emerging geographies that require new theorizations of urbanity? Michel Agier (2002) argued that theorizing camps is difficult because of the complexity of their urbanity. They are stuck between being humanitarian spaces and cities. However, he focuses on refugee spaces as being ‘on the margins of the world’, while remaining silent about informal settlements that are equally on the margins of society. Rural migrants, illegal immigrants living in slums and those in squatter settlements or labor camps in countries of the developing South are similarly subjected to the brutality of the state. In fact, one may argue that they have even fewer rights than refugees because they cannot avail themselves of international interventions. Liisa Malkki (2002) argues that choosing to talk about the camp as a ‘not yet city’, as Agier does, is unhelpful because it is a developmental jump that does not achieve much (if anything). While Malkki (1995a) raises important critiques of developmentalizing refugee camps, and provides critical reminders that refugee conditions around the world are qualitatively different from each other, she skirts continued debates in urban studies over how to ‘define’ a city and what the urban actually means. Cities are neither static geographies, nor are their subjects, networks or processes easily theorized as a whole (Amin and Thrift, 2002). Perhaps the objective should be to investigate the ways in which camps urbanize, and more critically question how their increasing informalization and marginalization can inform our current understanding of urban politics. This article suggests some ways of engaging with this subject by studying the refuge through its spatial development and politics. For the purposes of this article, however, I am mainly interested in interrogating the architectures of refugee spaces in cities as archives of spatial and political histories. In undertaking this, I hope to achieve two things. The first is to illustrate the ways in which the politics of refuge enable refugee camps to evolve into slums and squatter settlements through informal practices. Second, by doing so I intend to suggest that, despite the preordained assumption that camps are for non-citizens, they can be the spaces for the contestation of national citizenship and the production of new urban citizenship.

A transnational comparison

In order to undertake this study, I chose two case studies: Palestinian refugees in Beirut, Lebanon, and East Bengali refugees in Calcutta, India. In India, I conducted 45 semi-structured interviews with households or individuals who were original settlers in the East Bengali refugee colonies in Calcutta from partition in 1947. In the Palestinian camps in Beirut, Lebanon, I interviewed approximately 60 individuals or families from the 1948 exodus who had come following the establishment of the state of Israel. In Calcutta, my research was concentrated in the Jadavpur area (south Calcutta) where the bulk of the urban refugee population had settled. Of these, I chose to work in Bijoygarh, Netaji Nagar and Azadgarh as these were among some of the largest and oldest colonies. In Lebanon, my work focused on all four camps in the Beirut metropolitan area, namely Shatila, Mar Elias, Burj el Barajneh and Dbayyeh. While there is no dearth of camps around the world to investigate, these two case studies represent the largest, oldest and most contentious refugee flows of the twentieth century.

The site choices were deliberate as it is my contention that, in discussing the complexity of refugee spaces, it is useful to compare postcolonial states with one another. This allows the interrogation of different and problematic ways in which citizenship and shelter politics are articulated and intersect with each other in these countries. These I believe are different from the ways in which they are negotiated in the global North. Comparing the legal and social history of refuge between the global North and South is somewhat meaningless, because it fails to take into the account the brutality and informality of the postcolonial state in its relationship with its subaltern subjects, of which refugees are but one member. Hence, a South–South dialogue becomes useful to unsettle the normative understandings of refuge and explore the messiness of its practice. In addition, this is a transnational project that challenges the efficacy of comparing two distinct sites according to predetermined criteria. Instead, it allows each site to raise its own sets of questions that can be brought together to interrogate the conditions and politics of refuge at three different levels: the regional (Bengali, Arab), national (India, Lebanon) and urban (Calcutta, Beirut). In doing so it reveals how urban spaces play different yet equally important roles in providing refugees with rights to the nation-state (East Bengalis in Calcutta) and with rights to the nation without a corresponding state (Palestinians in Beirut).

The partition of India and Pakistan in 1947 caused the upheaval of over 10 million people, a number that remains unsurpassed. Violence accompanying partition announcements spread from the northeast to the northwest of India, forcing whole villages to uproot and relocate in what was the world's biggest mass migration, all in less than 9 months. Remarkably, the final borders were announced after the two nation-states had been declared independent (Daiya, 2008). Refugees migrated across the western and eastern borders of India (originally there was a West Pakistan and an East Pakistan, the latter subsequently becoming Bangladesh). Over 3 million came to India from the east in 1947, and others migrated in later periods of political crises and wars (frequent between the two nation-states). Many scholars argue that partition continues to this day as people continue to cross the Bangladesh–India border, motivated by the same reason as 60 years ago, namely persecution (Samaddar, 1999; Van Schendel, 2005). Mirroring the 1947 calamity in South Asia, the creation of the state of Israel a year later displaced hundreds of thousands of people in the Middle East. Although numbers are disputed, between 600,000 and 800,000 Palestinians are estimated to have been uprooted. In June 1967, another 100,000 Palestinians became refugees as a result of the Six-Day War. However, few of these refugees were registered with either UNRWA or with the Lebanese Ministry of the Interior, and most are stateless and status-less in Lebanon today (Brynen, 1990: 206; Halabi, 2004: 40). Palestinians are now the longest-standing refugee population in the world, their numbers increasing through natural demographic growth and continued episodes of violence in the Middle East rendering more of them stateless refugees. Both contexts and their concomitant production of refugees continue to provide crucial historical reminders of political turmoil in their regions.

Importantly, both examples are also exceptions to the norm of United Nations refugee populations and hence provide useful critiques of the system. In India, the 1947 refugees were never part of the UN mission. India is not a signatory to the Geneva Conventions, and hence the refugees continue to be assisted without UN intervention. The Indian state has had a particularly troubled relationship with partition refugees, vacillating between attempting to recognize and assist them on the one hand, and halting their migration and rejecting them from the nation-state on the other (Dasgupta, 2001). Palestinians were placed under the mandate of the United Nations Relief and Works Agency for Palestinian Refugees (UNRWA), established a few days before the establishment of the UNHCR on 8 December 1949, to aid Palestinians in the Near East. Its areas of operations include Lebanon, Jordan, Syria, the West Bank and the Gaza Strip. Because UNRWA was established before UNHCR, Palestinian refugees in its operations areas were excluded from the protection and assistance mandate of UNHCR. UNRWA is a strictly humanitarian agency. It has no mandate to protect Palestinian refugees. Initially, UNRWA attempted to create development projects for Palestinian refugees to help them ‘resettle’ into their host countries. However this was firmly rejected by Palestinian refugees and host states, who argued for the right of return of the refugees. Eventually, the opposition to the programs forced UNRWA to abandon them (Al-Husseini, 2010).

While the two case studies are quite different from each other, they are mutually relevant to the study of urbanization of refugee camps. The ambivalent and often contradictory approaches to ‘helping’ refugees led to the informalization of their spaces and livelihoods. Struggles over ‘rights’ to remain, property and identity had an impact upon urban politics, and the development of camps and colonies. While one turned into a slum, another turned into a squatter settlement. The examples are on the one hand used as historical studies of refugee spaces, and on the other to show the variety of different trajectories urban refugee settlements can take. Central to these divergences is the question of how the politics of identity intersect with the politics of space. Refugee camps around the world differ greatly, and the reasons for their dissimilarities are to be found in the ways that the politics of refuge intersect with the politics of resettlement/return/rights as articulated by refugee populations. In exploring this idea through these case studies, it is hoped that refugee camps can be seen as spatial forms that need to be grappled with seriously within urban literature. An interrogation of refuge allows such articulations of urbanisms to be called into question, links refugee flows to other forms of migration, and (importantly) explores how citizenship is produced through relations between the state and its many subjects.

Building refugee spaces

When India and Pakistan were partitioned in 1947, Hindus and Muslims were given the option of deciding in which country they wanted to settle. They were also promised protection should they find themselves to be minorities within their chosen countries. The role of the Indian government in drafting partition was therefore crucial in refugee relief and rehabilitation practices. As violence in the Punjab (to the west) was on a larger scale, relief and resettlement services for refugees from this area was commensurately greater. The migration from Bengal at the time was smaller in scale and less violent. The government's response towards Bengali refugees thus ranged from ambivalence to outright rejection. In some cases, transit camps were set up to ‘encourage’ people to go back (Chatterji, 2007). For the old and infirm, unattached women and children, permanent liability camps (PL camps) were created (Das, 2000). A few refugee camps, set up for very poor refugees, were scattered across various parts of the state; the largest, Coopers Camp, was located approximately 200 miles from Calcutta. Peasants and farmers who migrated across the border largely self-settled in rural areas of West Bengal or its cities. Cities were also the target for most middle-class upper-caste refugees, who dominated the initial wave of migration on the eve of partition (Pakrashi, 1971; Chatterjee, 1992).

Urban refugees, made up of formerly landowning bhadralok (genteel classes) and now destitute families, expected considerably greater help and support from the government and society in securing adequate shelter than they received. Refugees responded to the government's neglect and the prevailing housing shortage (particularly in Calcutta) either by attempting to share or rent space in houses of extended families and friends, or by squatting (Chatterjee, 1992). Organized by a few ‘leaders’, they began invading land on the fringes of the city and squatting there, creating ‘refugee colonies’. There were two types of properties claimed by and for refugees. One was the jabar dakhal (forcibly occupied) colonies that sprung up around the city, and the other was hukum dakhal (government-acquired evacuee property). The limited available space constituting the latter category was given to those refugees who could afford to pay for it (Bandhopadyay, 1970). The vast numbers of remaining refugees engaged in jabar dakhal, chalking out and parceling up plots for incoming residents, creating their own colonies, constructing their own homes and infrastructure (Ganguly, 1997). The majority of early homes were built in much the same way as those in urban squatter settlements — initially through sweat equity, with upgrades over subsequent years. Most interviewees in this research project claimed to have settled on empty land or rice fields (whose original ownership was a topic they generally avoided). Houses were made out of bansher bera (woven bamboo sheets tied together) with tiled roofs. The houses provided little protection from either the monsoons (they would often be flooded) or burglary (thieves could cut easily through the bamboo walls or dig under them to break into the houses at night). Families would upgrade or extend houses when they could afford to do so. Foundations of brick and cement were the first solid part of the house to be built. The walls of the houses were subsequently upgraded from bamboo to tin (Sanyal, 2009). While refugees battled hard for land tenure on their small plots of land (largely successfully), the fear of being uprooted remained strong. For nearly 40 years, they were not given land titles although there was de facto recognition of their right to remain. In many cases, such as Bijoygarh, the government gave tacit approval from the very beginning of the squat (Bandhopadyay, 1970).

Although the state stopped trying to evict the refugees from the colonies, it also did not aid them much in their attempts to acquire urban services. As a result, refugees often had to engage in building their own infrastructure. Male respondents in particular noted that, in addition to building their own homes and upgrading them from huts made of bamboo and tile to tin and tile, they worked together to build roads, dig ponds, set up clinics, markets and schools. The government moved slowly to regularize and formalize these colonies, many of which received 99-year leases only in the mid-1990s and finally title deeds to the homes over the last 15 years. With security of tenure, the process of upgrading proceeded rapidly in many of these areas. Houses until recently built with less durable materials were reconstructed using bricks and cement. Many titles were sold to private developers who constructed apartment blocks, thus contributing to the building boom in this area. Today, the refugee colonies are only recognizable by their names and their ward numbers.5 Otherwise, they resemble any other middle- or lower-middle-class suburb of the city, having evolved from being squatter settlements over a period of 60 years.

Among Palestinian refugee camps, the involvement of international aid agencies from the beginning of the displacement deeply influenced the ways in which spaces were constructed and fought over. In 1948, large numbers of Palestinian refugees migrated from northern Palestine into southern Lebanon. Those who could afford to bought or rented homes there. Others (many from peasant backgrounds who could not afford private property) eventually moved into refugee camps. At the outset of exile, camps for Palestinian refugees were set up by the Red Cross. These were scattered across the length and breadth of the country, from Tripoli in the north to Tyre in the south. Some of these occupied abandoned barracks, such as Wavell Camp in the Bekaa Valley. In other cases, the camps were established on land leased from the government or from churches, such as Burj al Barajneh and Mar Elias respectively. Camps were set up at different times, though UNRWA took over management of the camps and the refugees from the Red Cross in 1950.

Although initially welcoming, the Lebanese were also very wary of the overwhelming Sunni Muslim presence of Palestinian refugees in the country, and took a clear position that they were temporary guests in Lebanon. Palestinian refugees were long restricted from living in anything that suggested permanence (Sayigh, 1978; Peteet, 1991). Hence, they lived in canvas tents, with larger and smaller tents for different family sizes. The refugees themselves were disposed to live in temporary conditions, as they wanted to emphasize that their exile was not permanent and they intended to return to their homes in Palestine. For this reason, Palestinians also challenged UNRWA attempts to provide them with development projects (the works aspect of UNRWA) and resettle them. Over time, however, living in tents became increasingly difficult, as families grew and the tents proved to be inadequate shelter against environmental conditions. Refugees resorted to building with a variety of different materials, from stones to food tins that had been hammered flat. As they were banned from building solid structures, construction took place covertly under the cover of the tents. Eventually, they were able to add zinc roofs but these became unbearably hot in summer (Peteet, 1991). Those who were caught building would bribe Lebanese police (darak) and internal security officials (Maktab Thani), or collude with them in spying against their neighbors in order to preserve their rudimentary homes. Rosemary Sayigh (1978: 135) has pointed out that in these camps ‘a possibly exaggerated estimate of the monthly take of a police officer in charge of a camp was LL 5,000 … around 10 times his normal pay’. She further notes that often two officers were enough to control an entire camp (ibid.: 133–4). In urban areas, and certainly in Beirut where there were more economic opportunities, the demographic pressure on camps was significant, not only from natural population growth but also from refugees attempting to relocate from other camps. Often this was thwarted both by the government, which attempted to control the movement of Palestinians between camps (ibid.), and sometimes even by the camp directors themselves. In Shatila camp, for example, interviewees discussed the various negotiations they often had to engage in if they were to move from their previous camps to new ones using either wasta (connections) or other means to migrate. Over time, the increased population led people to take over open spaces around their tents, in essence squatting inside the camps within the limited confines of space they had been allotted.

As this article will further elaborate upon, Palestinian camps experienced a series of transformative events: the arrival of the PLO (Palestine Liberation Organization) in Lebanon in 1970; a civil war that raged for 15 years; and brutal oppression by the Lebanese state since the end of that war in 1990. Today, conditions there are among the gravest across all the UNRWA operations areas (Said, 2003). Palestinians have been held responsible for the civil war by many sections of the Lebanese population and continue to be viewed with suspicion. Long restricted from rebuilding, recent upgrading projects are underway in many camps. Overall, however, their social and physical conditions are abysmal as they suffer from poor infrastructure, overcrowding and poverty. Lebanon has the largest number of hardship cases of all the UNRWA areas in the Middle East, and this is evident from the way in which their camps have been ghettoized by the Lebanese state.

Organizing for spatial rights

The politics and processes of protecting the spaces refugees acquired give insights into the complex identity politics and ‘rights-speak’ employed by the refugees. In Calcutta, the squatter colonies were not welcomed either by citizen groups or by the state. In some cases, such as Bijoygarh in the Jadavpur area, where the land (a former US military barracks) belonged to the state, there was tacit approval from the state government for refugee-squatters to settle there (Bandhopadyay, 1970). In other cases, where the land was private, there were conflicting agendas over protecting private property rights and providing shelter for refugees. Many attempts at trying to evict squatters through force and the law were met with resistance and mass protest by refugees. Landlords often sent in thugs to beat squatters off their land. The state sent in police to remove refugees from the settlements. The state government tried to pass laws such as the 1951 Eviction Bill to legally remove refugees from their settlements. Refugees in turn became highly organized and formed umbrella organizations such as the United Central Refugee Council (UCRC), bringing together the various refugee colony leaders and working together with the Communist Party to advocate the right of refugees to migrate to and settle in India. These collaborations were often successful; oppressive laws restricting refugee status and rights to rehabilitation within India (and West Bengal) were rescinded and the Communist Party itself rode to power helped by the support of refugees (Chakrabarti, 1999).

Refugees' claims were premised on two things: first that, as Hindu Bengalis, they had the right to resettle in West Bengal and Calcutta. Historically, Bengal was seen as a single undivided area unified by language and culture. Its partition in 1947 was the second attempt at dividing its people (in 1905 British administrators had split the predominantly Hindu west of Bengal from the predominantly Muslim east of Bengal, only to rescind the decision six years later), and seen as a threat to the culture and heritage of the Hindu Bengali nation. Hindu refugees therefore saw their migration to West Bengal (part of Hindu India) as a way of protecting their way of life (Chatterjee, 1992). Secondly, the Indian government had guaranteed that those who were left facing persecution on the ‘other side’ of the border after partition would be welcomed into India if they sought refuge there. Meanwhile, Hindu refugees viewed the partition of India as a punishment that ignored the many sacrifices they had made for the protracted independence movement of India. Rather than rewarding them for their insurrection against British rule, bhadralok (genteel classes) refugees felt betrayed by the Indian state. The promise of refuge then was one way of recovering their dignity as freedom fighters and holding the state accountable to them for its independence. Refuge in this case was a right, not a privilege (ibid.). Interestingly, the practice of squatting in this instance wove two contradictory sets of claims together. On the one hand, the refugees demanded to be treated as equal citizens and not as refugees, a term they found demeaning. On the other hand, by taking over property that belonged to other citizens such as absentee landlords and Muslims, insisting that they had greater right to shelter, they demanded privileged treatment. The contradictions between equality and privilege point to the politics of refuge that can subvert the exceptional subject position into one that, rather than being reduced to ‘bare life’, can use it politically to its advantage.

The right to space within the city was the means by which larger claims to citizenship in the nation-state and the imagined Bengali nation were articulated. The right to shelter, to dwell within the city, and to regain respectability through the acquisition of land and housing, however problematically realized, became the foundation of the struggle between the state, which attempted to keep refugees out, and refugees who insisted on exercising their right to refuge. Urban citizenship, often seen as distinct from national citizenship, in fact became the conduit through which claims to belonging and identity were carved out.

In comparison to the Calcutta case, the Palestinian camps experienced different forms of organizing against restrictions on their movements or attempts to evict them altogether. Much of the response towards the Palestinian presence in Lebanon has been a product of the delicate confessionalist politics of the country. As is the case in similar situations around the world, the host state played a key role in deciding the limits of freedom the refugees could enjoy without upsetting the local population. As Julie Peteet (1991) points out, it is evident from the way the camps are spread across the country that there was a deliberate attempt by the Lebanese state to scatter the Palestinians in order to control them more easily. As a result, in addition to restricting building activities (as outlined earlier), economic activities were also severely constrained. Palestinian refugees did, however, work around such constraints by engaging in a variety of different informal economic activities, from construction work to running small businesses in the camps. When the PLO finally arrived in Lebanon in 1970, they made an agreement with the Lebanese government under the Cairo Accords to take over the security and management of the camps. Freed from the surveillance and oppression of the Lebanese state, Palestinian refugees were able to pursue building more aggressively. They formed camp committees and worked together to improve living standards within the camps. Further, they were also able to arm and equip themselves militarily, engaging in guerilla warfare with Israel and eventually becoming a destabilizing force within the state (ibid.). A powerful parallel militia and economy that in turn created a series of autonomous spaces which appeared to be beyond the control of Lebanese forces was obviously a threat to the political stability of the country. As Haddad (2003) argues, it further exacerbated divisions between Maronite Christians and Muslims. The former decided to train its own militia (known as the Phalangists) and clashes between the two ultimately led to the 15-year civil war.

At the end of that war, Palestinian camps were marked by several rounds of destruction and rebuilding. Palestinians themselves were faced with severe restrictions which included being barred from a range of skilled and semi-skilled professions and public sector employment, plunging even more of them into ‘ultra-poverty’.6 This was in addition to having the right to own or inherit property rescinded (Sayigh, 2001a). In essence, the Palestinians in Lebanon were reduced to a form of ‘bare life’ sustained mainly through UNRWA aid. The vehement opposition of the Lebanese towards a permanent resettlement or implantation of Palestinians in the country (towteen) was and is equally strongly rejected by the Palestinians, who insist on the ‘right of return’ to Palestine, but the Lebanese position on resettlement also translates into restrictive policies regarding social, economic and civil rights of the Palestinians as mentioned above (Nasrallah, 1997; Haddad, 2000; Sayigh, 2001b; Khalili, 2005). The fallout from this is of course intense demographic and economic pressure on the limited space of the camps themselves. The limits of Lebanese law, and indeed sovereignty, end at the borders of the camp, creating perhaps the only space in the state to which the Palestinians can lay claim. Here, within the confines of the camp, Palestinians can own, build, rent and inherit property, but the outcomes are tall, precariously balanced structures, narrow winding roads, and zigzagging electricity lines and water pipes. In addition to the overcrowding, years of restrictions on rebuilding or bringing construction materials into the camps led to overburdened and ill-maintained infrastructure, with remnants of war-ravaged buildings only now being upgraded. Camps are no longer solely Palestinian spaces. Today, many migrant workers from countries such as Syria, Egypt, Sri Lanka and Bangladesh rent cheap accommodation in them, often leading to conflictual situations (Peteet, 2005). Essentially the camps have evolved into ‘slum-like’ spaces, as UNRWA itself has begun to recognize.

It is important to note here the presence of several sovereigns, including at the very least the Lebanese state, UNRWA and later the PLO. Secondly, it is also evident from the building practices of refugees in the pre-PLO period that the relationship between refugees, the sovereigns and the space of the camps was much more complicated than that which can easily be reduced to ‘bare life’. Certainly, refugees were controlled in a variety of different ways, but they in turn were also able to subvert the system of control and engage in multiple forms of negotiations with sovereign representatives in order to advance their agenda. UNRWA, for example, had to modify its services to the refugees. The Lebanese state itself clearly could not establish full control over the Palestinian camps. However, Palestinian refugees did not engage in direct confrontation with the state over the provision of shelter. Rather, they practiced a ‘quiet’ form of squatting which is common in many parts of the Middle East. In Egypt, Iran and other countries, the urban poor squat in ways that seek to avoid the notice of the state for fear of demolition and eviction (AlSayyad, 1993; Bayat, 2000). Clearly, there are paralleling interventions by the state into squatter settlements and refugee camps except that, in the latter case, it is tempered through the presence of another ‘sovereign’ — the UN.

In the Palestinian case too then, space plays a critical role in the articulation of rights among refugees. While Palestinians are recognized as part of the conceptual Arab state, this has no bearing on their spatial rights. Hence, their insurgent nationalism is not aimed at gaining citizenship or nationality rights in Lebanon. Rather, it is indicative of the rights as a displaced nation of Palestinians to return to their homes. As Robert Bowker (2003) points out, the existence of UNRWA and the refugee camps is important for refugees persisting in their claim for the right to return to Palestine. In this regard, the organization and its spatial responsibility are political tools that seek to remind the international community of the continuing Palestinian exile (ibid.). The camps remain as permanent–temporary spaces, pertinent reminders that the condition of statelessness is an act of defiance against the persistent efforts of the international community to resettle Palestinians in the host states. Symbolically, the camps in Lebanon have prominently placed markers specifying their distance from Palestine, the homeland, keeping alive the hope of returning. In a sense, the camps are political claims of return rendered through bricks and mortar.

Refugee informality as a ‘new’ form of politics

These ‘urban’ practices described in the Indian or Palestinian cases are no longer the exception but rather the norm, as refugees continue to move to cities in developing countries and their camps continue to grow to the size of ‘cities’ as well. Building practices described in the Palestinian case find echoes amongst refugees around the world, compelled to engage in similar forms of survival in a variety of circumstances, as the need for more durable shelter and sustainable livelihood becomes imperative. In various camps, therefore, it is not uncommon to find makeshift shacks built using discarded ration bags, boxes, tarp and so forth, and refugees eventually finding ways of earning a living by selling excess rations to other refugee families, running video shops or tea stalls (Agier, 2002).

Property relations in refugee spaces are remarkably telling examples of how refugees and slum dwellers have come to constitute a larger informalized population. Refugees in camps in Lebanon can rent out rooms in their homes, homes that they do not actually own. In Calcutta, refugees could earmark plots of land and draft maps thereof, despite not having legal rights to such spaces. The only difference perhaps is that in the Palestinian camps there is a perverse rendition of ‘security of tenure’ that exists only because it would be difficult to evict Palestinians from their camps.7 Indeed the relationship between the urban poor and refugees is similar and yet different in that they share the same position as marginal populations within the urban environment, but while the urban poor can often legitimately gain access to employment, the refugees cannot. On the other hand, while the urban poor can be and often are abandoned wholesale by the state, refugees can ostensibly rely on humanitarian organizations to provide them with basic aid. Such complex relationships between populations of people that mirror each other's political, economic, social and spatial conditions and yet remain distinct from one another pose difficult questions for articulating a ‘refugee urbanism’.

Refuge is clearly no longer a state of temporariness, even if in practice it continues to be imagined as such. ‘Durable solutions’ for an inevitably long exile for refugees begin to take on familiar tones, as basic minimum standards for refuge compete and collude with the poverty of the urban margins (Corsellis and Vitale, 2005). As the forms and standards of the global marginals and the urban marginals begin to converge, they raise important questions and possibilities of understanding the condition of refugees through the lens of urban informality. Thus, the city provides a template (albeit a pernicious one) to understand camps instead of the other way around.

The idea that the camp is being ‘informalized’ should perhaps be obvious given that organizations such as UNRWA (alongside previously mentioned writings on camps in both the media and academia) point to how camps are becoming like ‘cities’ or are ‘slum-like’. However, what aspects of the camp render it city-like or slum-like is what is at stake. Is it simply the aesthetics of the camps themselves, abysmal spaces of poverty and deprivation, endlessly waiting for a ‘solution'? Or is it perhaps that refugees and their spaces, which are meant to be protected by international law, are increasingly resorting to survival mechanisms not unlike those employed by the urban poor, who are also equally subjected to and left outside the law itself? Building on Lois Wirth's (1938) seminal work on urbanism as a way of life, Ananya Roy and Nezar AlSayyad (2004) argue that urban informality is a new way of life, a mode of existence that is now not only pervasive in the developing world, but also the means by which urbanization itself is conducted in some parts of it (see also Roy, 2009). It is not only in the ambit of the poor, but also a means exercised by the rich to appropriate spaces of the city and rights to it. The difference perhaps is that, while the rich can gentrify the state and co-opt the process to meet their needs, the poor in many countries continue to inhabit the margins of urban life and claim rights to it surreptitiously or through insurgent means (Holston, 2008; Ghertner, 2011). In this sense, the urban poor also fall into a state of exception, with multiple sovereigns (the state, the local or extra-state authorities, gangs) determining how they inhabit the city. However, such groups have agency, as is enacted through their insurgence, despite remaining trapped by the idiom of informality (Roy, 2009). This is perhaps the challenge of the postcolonial state which complicates our understandings of how citizenship, politics and agency are formed and reformed, particularly within urban spaces of informality.

The city itself remains central to how refugee identity and spaces emerge to stake out new politics through the ‘cracks’ in the system. As centers of heterogeneous mixing, conflict and claims-making, cities are the symbolic and strategic sites through which struggles take place. They have been used to understand the emergence of unique ‘urban citizenship’, distinct from and disruptive of ‘national citizenship’ (Holston and Appadurai, 1999). Cities also become the salient sites where refugees can make their claims most visibly. It is at this intersection between the city and the camp where refugee politics and spaces can be recuperated.

Here, Oren Yiftachel's theorization of ‘gray spacing’ of urban politics is perhaps most useful. Yiftachel (2009: 250) describes ‘gray spaces’ as ‘developments, enclaves, populations and transactions positioned between the “lightness” of legality/approval/safety and the “darkness” of eviction/destruction/death. Gray spaces are neither integrated nor eliminated, forming pseudo-permanent margins of today's urban regions which exist partially outside the gaze of state authorities and city plans’. Born from new urban colonial relations, ‘gray spacing’ as a process is neither clear cut, nor does it produce finely defined agendas. Instead, this process, produced from the ‘periphery of peripheries’ exploits the ‘cracks’ in the workings of oppressive power, and through small movements upsets the prevailing urban order (ibid.). While the oppression of the state, the middle classes and developers is heavily skewed against the subaltern classes, Yiftachel (ibid.: 249) argues that a point comes when struggles no longer target citizenship, equality and integration as dominant goals, rather they are ‘intertwined with efforts to create autonomous ethnic spaces of development and identity’. ‘Gray spacing’ allows urban scholars to think of a different kind of urban politics to emerge from spaces and practices that seeks to silence and strip subjects of political voice. It allows us to see how agency is exercised through ‘descending into the ordinary’ (Das, 2007) and reclaiming livable spaces. In discussing the everyday ways in which women have coped with partition-generated social ostracization, Veena Das (ibid.: 7) argues that ‘our theoretical impulse is often to think about agency in terms of escaping the ordinary rather than a descent into it’. She points to the fact that actually life can be recovered ‘not through some grand gestures in the realm of the transcendent but through a descent into the ordinary’ (ibid.). In the analysis of words, gestures and the slippages between language and emotion, what she reveals is that this agency occurs through the practice of ordinary activities in the states of exception. Such an analysis is useful in understanding how, in intimate spaces of refuge, the acts of agency are not linked to a grand attempt at rebellion. Rather they are attempts to make spaces ‘ordinary’ through the processes of squatting and building that try to reclaim ‘normal’ life and create a ‘home’.

The informality of camps could perhaps be viewed as another geography of urban colonial relations, where subjects produce new forms of politics that are in the ‘gray space’ of legality and illegality. This other ‘periphery of peripheries’ — the refugee camp/colony — could be seen as the site from which new urban citizenships emerge that may lay claims to national citizenship (in the Indian case) or to alternate visions of life, community and rights (in the Palestinian case). As with the fragmented spaces of cities, perhaps refugee camps too can offer possibilities for elaborating livelihoods and politics that do not conform to normative frameworks (Simone, 2004).

On the spaces of displacement

As this is an article that attempts to grapple with the consequences of biopolitics, agency and space, it is important to shift from the politics of space to the politics of the body. Refugee spaces are not just potent political symbols but also strategic tools for negotiations. Neither Palestinian camps nor partition colonies exist on the margins of the world. Indeed, the productions of the camp and the colony are important ways by which political ends are met. In these cases, spatial politics allow each refugee group to access rights at three different scales: the conceptual, the national and the urban. The Bengali refugees are able to access all three through the realm of the urban. The act of invading land, setting up refugee colonies through sweat equity, and pressuring the state and national governments to formalize and regularize such colonies were efforts to ultimately gain citizenship in the Indian state and recover the Hindu Bengali nation. The Palestinian refugees on the other hand are able to access rights at the urban and the national scales. The existence of camps is significant, for they are physical reminders to the wider international community of the enduring impasse of the Israeli–Palestinian conflict and the refusal of the Palestinian people to give up their demand of the right to return to their land. However, they are yet to prove to be meaningful tools to achieve the nation-state itself. Perhaps if the Arab of Palestine had a state, the Palestinians would have a different fate and not just have access to urban citizenship elsewhere in order to survive. The two case studies have illustrated the ways in which spaces of displacement can be used as meaningful sites of politics and citizenship. These processes of claims-making may be messy and incomplete, but they mark the sites within which voice and agency can be recovered. To discuss ‘camps’ as sites of ‘bare life’ and ‘exception’ that fold into spaces of the city is not only disingenuous, but denies the possibility of articulating these new politics at the margins.


  1. 1

    Here I invoke specifically the figure of the refugee as imagined through the cold war discourse on asylum. As Guy S. Goodwin Gill and other scholars have pointed out, the wording of the 1951 Convention on the Status of Refugees (which provides protection for people seeking asylum on five enumerated grounds: race, religion, nationality, political opinion and social group) sets up a scenario in which the ‘ideal’ refugee is one fleeing political persecution from Eastern Bloc countries to the West. For a more complete discussion, see Goodwin Gill and McAdam (2007).

  2. 2

    See UNHCR (2007) for further reading on this.

  3. 3

    Article 33 of the Geneva Conventions prohibits countries from returning refugees against their will under the principle of non-refoulement. Non-refoulement is customary international law meaning that, even if countries are not signatories to the Geneva Conventions, their adherence is obligatory.

  4. 4

    Here I deploy the term ‘emergency urbanism’ as used by Jim Lewis (2008).

  5. 5

    The naming of colonies is very significant in the Calcutta case. Many colonies were named after famous freedom fighters in order to show allegiances that existed with the Indian Congress Party and the independence movement. The hope was that such patriotic displays would dissuade authorities from engaging in eviction processes.

  6. 6

    Palestinians have always faced economic restrictions in Lebanon from the beginning of exile. However, as Rosemary Sayigh (2001a) pointed out, the Lebanese parliament's passage of revisions to law 11614 (1969) concerning ownership of real estate by foreigners created ‘a new threshold of exclusion through a clause forbidding “anyone who does not have citizenship in a recognized state” from owning property’.

  7. 7

    Although Lebanon has done so, with the complete erasure of camps including Nabatieh and Tal al-Za'atar during the civil war.