1. Introduction
Texte intégral
1This study sets out to do two things. Firstly, it seeks to contribute to the burgeoning literature on ontological security in International Relations (IR). In much of the literature to date, there has been a conflation of identity and selfhood and the casualty of this has very often been the ontological. Drawing on the work of Huysmans and Bauman, among others, I seek to sketch an account of ontological security that recentres the role of ontology and selfhood, which is what separates ontological security from identity security. In doing so, I creatively reinterpret the concept of ontological security, looking in particular at the role of indeterminacy in disrupting it. Secondly, I hope to say something about Indian nationalism by making the case for Bangladesh’s importance in the project of nation-curation. I show how the uncodability of the Bangladeshi migrant and the Indian citizen presents an ontological threat to the Indian nation, portending an implosion of selfhood by undermining claims to an ontic reality for something called the Indian nation.
2To make my argument I analyse Indian newspaper discourse about Bangladesh and Bangladeshis, looking both at how the anxiety over indeterminacy is evident and at how the discourse works to abolish this indeterminacy by overcoding ambivalence. To do so, I use a combination of methods from critical discourse studies alongside a more quantitively informed corpus-based analysis. I then place this discourse within a wider set of practices, materialities and institutions also focused on the banishment of indeterminacy.
3The study continues as follows. The remainder of this chapter introduces in more detail the arguments made and provides contextual information, both about ontological security and about India, Indian nationalism, and Bangladesh. In the second chapter, I develop my theoretical argument, situating the study within and against the literature on ontological security. Chapter three details the methods and research design of my discourse analysis, which occupies chapter four. Chapter five moves beyond the discourse, looking at those institutions and practices that work alongside it and showing how they too support my argument that national indeterminacy causes ontological insecurity.
1.1 Summary of Argument and Contribution to Ontological Security Theory
4The problematic of ontological security is the negotiation of chaos and uncertainty that ‘lurks’ “on the other side of what might appear to be quite trivial aspects of day-to-day action and discourse” (Giddens, 1991: 36). This radical uncertainty emerges from the inability to provide meaning and sense to the world, and ourselves. In international relations, as in sociology, the focus has primarily been on how self-narratives bracket out chaos by providing a structure and a story that makes action and cognition possible. Narratives about ourselves produce a sense of self-identity that generates certainty in our world. The implication is that when these narratives begin to break down, we are once again faced with radical uncertainty about our being – that is to say, ontological insecurity. The problem with this literature, particularly as it has been used in international relations, is that ontological security often becomes operationalised as little more than identity security. The focus has thus been on the creation, maintenance and insecurity of ideas of ‘who we are’, rather than more explicitly ontological questions of ‘are we?’. The result is that selfhood has often been reduced to a function of identity. While this literature has proved valuable, and while I am not suggesting that identity is not crucial to notions of self, I suggest that the focus on ontology need not and should not be confined to questions of identity, and that delinking selfhood from identity allows for a greater focus on radical uncertainty, chaos and questions of ontology more generally.
5In this vein, I look at the radical uncertainty that derives from indeterminacy. Indeterminacy is that condition of not being able to make a judgement as the value of something; the inability to code a given phenomenon. This uncodability gets to the heart of radical uncertainty; without determinacy the world is amorphous, unknowable and unthinkable. Determinacy is the product of an order; it institutes a way of organising chaos that makes the world coherent and actionable. The indeterminacy of selfhood is thus what could properly be described as ontological insecurity.
6I look for the signs of ontological insecurity in India, paying particular attention to those margins where indeterminacy most visibly raises its head. I hope to show how such indeterminacy can, in Spivak’s (1993: 70) words, “show the irreducible margin in the center”. This is particularly relevant to international relations because it can help to account for numerous phenomena in world politics, phenomena that often have concrete consequences for the lives and security of a great number of people. For instance, the India-Bangladesh border fence is the longest border structure in the world, running the equivalent of Greece to Somalia, and is one of the world’s deadliest. Beyond insufficient realist and functionalist accounts, ontological insecurity offers to make this enormous disciplinary project intelligible. Indeed, it coheres quite a lot of what counts for politics more generally, as I shall attempt to illustrate. Certainly, the focus on ontological security places the border within a larger set of practices, materialities and representations that both bespeak an ontological anxiety and are aimed at its amelioration. What links them is their attempt at banishing the indeterminate that threatens the possibility of national selfhood. In making my argument, I show how the Indian nation is afflicted by a distinct ontological insecurity concerning Bangladesh and Bangladeshis.
7Krishna (1994: 509) noted many years ago how India, as “this child of partition… has cartographic anxiety inscribed in its very genetic code”. This cartographic anxiety derives from the obvious arbitrariness of India’s borders (physical and otherwise) that are often indeterminate. This indeterminacy challenges the idea of the nation’s ontic status. I explore this cartographic anxiety largely through discourse and the institutions of citizenship. I look at how indeterminacy can be shown to cause anxiety, and how this anxiety manifests itself discursively, but also materially, in practices, institutions and infrastructures, all of which seek to banish the indeterminate and create a sense of illusive-elusive ontological security. Part of this involves looking at how India narrates itself and its neighbours, and how these narrations work to make possible a sense of coherent and unitary national selfhood by strongly overcoding the arbitrary and liminal nature of nationhood. The liminal is banished by producing a set of representations that position Bangladesh(is) as devalorised, dangerous and deviant. In creating Bangladesh as a kind of postcolonial dystopia, India is able to distinguish itself as progressive, advanced and agential; a contrast space emerges that is the very possibility for sovereign selfhood. The overcoding of indeterminacy offers to restore a sense of difference that underpins ontological status. The threat of indeterminacy is therefore the threat of the annulment of such a status.
8Representations of Bangladesh therefore work to secure Indian exceptionalism by erecting a clear line between the self and the other. This line becomes central to Indian ontological security, though, because it is merely a discursive effect, it is forever unstable. The risk is that this line could be broken and the difference between the exceptional India and remedial Bangladesh fades. This constant threat undergirds Indian ontological security; the reinscription of the line becomes the condition not just of Indian identity, but of Indian ipseity. In this paper I trace this line in discourse, analysing how Bangladesh(is) are narrated in India’s two largest English-language newspapers. I argue that this discourse works analogously to the fence in asserting a sovereign presence.
1.2 Contextual Information – The Case of India and Bangladesh
9A wealth of scholarship has looked at Indian nationalism over many decades, from numerous angles. Much of the literature has focused on Pakistan, or its fraught relationships with religion, secularism, diversity, masculinity and modernity, among others (e.g. Das, 2008; Nandy, 2003; Krishna, 1999). Almost no scholarship to date has sought to interrogate Indian nationhood vis-à-vis Bangladesh, and while many studies have looked at issues of Bangladeshi migration very few have systematically brought this analysis back to a discussion on national identity or nationalism. Much of the most interesting scholarship in this area is anthropological (e.g. Ludden, 2003; van Schendel, 2004; Sur 2013; Hussain, 2015; Cons, 2016) and has sought to problematise the ‘coherence’ of national identities and to emphasise the lived reality of borderland communities, whose ontic modalities blur the boundaries between nations. What this scholarship has not done so explicitly is to link this problematisation to broader discussions of Indian nationalism, and to use them to place Bangladesh and Bangladeshis at the centre of national self-making practices. One key exception is the work of Sankaran Krishna (1994; 1999) whose work on ‘cartographic anxiety’ sought to link the permeability of the border to a wider postcolonial anxiety over nation-building and secessionism.
10Beyond this, there has also been some scholarship on Bangladeshi migrants in Indian cities, particularly by Ramachandran (2003, 2006), though as she notes (2004), this research is often difficult to carry out and, since her work in the early 2000s, limited additional scholarship has looked at Bangladeshi migrants outside border regions. Some scholarship on ontological security has taken South Asia as an empirical case. Most notably, Kinnvall’s Globalization and Religious Nationalism in India: The Search for Ontological Security (2006) has become a key text in ontological security studies, and has since be complemented by additional work on the region (e.g. Kinnvall, 2019; Kinnvall and Svensson, 2017). A few others, including Chacko’s (2014) study of ontological security in India-US relations, have also theorised ontological security from South Asia. None of this scholarship, however, has looked at India and Bangladesh.
1.2.1 General Overview of Bangladeshi and Indian History
11India gained independence in 1947 with the partition of India and Pakistan. Partition emerged as a demand of the Muslim League, led by Jinnah, who collaborated with the British during WWII, while much of the Congress Party were in gaol, in return for a separate homeland for Muslims (Pakistan means ‘Land of the Pure’). The Congress Party ruled India for three decades consecutively, first under Jawaharlal Nehru and then under his daughter Indira Gandhi. The 1970s and 1980s were a period of distinct instability in India, when a number of wars (e.g. the 1971 Bangladesh Liberation War) and insurgencies were seen as very real threats to the coherence of the Indian state and nation. Between 1947 and 2014 the Congress Party was in power for all but nine years.
12In 2014 the Hindutvavadi Bharatiya Janata Party (BJP) came to power under Narendra Modi, who won a second landslide victory in 2019. The BJP is a member of the Sangh Parivar, a ‘family’ of right-wing organisations rooted in the ideology of Hindutva. Hindutva or ‘Hindu’ nationalism sees Hinduism as the defining feature of an Indian ‘civilisation’ that has been subverted over centuries. The political project is to protect and resurrect this Hindu civilisation. A ‘Hindu’ in Hindutva ideology is anyone whose ‘fatherland’ is coterminous with their holy land (Hindus, Jains, Sikhs, Buddhists),1 though it is possible to become Hindu through assimilation, as in the case of Jews, Christians and Parsis (see Varshney, 1993). Muslims, who comprise 14% of the population, therefore figure as the key non-Hindu in Indian society from the Hindutva perspective. It is important to note that Hinduism consists of very disparate beliefs and practices, and the form of Hinduism valorised by the Sangh Parivar is a very particular form of Hinduism (high-caste, North Indian) (see Menon, 2019).
13What is now Bangladesh was carved out of India by Partition in 1947, then as the eastern wing of Pakistan. The political and economic domination of East Pakistan by West Pakistan resulted in popular unrest and the rise of a strong confederate and later secessionist movement. Following violent repression and genocide, Bangladesh gained independence in 1971 during the third Indo-Pakistan War, when Indian forces invaded East Pakistan with the support of local militias. The Awami League (AL) government of Sheikh Mujib governed until 1975, when a military coup presaged a decade and a half of dictatorship under Generals Zia and Ershad. In 1990, Bangladesh once again held elections. The Awami League, led by Mujib’s daughter Hasina, and the Bangladesh National Party (BNP), led by General Zia’s wife Khaleda Begum, came to dominate electoral politics. Since 2008, the Hasina government has consolidated its power through political repression and crackdowns on opposition parties. The 2014 election was boycotted by opposition parties and the 2018 general election saw widespread violence and vote-rigging. Bangladesh is the eighth most populous country in the world, with 165 million people, 98% of whom are Bengali. Islam is the majority religion (88%), though 10% are Hindu (Bangladesh Bureau of Statistics, 2018).
14There is animosity in Bangladesh towards the Indian government. From the 1970s the military felt as if their victory had been stolen by the Indian army and that the AL’s closeness to India was both unbecoming and a threat to sovereignty. This was one of the issues that led to Mujib’s assassination by the military (van Schendel, 2009: 182). Bangladesh’s relationship with India has been troubled ever since, though the return of the AL has led to notable cooperation between Hasina and Modi in recent years. Nonetheless, there is a great deal of anti-Indian sentiment in Bangladesh, with many seeing India as domineering. Water sharing issues, the construction of the border fence, frequent killings by border guards, support for militant groups, and anti-Bangladeshi rhetoric have been recurring issues in Bangladeshi politics. Many in Bangladesh are also deeply critical of violence against Muslims in India, and of the Modi government. Instances such as the Gujarat Riots (2002), oppression in Kashmir, and the destruction of the Babri Masjid (1992) have evoked strong emotions in Bangladesh, and have even led to reprisals on non-Muslim communities, whose religion is often conflated with pro-Indian sentiment (van Schendel, 2009: 208). Bangladesh cancelled Modi’s visit to Dhaka in 2020 amid widespread public backlash and protest.
15In India, its intervention in Bangladesh in 1971 gave the country perhaps its greatest foreign policy success and demonstrated its emergence as a major power. By creating an independent Bangladesh and confining Pakistan to its western wing, India established its pre-eminence in South Asia and made up for its humiliating defeat by the Chinese a decade earlier. Indira Gandhi used Bangladeshi independence as a way of entrenching Indian dominance in the subcontinent and Indian leaders ever since have sought to incorporate Bangladesh into a system of regional hegemony. Indeed, its intervention in 1971 is widely perceived in India as entitling it to special privileges, and as obliging Bangladesh to accept Indian leadership. As noted above, this has created resentment in Bangladesh, and has defined a politics that has sought to push back against perceived Indian domineering. In India, anti-Indian sentiment and the decline of secularism and democracy in Bangladesh are seen as acts of betrayal for the gift of independence (Van Schendel, 2009: 184).
1.2.2 Indian Nationalisms
16There are two mainstream nationalisms in India, Hindutvavadi and secular.2 While the BJP and Sangh Parivar head a nationalist movement based on Hindutva, the Congress and other parties have traditionally sought to define a secular nationalism. Within this, two strands can be identified: Nehru’s highly modernist secularism and Gandhi’s ‘Hindu-inspired’ secularism that drew on Hindu ideas and morality, and sought not so much the separation of religion and politics (as Nehru did), but rather an equidistance of politics from all religions (Nandy, 2003). Gandhi’s nationalism therefore was deeply informed by his faith, but it was not Hindutvavadi. Secularist parties have also long been pressured to adopt a ‘soft-Hindutva’ position that has ultimately undermined secularism in the country. Importantly, while there are two key forms of nationalism, most people hold elements of both, and indeed, even the BJP has found it expedient to use the language of secularism to win votes.
17Hindutvavadi nationalism emerged in the 19th century as an attempt to frame a Hindu identity that could be the basis for political mobilisation. The idea of Hinduism that Hindutva takes to be Hinduism proper is based on upper-caste North Indian practices and beliefs (sanartan dharna), many of which are in fact not practiced by many members of India’s so-called Hindu majority (Menon, 2019; Shepherd, 2019; Oddie, 2006; Udayakumar, 2005). It is based on the idea of reinstating a ‘golden age’ of Hindu civilisation (Hansen, 1996) that was corrupted by Muslim invasion and Buddhist pacifism (Savarkar, 1922; Golwalkar, 1939) and more lately by secularism. The notion of Hinduism as a unitary ‘religion’ emerged through the colonial encounter, with a Christianity that was seen to be unitary and a nationalism based on a common bond. Hindutva arose as an attempt to build a notion of a Hindu nation that could harness the mobilising power that western nationalism and Christianity had supposedly done (Menon, 2019).
1.2.3 Partition and Borders as Problematic
18The border between India and Bangladesh is one of the least naturalised and is coterminous with no pre-existing boundaries. In the border areas of West Bengal, Assam and Tripura, Bengalis are the majority on both sides, and both Hindus and Muslims are common on either side.3 Often accent and dialect are the only ways to differentiate Indians and Bangladeshis, but in the borderlands this difference is non-existent, and some migrant communities who have become legalised in India (those who came before 1971) have maintained their dialects, further complicating the codability of Indian and Bangladeshi. Outside the border regions local populations cannot differentiate between Bengali accents, and most Bangladeshi migrants outside eastern India speak Hindi.
19The border came into existence in 1947, first as the border between India and Pakistan, and after 1971, as the border between India and Bangladesh. The idea that Partition divided Muslim majority and Hindu majority areas is a myth that serves to vindicate both British expediency and post-independence nationalisms. In fact, 42% of Bengal’s non-Muslims found themselves in Pakistan in August 1947. Khulna and the Chittagong Hill Tracts, which became part of East Pakistan, had a non-Muslim majority, while Murshidabad, which became part of India, had a Muslim majority (see van Schendel, 2009: 99). The result was that 20% of East Pakistan’s citizens were non-Muslim in 1947 and that number remains 12% today (van Schendel, 2009; Bangladesh Bureau of Statistics, 2018).
20The memory of Partition in India tends to be dominated by the Punjab experience, which saw a fast and near-complete exchange of Muslims and non-Muslims, accompanied by widespread violence. The eastern partition, however, was different, with a far slower, less violent, and less total exchange of populations. The eastern partition accounts for only 23% of Partition migration. Indeed, the particular forms of territoriality that could be said to accompany and define nation-statism were far from intuitive in the east. For centuries, migration, both seasonal and permanent, had been the norm in Bengal and beyond. Political boundaries had little effect on everyday territoriality and it was widely perceived that the partition of 1947 would consequently also have little effect. Indeed, Bengal had been partitioned in 1905 but had been reunited six years later and many believed that the partition of 1947 may be equally as short-lived, a belief that continued among some until the 1970s (Ghosh, 2015: 277; van Schendel, 2009: 101; Chakravartty, 2005).
21Consequently, post-Partition migration in the east was far less sudden or total than it had been in the west, and occurred for reasons often unrelated to communal violence (van Schendel, 2009: 132). Unlike in the west, large numbers of non-Muslims chose to stay in East Pakistan after Partition, with far more staying than leaving. The migration that did occur was often temporary, with many families moving for employment or in response to political and environmental disasters, only to return after a period of time (van Schendel, 2009: 132). Indeed, prior to 1952, documents were not legally required for travel between East Pakistan and India, in stark contrast to the west where documents had been required since 1948. Before 1952 some even maintained government jobs on one side, and kept their permanent residence on the other, as in the case of railway workers (Roy, 2016). It was not until 1955, eight years after Partition, that India adopted its first citizenship laws (Roy, 2010), reinforcing once again that current forms of territoriality and belonging were not hegemonic from the ‘beginning’.
22Since then, several waves of migration have occurred, particularly following bouts of violence, and most notably following the 1971 war when as many as 10 million refugees took shelter in India (van Schendel, 2009: 164), most of whom returned. There has also been a notable growth in Bengali populations in the Indian states of Assam and Tripura. This has been blamed on migration from Bangladesh, which has certainly occurred, though scholars are divided on whether immigration or fertility is the prime cause (Mannan, 2018; Saikia et al., 2020). Certainly, Bengali immigration into these regions began at least a century ago and constitutes the latest in several centuries of eastward agrarian expansion (van Schendel, 2009; Saikia, 2019). For centuries forest and swamp lands have been pushed back along with the indigenous peoples that lived there. Just a century before independence, Sylhet, now a major city in Bangladesh (but in Assam until 1947), was described as “outside the pale of human habitation” (Ludden, 2003: 5082). This expansion boomed in the colonial period as Bengali populations spread eastward and northward into East Bengal, Assam, Tripura and Myanmar (Rohingyas) often sponsored by the colonial government. This expansion has met with resistance from indigenous communities. In Assam and Tripura in particular, violent protest movements and insurgencies targeting Bengali immigrants have occurred, and continue to this day.
23The point of this historical tangent is to make the point that the border between Bangladesh and India is fraught. The border is inherently compromised both by everyday practices that refuse or subvert it, and because the very imposition of a border in these spaces problematises the codability of nationhood and belonging. As one border-security commander lamented, “these people speak the same language, wear similar clothes and look no different. It is impossible to differentiate a Bangladeshi and an Indian” (Krishna, 1994: 515). The following vignettes reinforce this indeterminacy and uncodability:
Hoseb Ali, a resident of Nabinnagar village [India]… sat in his courtyard, lit a bidi [cigarette] and gently tossed the matchstick away. The matchstick, still smouldering, landed in Bangladesh. “Uncle, come over, I have something to tell you”, he shouted. (Banerjee, 1993)
Panitar’s division is as direct as it is arbitrary: The houses on either side of a dusty lane occupy two neighbouring countries… Dotty, like several others, crosses the border, both ways, almost every day to play a game of cricket, catch a film, visit family, shop at the markets. (Vijayan, 2016)
24While the border fence has gradually disrupted cross-border ontic modalities such as these, it has been unable to impose any degree of determinacy. The fence may discipline movement, but it has failed as of yet to discipline uncodability. The fact that it remains largely impossible to differentiate Bangladeshis (even those who do not live in border areas) from Indians, and thus Indians from Bangladeshis, is evidence of this. If the vision of an independent South Asia was one of sovereign nations, clear territorialities, and ‘western’ notions of citizenship, the project of post-independence has been to make this image a reality. The importance of achieving this goal is (perhaps ironically) directly related to the politics of decolonisation, where a strong nation and state become the vehicle to reclaiming agency and subjecthood in world politics, and for development and change domestically. To lose a sense of nationhood is thus to lose the decolonial project; this loss is precisely what is threatened by the indeterminacy of nationhood that the concern with Bangladesh and Bangladeshis is all about.
1.2.4 The Politics of Anti-Bangladeshi Sentiment – Assamese to National Politics
25The anti-immigrant movement began in the 1960s and initiated the narrative of Bangladeshi ‘infiltration’ that became part of right-wing Indian discourse in the 1980s and has since become mainstream. In Assam, this narrative began as the ‘Bongal Kheda’ (evict the Bengalis) movement and re-emerged in the Assam Agitations of the 1980s that saw several massacres of Bengali villages. Assam continues to be at the centre of anti-Bangladeshi politics, and the presence and growth of Bengalis in Assam has been used by Assamese politicians as a political tool. Assamese politicians initiated the narrative of ‘infiltration’, the demand for border fencing, for citizenship testing and for deportation. Importantly however, anti-Bangladeshi sentiment, and the anxiety that I shall argue pervades the border and those seen to cross it, cannot be reduced to the politics of Assam. The politics of Bangladesh(is) in India has gained its own life with its adoption into national politics. In the following chapters, I shall show that the issue of Bangladesh and its indeterminacy is at the heart of Indian ontological insecurity.
1.3 Against ‘Rationalism’
26In making the case of a pervasive cartographic anxiety produced by indeterminacy and the need to counter it, I need to show that the preoccupation with the border cannot be explained simply in terms of rationalist arguments from a security or economic perspective. One of the claims made by politicians is that Bangladeshi migrants are an economic burden on India, preventing genuine citizens from accessing employment opportunities and state aid, and that this accounts for the obsession with illegal migration. From an economic point of view, this argument is hard to maintain for several reasons. Firstly, as studies have shown (e.g. Ramachandran, 2004), most Bangladeshi migrants in Indian cities are in the informal sector, often working as manual labourers and domestic servants. Their labour supplements rather than competes with Indian labour (Sen, 2003), and many of the jobs they do are ones not wanted by most Indians, for instance manual scavenging (Sen, 2019). The idea that politicians should be enraged with Bangladeshis ‘stealing’ these jobs seems ironic given political parties do not regularly advocate for more employment in these ‘sectors’. Seasonal agricultural labour is also thought to have been a key draw for Bangladeshis, but at certain times of the year there is a shortage of agricultural labour and so migration is once again complementary.
27The vast majority of Bangladeshi migrants, though, have most likely settled in remote areas and practice subsistence agriculture. It is no coincidence that many migrants from Bengal (whether Bangladeshi or ‘Indian’) have made Assam and the chars their home, given the availability of land there. The chars in particular, where many Bengalis in Assam live, constitute some of the most marginal ‘land’ in India. It is difficult to see how their presence constitutes much of a burden on the economic opportunities of other ‘properly Indian’ populations in the region.
28Secondly, the argument that Bangladeshis might be taking advantage of Indian state support is also spurious. Even for most of the ‘legal’ poor in India, reliable government support is a pipe dream: more a fiction of the middle-class than a reality among those who need it (e.g. Chatterjee, 2004; Sen, 2003). One need only look at the places where Bangladeshis are supposed to ‘settle’ to reinforce the point. The chars of Assam and the urban slums of large cities represent some of the least ‘supported’ parts of India. Most of the people in these areas do not have access to even the most basic services, so the idea that they are a drain on resources is a misnomer. As Mashiur (2018) notes, “there is nothing for them to leech off of, even if they were capable of leeching in the first place—education, health care, all of that tends to be a lot better at policing the identities of the people they serve than we imagine... The charity argument is the home of the bigot and the idiot. There is no charity available”.
29It should also be noted that, despite severe data challenges, evidence suggests that Bangladeshi immigration has slowed substantially over the last couple of decades. Economic improvement in Bangladesh, access to wealthier countries such as Malaysia and the Gulf, and political hostility have made India an increasingly unpopular destination. Although one should be deeply cautious of census data in this respect, the number of Bangladeshis in India dropped significantly between the 2001 and 2011 censuses, largely because of deaths of older-generation migrants and ‘refugees’, and the overall immigration rate is down from 0.6% to 0.4% (Tumbe, 2019).
30While such statistics must be taken with caution, it is to be noted that even the largest figures of Bangladeshi immigrants in India (which have no evidence behind them) put the number at 20 million. This represents 1.5% of the Indian population. The economic threat posed by Bangladeshis is therefore hard to argue, and certainly fails to account for the zeal with which the border and its so-called transgressors are securitised. Indeed, if one was to make a purely economic argument, the cost of border fencing, infrared cameras, drone surveillance, the 2,500 NRC test centres and 40,000 civil servants, the 100 foreigner’s tribunals and the world’s largest border force most likely cost far more than any economic damage that may be caused by illegal migration.
31The other issue with economic arguments is the case of Nepal, which also sends a large number of migrants to India. As with Bangladesh, figures vary widely, from 4 million (Kharel, 2019) to 7 million (Bhattrai, 2007). Nepal and India signed a treaty in 1950 that guaranteed free movement between India and Nepal. While there are some concerns in India over Nepali migrants (e.g. Subba, 2018) and illegal activity, it is nowhere near the scale of concern over Bangladesh. This is partly because of Nepalis’ religion and the ambivalence of categories of nationhood and citizenship, and partly because, as legalised migrants, they are not forced to pass off as Indian.
32The other argument that frequently occurs in the corpus and more broadly, is the security threat posed by Bangladeshis. Migrants from Bangladesh are held accountable for crime and terrorism in India, and even insurgency, but again, the evidence is thin at best. Bangladeshi gangs are alleged to operate in numerous Indian cities, and have been linked (in the corpus) to robbery and theft. In addition, numerous articles deal with fake currency crossing from Bangladesh to India. Islamic terrorism is an issue in India and Bangladesh, and a number of high-profile terror attacks in Bangladesh have occurred in the last decade. Nonetheless, it is unclear that the preoccupation with Bangladeshi immigration can be explained by a concern with terrorism or crime. Testing the citizenship of millions of people is an extreme length to go to root out terrorism, and is also an inefficient one, as foreignness is clearly not a useful proxy for terrorism. The fencing of the border also seems an ineffective strategy given that terror networks in India operate with the support of international governments and non-state actors for whom a border fence is unlikely to be problematic. It is also hard to explain the transformation of citizenship laws with reference to terror and crime, and thus other explanations are needed.
33Bangladesh used to provide sanctuary for insurgent groups operating in India’s north-east, including the United Liberation Front of Assam (ULFA), however this was largely in retaliation for Indian support for insurgency in the Chittagong Hill Tracts (CHT) (Bhaumik, 2009). This support ceased in the 1990s, following the CHT peace accord and the arrest of ULFA’s leader Anup Chetia in Dhaka, both in 1997. The security threat posed by insurgent support is therefore not present today, and indeed is far more of an issue along the Myanmar border, with the still very active National Socialist Council of Nagaland operating from there (Bhaumik, 2009). Interestingly, there has been little push to fence this border. The insurgent threat is therefore also unable to account for the state and nation’s obsession with Bangladesh.
1.4 Structure of the Study
34The study is structured as follows. Chapter two reviews the work on ontological security to date, arguing that by reducing selfhood to a function of identity, the ontological has not been fully theorised. I then sketch a theory of ontological security that foregrounds the ontological, focusing on indeterminacy and uncodability. Chapter three details the methods and research design. Here I introduce corpus-assisted discourse analysis, address my ‘data’, and discuss what would constitute evidence of my theoretical framework. In chapter four I conduct a discourse analysis, looking at how the ontological insecurity posed by indeterminacy is both evidenced and mitigated. In chapter five, I go beyond the discourse and link it to developments in Indian politics, notably changes in citizenship law, the materiality of the border, and the practices and politics of citizenship testing.
Notes de bas de page
1 While these groups are considered “Hindu” by Hindutvavadis, it is important to note that many of them do not accept this categorisation, and Jainism, Buddhism and Sikhism were all in some way political responses to Hinduism (Brahminical rule) with which they sought to split (see, Menon, 2019).
2 There are many regional nationalisms, and forms of dalit nationalism. While both have played a strong role in Indian politics, neither have had the reach or power that secular and Hindutvavadi nationalisms have had.
3 Bengali speakers number 79 million in West Bengal, 9 million in Assam and 2.5 million in Tripura, with significant numbers elsewhere in India, largely due to labour migration.

Le texte seul est utilisable sous licence Creative Commons - Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International - CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
The Development of International Refugee Protection through the Practice of the UN Security Council
Christiane Ahlborn
2010
The SWIFT Affair
Swiss Banking Secrecy and the Fight against Terrorist Financing
Johannes Köppel
2011
The Evolving Patterns of Lebanese Politics in Post-Syria Lebanon
The Perceptions of Hizballah among Members of the Free Patriotic Movement
Fouad Ilias
2010
La justice internationale à l'épreuve du terrorisme
Défis, enjeux et perspectives concernant la Commission d'enquête internationale indépendante (UNIIIC) et le Tribunal spécial pour le Liban
Sébastien Moretti
2009
Aut Dedere, aut Judicare: The Extradite or Prosecute Clause in International Law
Claire Mitchell
2009