Islam, Politics and Violence in Eastern Sri Lanka

Abstract

This article bridges Sri Lankan studies and the academic debate on the relation between contemporary Islam and politics. It constitutes a case study of the Muslim community in Akkaraipattu on Sri Lanka's war-ridden east coast. Over two decades of ethnically colored conflict have made Muslim identity of paramount importance, but the meanings attached to that identity vary substantively. Politicians, mosque leaders, Sufis and Tablighis define the ethnic, religious and political dimensions of “Muslimness” differently and this leads to intra-Muslim contradictions. The case study thus helps resolve the puzzle of Sri Lankan Muslims: they are surrounded by hostility, but they continue to be internally divided. Akkaraipattu's Muslims jockey between principled politics, pragmatic politics and anti-politics, because they have to navigate different trajectories. This article thus corroborates recent studies on Islam elsewhere that argue for contextualized and nuanced approaches to the variegated interface between Islam and politics.

Introduction

When Muslims pray on a ferry, an Indonesian friend once told me, they are confronted with a challenge to their everyday routine. As the boat maneuvers across the ocean, its orientation in relation to Mecca, and thus the direction of prayer, changes. During challenging routes and stormy weather this may result in frequent changes in the mise en scène (and frequent runs to the railing or the bathroom for those with a weaker stomach). This metaphor illustrates that religious practice and affiliations do not remain unaffected when navigating through new terrain. Changes in context generate questions of orientation, identity and boundaries.

The interaction between religion and politics features saliently in the academic debate on contemporary Islam. Globalization, Islamic revival and reform in conjunction with modern politics have made for a tantalizing mix of processes (see Soares and Osella 2009 for an overview of the discussion). While Islamists revived the concept of a caliphate and liberals charged that democracy is inherently secular, academic research has sought more nuanced ways of scrutinizing the nexus between religion and politics. It has highlighted the manifold ways in which religious symbols and ideas have penetrated politics. Similarly, political contestation and patronage have blended with religious identity, religious movements and religious governance. Various authors have pointed towards the socio-economic processes of modernization: how education, rationalization, urbanization, and the emergence of a middle class produced Islamic revival movements that challenged and overthrew the “traditional” religious elites (e.g. Bayat 2007; Eickelman and Piscatori 1996; Hasan 2007; Hefner 2000). The interface between Islam and democratic contestation thus came forward as multi-faceted in various case studies (Bayat 2007; Hefner 2000; Sidel 2007; Tepe 2008). It is in this tradition of exploring the paradoxes and complexity of everyday religious and political practice that this article is written.

Sri Lanka's Muslims – a group of about 1.6 million people, constituting 8.7 per cent of the national population (Census Department 2008¹) – have barely featured in these wider debates on Islam. Apart from its historic connection to Islam around the Indian Ocean (McGilvray 2008), this group is of particular interest because of Sri Lanka's recent history of violent conflict, which has impacted on the connections between religion and identity politics. Competing ethno-nationalism between Sri Lanka's two largest population groups – Sinhalese and Tamils – put pressure on societal boundaries and identities (Goodhand and Klem 2005). From the 1980s onwards, the Muslims in the north and east of the country found themselves on the frontline between Tamil separatists and government counter-insurgency troops. The case of the Sri Lankan Muslims thus connects contemporary Islam to a context of war and ethnic tensions. That brings us to the questions this article aims to address: how did the war interact with Muslim identities? And how, in turn, has this encounter affected the interaction between Islam and politics? Surprisingly, the Sri Lankan literature on these questions is very thin. There are relevant accounts of Muslim history, of ethnic contestation, and of contemporary politics as well as some ethnographic work on Islam, but very little that draws connections between ethnic, religious and political realms, while it is those connections that explain many of the tensions and paradoxes we witness among Sri Lankan Muslims. This article will explore these inter-relations using a detailed case study of Akkaraipattu, a town on Sri Lanka's east coast. This coastal strip is the place where Muslims are demographically most significant in Sri Lanka. Akkaraipattu is one of various Muslim pockets, but it is of particular relevance because the town is one of the few places where Muslims and Tamils live in similar numbers. The availability of good anthropological and historical work on Akkaraipattu (McGilvray 1982; 2008) was another reason to select this town.

Faced with common threats and enemies, one would expect Muslims to stand together in a place like Akkaraipattu. A bolstered ethnic and religious identity would converge with a more militant political outlook, and even jihad inspired violence could easily be imagined. But this proves to be a mistake. It becomes clear from my findings as well as existing literature (Lewer and Ismail 2011; McGilvray and Raheem 2007) that decades of ethno-nationalism and armed conflict have not produced homogeneity or unity among Sri Lanka's Muslims in the war zone. On the contrary, I argue the war has affected Muslim identity in paradoxical ways and divergent interpretations of that identity have resulted in new intra-Muslim fault lines and contradictory political orientations. I contend that Akkaraipattu's Muslims have become more politically engaged, but – both for religious and practical reasons – they have also turned their back on politics. The result is an everyday jockeying between political and anti-political behavior, an everyday survival strategy that navigates between multiple boundaries and discourses. Apparently commonsensical contradictions – between religious fundamentalism and ethno-nationalism, between personal piety and collective politics, between modernity and tradition – exist, but they explain little as people blend and circumvent them.

The empirical material presented in this article is based on a sequence of fieldwork visits to eastern Sri Lanka in 2007 and 2008, and more generally on visits to the eastern region, more or less annually, from 2000 till the present. I stayed in Akkaraipattu for several weeks at a time to collect data, make observations and interview (in English or with Tamil or Sinhala translation) a wide range of people from different ethnic, religious, class, and political backgrounds. During the fieldwork in 2007 and 2008, 122 interviews were held² and it is from these encounters that I draw the core empirical material. The analysis emerged through a sequence field visits, group discussions with colleagues and academics from the region and joint seminars with a broader research team.³

Some of the interviews had a highly performative character, because the informants saw it as a pious act to educate me.⁴ In other cases I had difficulty getting access as an outsider, for example with the leaders from Tawhid Jamaat (the smallest Islamic reform movement in Akkaraipattu) who refused to meet me⁵, or in capturing the view of female Muslims. In most cases, however, people were very willing to talk to me openly and took pleasure in telling me about their religious views and experiences. This was in part dependent on the time period. The study period was a volatile time of change, due to the turbulent political and military events. The everyday ebb and flow of tensions and incidents affected the space to talk. There were no military clashes in Akkaraipattu itself, but battles in the region as well as assassinations, round-ups and other forms of violence obviously affected people's willingness to speak and the way they emphasized or de-emphasized sensitive issues.

Islamic Revival and Politics

The primary aim of this article is to help fill the voids in our understanding of religious and political practice among the Muslim community in contemporary Sri Lanka. Before I delve into that context, it is necessary to conceptually scrutinize religion and politics.

In general …

Studies of religion are impeded by a definitional problem – its institutional, behavioral, spiritual and other facets are not easy to reconcile. In this article I de-emphasize the theological angle. I do not treat religion as an all-encompassing system of truth or an ensemble of personal or collective beliefs. Instead, this article takes a sociological focus and analyses religion through its social and institutional manifestations, though for many of my informants such a worldly interpretation would amount to heresy, as it transfers agency from God to people and reduces the divine, the magical and the spiritual to mere constructs that shape human behavior. This sociological perspective draws from Bourdieu's work and as pointed out by Sidel (in a different context), it treats religion as a “field structured by its own institutions, authority relations, instilled dispositions (habitus), means of production and accumulation, and representation of symbolic or spiritual capital […].” (Sidel 2007, xi)

Though inherently worldly, the political is no less elusive conceptually; its red-taped boundaries tend to shift depending on time, place and interpretation. Labeling something as “political” or “apolitical” is itself a political act that delineates a normative space and apportions (il)legitimacy. Jonathan Spencer's (2007) book on politics in South Asia underlines the conceptual problems of defining politics. He takes issue with the separation of formal and everyday politics and highlights the paradoxes in people's engagement with it: they tend to see politics as “dirty” and “disturbing”, but are meanwhile attracted to the spectacle of it. They have high expectations of it; they work hard to be part of it and they dress up for electoral rallies. Spencer follows the path forged by Carl Schmitt and Chantal Mouffe (without of course adopting the ideological outlook of either author) in arguing that we miss the point when we think of politics as an arena of deliberation and consultation, a privileged space of rational moderation of conflicting interests. Instead, politics is inherently about defining friends and foes. “The political” in Mouffe's terms is inevitably antagonistic and its currency consists of passions and group identities, rather than just interests (Mouffe 2005). Conflict and politics are by no means oppositional categories and politics does not put violence beyond use. Rather violence is “the heightened and intensified continuation of normal politics” (Spencer 2007, 120).

This view takes politics beyond a place like parliament and into the arena of communities and the everyday practice of rivaling identity groups and political entrepreneurs. The difficulty of separating religion from politics becomes obvious. Religion is dynamic and heterogeneous and does not escape antagonism. It can be seen as a “discursive tradition” that produces historically contingent categorizations of doctrine and practice (Asad 1986), framed and crafted by struggles between groups who hold social power (Bayat 2007). Divergent religious interpretations thus upset pre-existing allegiances and orders and in some cases produce severe turmoil. The notion of antagonism between identity groups thus seems well positioned to explore the connections between religion and politics. Indeed, the case of Akkaraipattu foregrounds the continuous group tensions and the juggling and negotiation of discursive boundaries and collective identities, and it is these processes that make the interaction between religion and politics legible.

The fact that people maneuver between positions and manipulate and reproduce boundaries reminds us that antagonism and group identities are not static structures, but subject to agency and shaped by people's everyday behavior. People navigate the religious field in everyday life. More in line with the sociological perspective taken here, Soarez and Osella advocate analyses of the ways in which Muslims “produce themselves as modern religious subjects”, a “self-fashioning” on the basis of a wide range of influences, uncertainties and sources of inspiration (2009, 11). Rather than reducing the religious and the political to each other, or alternatively, declare them incompatible, such an approach investigates how people operate both in the religious and the political field. Similar views can be found in the work of Bayat (2007), whose analysis of Muslim engagement with politics in Egypt and Iran concludes that it is through “the politics of presence” that people determine a society's religious and political outlook. He highlights the blending of the public and the private in the way Muslims forge boundaries, adjust practices and engage with politics.

… and in Sri Lanka

Studying the nexus between religion and politics thus requires qualitative and fine-grained empirical analysis and it is to that approach that this article tries to make a contribution. Rather than exploring the religio-political nexus by pinning down the divide between the political and the spiritual, between the secular and the religious, it focuses on the way people continuously define and redefine boundaries and identities. To the extent that such analyses exist in relation to Sri Lanka, they have been applied to Buddhism (e.g. Gombrich and Obeyesekere 1988). In his eloquent book, “The colors of the robe”, Ananda Abeysekera proposes “an alternative understanding of religion and politics,” (2002, 80) and ventures to “explore the Sinhala native configurations of narratives about Buddhism and politics, locating arguments and counter-arguments about them.” (2002, 80) The analysis of Sri Lankan Muslims, however, is much less developed.

In fact, it was only very recently that Muslim society emerged as a significant issue area in Sri Lanka. There was some interesting historical (Dewaraja 1994; Mohan 1987) and ethnographic work (McGilvray 1982) pre-dating the violent conflict between the Sri Lankan government and the Liberation Tigers of Tamil Eelam (LTTE) that escalated in the 1980s. A limited number of studies in the 1990s subsequently described the way the conflict transformed Muslim discourses and political practice (Wagner 1990; O'Sullivan 1999; Ismail 1995; Knoerzer 1998). A new wave of studies appeared with the advent of the peace process in 2002. Through the peace talks, political leaders and analysts were reminded of the fact that Sri Lanka's third ethnic group is deeply entangled – politically, geographically and otherwise – in what was sometimes mistakenly seen as merely a Sinhala-Tamil conflict. The December 2004 tsunami, which affected the Muslim community particularly badly, further boosted the attention for this community (Hasbullah and Korf 2009; Lewer and Ismail 2011). In the ensuing resumption of war, the Sri Lankan government pushed the LTTE out of the Eastern Province in 2006 and declared a countrywide victory in May 2009. Throughout these years, the Muslims played a pivotal role, in terms of political contestation and inter-ethnic relations, particularly in the east of the country, which is ethnically and religiously the most diverse.

Recent publications focus on political dynamics and localized ethnic antagonism (Goodhand and Klem 2005; ICG 2007 and 2008; Ismail 1995; Lewer and Ismail 2011; McGilvray and Raheem 2007; Salman 2008; Uyangoda 2007). Despite occasional activist and political overtones, some of the most insightful and useful work on these issues was produced by Sri Lankan Muslim scholars (Ali 2001; Ameerdeen 2006; Ismail, Abdullah and Fazil 2005; Nuhman 2002; Zackariya and Shanmugaratnam 1997). Coming from a different angle, a much smaller number of authors have written about important social and religious transformations in the Muslim community, such as new religious practices in relation to ethnic and gender boundaries (Haniffa 2008) or changes in the application of the traditional kudi (matriclan) system (McGilvray 2008; Ruwanpura 2006).

Unlike in most other parts of the world, the Muslims of Sri Lanka adamantly define themselves not just as a religious, but also as an ethnic group. Scholars, however, tend to take a constructivist perspective on ethnicity. In contrast to local perceptions and everyday usage in Sri Lanka, they see it as a cognitively and socially produced category. In his provocative contribution “ethnicity without groups”, Rogers Brubaker argues that ethnic groups “are not things in the world, but perspectives on the world.” (2004, 44) Such a perspective, tends to encompass “a myth of common ancestry, shared memories, and cultural elements; a link with a historic territory or homeland; and a measure of solidarity” (Smith 1993, 28). In the case of Sri Lanka's Muslims, this list creates as many questions as it answers. Indeed, the Muslim ethnic label is a somewhat empty category, a product of the struggle against its denial, against the counter image that Muslims are none other than Islamic Tamils (for a discussion see Ismail 1995; McGilvray and Raheem 2007).

The literature on Sri Lankan Muslims tends to analyze “ethnic”, “religious”, and “political” issues separately. This is a major drawback, because it is evident that these categories are strongly inter-related. In fact, the interaction between these three realms plays a leading role in Sri Lanka's present predicament. Some of the writings produced by the Muslim polity – documents produced by political and civil society leaders (e.g. Mohideen 2006) and more scholarly analyses (e.g. Ali 2001) – conveniently marry the religious and the ethnic sphere to substantiate historic claims, minority rights and political aspirations. Such sources thus fail to scrutinize the tension between the religious and the ethnic sphere. Other scholars in turn adopted this rhetoric rather uncritically. O'Sullivan (1999) for example reviews the rise of the Sri Lanka Muslim Congress (SLMC) in the 1990s and connects the ethno-political outlook of the party to its Islamic discourse. She cites SLMC sources on “the Islamic notion that political and religious leadership should not be separated” (1999, 258) and attributes a call for jihad (335) to them, but in my view she misreads party political rhetoric as the ascendancy of Islamist politics. When we move beyond the written documents and political figureheads that she consulted, the presumed convergence of ethnic sentiments, militant politics and religious fundamentalism proves over-simplistic.

O'Sullivan's account does not stand alone. More widely, the juxtaposition of the terms Islam and armed conflict tends to invoke political and analytical knee-jerk reactions in a post-9/11 world. Implicitly or explicitly, policy makers and analysts have searched for emergent jihadi tendencies in eastern Sri Lanka (see for example ICG 2007). Rumors of frustrated Islamic youth, wahabi influences, uncanny armed groups, and inter-ethnic tensions have easily blended into compelling stories, but they are typically void of robust empirical evidence. There is thus a need for studies that scrutinize these interactions. This article does so with a place-based study of Akkaraipattu, one among numerous towns along the east coast where the context of war, ethnic tensions, religious revivalism and political contestation collide (and converge) with each other.

The View from Akkaraipattu

First time visitors to Akkaraipattu (see map 1–4) will notice the clock tower and the common features of a Sri Lankan rural town: a bustling market, bus stand, shops and local restaurants lined up along the main road, which occasionally gets congested with bullock carts, tractors, a flurry of motorbikes, and school-going children. Travelers passing the town may note the striking difference between the Islamic features of the northern half of the town and the Hindu kovils (temples) and Christian churches in the southern half (see map 4).

Figure 1. Maps of Sri Lanka, the East, Greater Akkaraipattu, Akkaraipattu Town.

The geography of ethnic enclaves that characterizes the east coast manifests itself in the heart of the town: a fairly clear-cut division can be made between the Muslim and the Tamil side. The former is a densely populated (37.000 people) grid of walled alleys and lanes, accentuated by numerous shops and an occasional villa. The southern half houses some 23.000 people and encompasses the cradle of the town – the Pilliyar temple – and the old post office and hospital, schools and government buildings. This side of town has a different, quieter feel to it. Its appearance is greener and more spacious; the numerous traditional Tamil houses – some of which are fairly run-down – are often separated by barbwire or palm leaf fences, which are cheaper than walls. Both sides of town have their own ethnic “hinterland”: neighboring northern settlements like Addelaichennai and Nintavur are Muslim, while one finds Tamil settlements (Tampaddai, Tambiluvil and Thirukovil) to the south. Further inland, there are Sinhala villages and then Ampara town, the Sinhala dominated capital of the district (map 2 and 3). Scholars with an interest in the region will know Akkaraipattu through Dennis McGilvray's detailed ethnography of the town (1982; 2008). His primary focus lay with castes and kudis (clans) among the Tamils, but part of his work highlights the deep historic inter-connections with the “Moorish” community as well. Whilst the Muslims do not have castes, their kudi system is largely an extension of the Tamil tradition. The system is matrilineal and matrilocal and in both communities it is the key regulatory mechanism for marriage, dowry, and inheritance and thus reproduces socio-economic relations and hierarchies. As elsewhere in Sri Lanka, Islam came to Akkaraipattu through Muslim traders who crossed the Indian Ocean centuries ago, settled and married local (Tamil) women who converted to Islam. Religious practice was shaped by the arrival of a Yemenite sheikh, who established a Sufi tradition (McGilvray 2008). McGilvray's account of social practice provides an exceptionally luxurious basis for understanding a complicated town like Akkaraipattu. Yet, many questions remain unanswered, particularly those concerning recent religious transformations in the Muslim community and the political developments associated with the escalation of violent conflict in the past twenty-five years.

The Tamil militancy, military intervention, ethnic clashes, political strife, breakaway factions and “gunmen” marked Akkaraipattu from the 1980s onwards. As elsewhere in the north and east of the country, recruitment, extortion, assassinations, and intimidation became hallmarks of everyday life. Ethnic discrimination by the state, the language policy, and Sinhala-dominated colonization schemes in the post-independence period raised anxiety among both Tamils and Muslims. Particularly in the east, numerous Muslim youth are said to have joined Tamil militant groups in their struggle against the state in the 1980s. For reasons that were never fully elucidated, however, the Tamil rebels and the Muslims broke apart. Throughout the 1990s, a long sequence of violent incidents occurred between the Tamil and Muslim community, tearing apart the deep historic socio-cultural ties between the two groups. The restoration of “normalcy” announced with the 2002 ceasefire between the government and the LTTE was short-lived. Despite the truce, ethnic tensions, assassinations, extortion and intimidation continued and peaked when a split occurred within the LTTE in March 2004. The December 2004 tsunami further disrupted people's lives. War resumed in 2006 and resulted in a government victory over the LTTE in the east (2007) and the country at large (2009). The period of study was thus rife with uncertainties, tensions and violent incidents.

The Transformation of Ethnic and Religious Identity

The following paragraphs will discuss some of the remarkable changes that have taken place in the ethnic, political and religious realm in the past twenty-five years. The war has had multiple and paradoxical impacts on Akkaraipattu's Muslim community. The following four inter-connected developments form the empirical core of this article: 1) a hardening of ethnic fault lines, 2) the increasingly ethno-territorial meaning attached to Sufism, 3) the attempt to unify the Muslim community through the creation of the Mosque Federation, and 4) the project of Islamic purification propagated by Tabligh Jamaat.

Hardening ethnic fault lines

The eastern coastal belt is known for its long history of interconnections between Tamils and Muslims. This is reflected in cross-group similarities in the kudi system and social organization. Akkaraipattu is one of the few places where a Muslim and a Tamil enclave actually form one town, but the ethnic rift is as manifest in Akkaraipattu as it is throughout the east. Already prior to the escalation of hostilities, political contestation resulted in an administrative break-up. In the early 1980s, separate divisions were created for the Tamil (Alayadivembu division) and the Muslim (Akkaraipattu division) part of town. This resulted in a tug of war around the exact location of the new boundary, as the ethnic geography and the position of key facilities – the market, the bus stand, the Grand Mosque, the Catholic Church – defied a clear-cut divide.

The escalation of conflict further reinforced the ethnic fault line. In an attempt to establish control, both the Tamil rebels and the state security forces resorted to severe measures. They targeted the two ethnic communities in different ways and thus compounded the Tamil-Muslim rift. The LTTE came to see Muslims as menacing traitors and acted accordingly. The security forces saw the Tamil community as a potential security threat; intimidation, round-ups, arrests, torturous interrogations, killings and disappearances thus affected them particularly badly.

The national conflict often became manifest in Akkaraipattu through more localized struggles. In 1985, a fight between Muslim and Tamil users of the market place got out of hand and the rebels took their revenge on Muslim traders by burning it down. In 1990, ethnic anxiety and contestation over land led Muslims to destroy a Hindu Temple (see the shattered image on the right of map 4). Grenade attacks and sudden escalations of violence have been rife. Carved into the collective local memory, such incidents fortify the division between the two communities. Traditional linkages and inter-personal relations have been under pressure.

The schism between the two communities is aggravated by unequal access to political power and patronage networks. The ability of Muslim political leaders to tap into such networks has borne fruits for the Muslim division, as evidenced by the general state of the infrastructure: the base hospital, roads, a conference hall, and the prestigious Islamic gates at the northern entry of town. The numerous shops, petrol sheds and restaurants are evidence of greater mobility among Muslims. They could cross military checkpoints to do business in Colombo more smoothly than Tamils, though the general trading tradition attributed to the Muslims may be a factor here as well. It would nevertheless be a mistake to single out a Tamil and a Muslim economy. Both are in fact deeply connected: economic exchange between the Tamils and Muslims is highly common in Akkaraipattu. Tamil labor is vital for Muslim construction projects and agriculture. Hiring, selling and buying take place on a daily basis, though it must be added that the Muslim community tends to be the more powerful actor in economic exchange.

The paradoxical combination of ethnically colored conflict and economic inter-dependence drives the pulse of life in Akkaraipattu. The situation oscillates between sudden eruptions of tension and violence and subsequent resumptions of “normalcy” through trade, agriculture, school and so on. Community leaders as well as workers and traders on both sides of the divide agree that economic interests are a driving force in ending temporal crises. The common cadence tends to encompass general tensions, a violent incident and possibly a reprisal, which in turn result in a public shut down (hartal) imposed by either side. Whilst efforts are made to defuse the crisis, people in both communities – employees, employers, sellers, consumers – are eager to continue their everyday struggle for income and the situation gradually returns to square one. Neither side can afford to let enmity get in the way for more than a few days.

The reinvigorated genealogy of Sufi shrines

Wet rice has historically formed the heart of Akkaraipattu's economy. The irrigated rice fields surrounding the town are subject to increasing competition due to population growth and ethno-political boundaries. This has resulted in local conflicts, which were further complicated by the context of state sponsored land colonization, ethnic contestation, guerrilla warfare, and tsunami relocation. Land, in fact, became a conjunction of economic interests, ethnic anxieties, and genealogical claims. The shrine at Ambalattaru forms an interesting example.

The mosque was erected in a rather remote location some 17 kilometers southwest of town (see map 3). There are contesting accounts of the date of erection. Some claim it dates back to the 17^th Century, but the more plausible account that came out of McGilvray's work (private communication) suggests the shrine was built in the early 1950s when the jungle was cleared for the Gal Oya irrigation scheme. Legend has it that the earth-moving equipment froze up and the previously unknown saint Sikander emerged in a dream to explain he had been buried in the forest for centuries. The mosque was thus built to appease him with a saint tomb. There is a long history of Sri Lankan Muslims worshipping such tombs inside mosques to make vows and ask for help with their problems. Whilst the rituals and symbols are Islamic, shrines like this have attracted Hindus and Christians as well (McGilvray 2008, 273–275).

The location of the Ambalattaru mosque bestowed it with special importance. What used to be a peripheral jungle later became a controversial ethnic borderland. The shrine became the marker of a fault line that not only defined claims on farmland, but an ethnic space and a historic right to inhabit that space as well. It lies on the very edge of the paddy fields, on the verge between the Muslim and Tamil dominated coastline and the Sinhala dominated inland. Beyond the mosque, one finds jungle on one side and the Sinhala settlements created under the Gal Oya irrigation scheme on the other. Poor Sinhala farmers from the south were brought in to exploit the newly gained agricultural potential. As elsewhere in the north and east, these developments invoked Tamil and Muslim anxieties about Sinhala encroachment (McGilvray 2008; Moore 1985). This led the Muslim leadership to attract new Muslim farmers to the area. They were brought in all the way from Welimade (a Sinhala speaking area in Sri Lanka's central highlands). Apparently, the political compromise at the time required the new settlers to speak Sinhala. Ambalattaru was given a Sinhalized name: Ambalam Oya.

With time, the mosque started to decay, but when ethno-political contestation flared up again in the 1980s and 1990s, it resurfaced on the political radar. A local Muslim politician renovated the building and SLMC leader Ashraff provided electricity. The mosque thus experienced a rebirth, which had little to do with religious dynamics, but everything with the land politics of the time. Every morning there is a vow-making ceremony and pilgrims come from all across the east coast to pray and ask for help. Over 10.000 attend the annual festival, dignitaries at the mosque told me, and they bring cows and other offerings. Sufi followers in Ambalattaru and other Sufi shrines in the region, voiced anxiety during interviews that more puritan Muslim reform movements (mainly Tawhid Jamaat) oppose these traditions and heritages. But in the east of Sri Lanka, Muslims have to cling on to their history, they explained, because they are under threat from Sinhala and Tamil nationalists, who argue that Muslims are not from this soil. Shrines like the one in Ambalattaru are pivotal for the Muslims’ sense of belonging and they play a vital role in contemporary interpretations of “ancient” history.

The mosque thus assumed new significance in the region's ethno-political field. Genealogy, belonging and the historical nexus between place and religion became paramount for a minority community that perceives itself as “under siege”. Rather than a mere prayer house, the mosque provides meaning to contemporary notions of place and belonging, a boundary between “us” and “them”. It became a marker of a discourse that implicitly emanates notions like “traditional homeland” and “sons of the soil”, both dominant features of Tamil and Sinhala nationalism.

The mosque federation

Apart from transforming the symbolic and territorial meaning attached to mosques, the conflict has also affected the role of mosque organizations in society. Mirroring the institutional setup of Hindu temples, mosques have long been administered by kudi-based trustee boards. Until the 1980s the role of these boards was mostly confined to the religious realm and its wider worldly significance was declining. This becomes clear from McGilvray's work (2008, 275–279) as well as interviews with one of the former chairmen of the federation and other Muslim leaders. Board members were selected from the mosque-affiliated kudis to settle disputes within the mosque constituency and run mosque affairs, like maintenance and religious festivals. In response to the insecurity and ethnic tensions, that role has changed in a major way.

Akkaraipatttu's three main mosques jointly founded a federation in 1983 to oversee social activities and facilitate the implementation of shariah law. Neighboring towns followed suit with similar bodies. Soon after, in 1985, the first ethnic clashes caused havoc in Akkaraipattu and the Muslim community felt an urgent need to get organized. People expected the Islamic leaders to settle inter-ethnic problems and disputes on their behalf. In the following fifteen years, the federation became strongly preoccupied with preserving the unity and safety of the Muslim community.

The 2002 peace talks between the government and the LTTE further intensified this role. Instigated riots, high profile killings, extortions and disappearances ravaged the region as the LTTE, the army and Muslim politicians competed for control in the shadow of the formal negotiations about the region's political future. The following incident is illustrative of the federation's capacity to settle such unruly situations. During a Friday morning prayer in November 2005, a volatile period because of the presidential elections at the time, Tamil rebels threw a grenade into the praying crowd at the Grand Mosque, killing eight and wounding some thirty-five Muslims. The mosque lies right on the ethnic fault line. The rebels quickly made their way out, leaving behind a shocked and furious Muslim crowd, which was ready to take revenge on the Tamil part of town. Sri Lanka's history is rife with ethnic riots, even pogroms, sparked by incidents like this. In this particular case, however, the escalation of violence was prevented, because the religious leaders stepped in. The maulavi preaching in the mosque when the grenade exploded told me he addressed the crowd, saying: “please behave like real human beings. Don't do any harm to anyone.” He said this through the mosque loudspeakers immediately after the incident. They turned the volume up, so even in the Tamil area it could be heard. Meanwhile, the police moved in with armored cars and all the troops they could spare. Muslim leaders continued to talk to the crowd saying: “don't go harm anyone. This is the result of madness. Don't go hurt the Tamils. […] Allah will take care of it, you go home.” The maulavis were successful and an outright ethnic clash was averted (see also Goodhand et al. 2009).

Islamization and Tabligh Jamaat

In parallel to the war-related developments discussed above, remarkable changes occurred in the religious sphere. In the past three decades, the number of mosques has increased rapidly and the practice of alms giving has been formalized in a Zakat foundation. Public displays of Islam – dress, haircut, beard – have changed as well. More pious inhabitants of the town prided themselves in pointing out that an increasing number of men shave their heads and grow their beards. Some women have started to wear a black niqab that fully covers their face (though this is not as common as in some other towns in the east, most notably Kattankudy). This Islamization process was mainly spearheaded by Tabligh Jamaat, an Islamic reform movement. It was founded by the Indian Islamic scholar Muhammad Ilyas in the 1920s to call on Muslims to bring their everyday life in line with Islam. The credo of the movement can be summarized in his quote “O Muslims, become Muslims”. Whilst relatively little has been written about Tabligh, they have become a big player, certainly in South Asia, but also globally (Ahmad 1991; Metcalf 1982; Robinson 2007; Sikand 2007). The movement spread along Sri Lanka's east coast and to Akkaraipattu in the 1970s. Other groups – Jamaat-i-Islami and Tawhid Jamaat – came to the town as well, but their presence remains small. In many mosques and in the mosque federation Tabligh Jamaat has a strong presence and presently they are building a large dedicated Tabligh mosque along the main road.

Tablighis aim to “purify” Islam from “undesirable” customs that Islam has blended with during the centuries of its expansion. Sufi practices like mysticism, meditation and saint worship are tolerated as an advanced form of Islam, but they are not encouraged. Tablighis see themselves as a da'wa movement, that is they educate the rural masses and disentangle them from “un-Islamic” habits (Robinson 2007; Sikand 2007). During my fieldwork Tablighis proudly presented their teachings as “simple”, as a way of life that anyone can adopt. They shun theological debate and interpret purification in terms of doing everyday things – eating, praying, dressing and so on – in an Islamic way (see also Ahmad 1991; Metcalf 1982). All can reach the afterlife by following this very simple path, according to their belief.

Tabligh Jamaat is horizontally organized and claims not to have elaborate financial or institutional structures. Members are not recorded and for this reason it is hard to estimate the size of Tabligh's following, but one Tablighi estimated the number of active members in Akkaraipattu at one thousand at the time of research. “Jamaat visits” form the core of its activity: members of the movement travel to other places (neighboring villages or abroad) for some days, weeks or months for devotional practice and spreading the word to other Muslims. These visits take place on a regular basis, often by men, but in some cases women are brought along as well.

I have not been able to reconstruct how exactly Tabligh Jamaat entered Akkaraipattu, but Victor de Munck's analysis (1998) of the arrival of the movement in a village in Moneragala District, further south, provides a plausible illustration. Interestingly, the non-Tablighi villagers De Munck describes acknowledged that their white-clad, bearded visitors propagate the proper Islam, but they avoided interaction with them, sighed with relief when they survived the visit without being converted and made inappropriate jokes about them at the local store afterwards. They thus averted absorption into the movement, but did not challenge the Tablighi concept of purity. This purification concept is the leading theme for those who did join the movement as well. One member I interviewed in Akkaraipattu said the following: “Before I joined Tabligh, I had no life. I didn't go to the mosque and I didn't obey my parents. […] Before, I thought my life was spoilt. I joined Tabligh because I felt I needed to correct my life. Go the path of Allah.”

Remarkably, the Tabligh purification project steers clear of ethnicity altogether. They actively try to shut out key planks in above-discussed Sufi discourse of genealogy and a local sense of belonging, and the inter-ethnic dimension escapes their narratives. Tabligh is all about Muslim identity, but the movement explicitly disconnects itself from worldly matters like identity politics and ethnic strife, because they are seen as impure, inherently divisive, and a source of trouble. “We don't worry about politics, shariah, or an Islamic state,” one member said. Another added: “I don't believe we can decide by politics. God is the decision maker. Politicians can't decide over the life of the people.” Politics were the cause of Sri Lanka's “ethnic problem”, and they will not help resolve it. “The solution will come if the Muslims obey the word of Allah. With politics, one problem is solved and another one will come.”

Divergent Muslim identities, demarcations and notions of purity

As becomes clear from the above developments, the war and the hardening of ethnic fault lines have had great impact in Akkaraipattu. That impact goes beyond a simple us/them divide, however. It is not just a story of violence and the mobilization of identity leading to a segregation of Muslims from Tamils and in turn a unification of Muslims around Islam. The religious field harbors multiple trajectories. We have witnessed how a Sufi shrine became an ethno-territorial marker in the politics of place, a mosque federation that expanded its role to deal with violence and contention, and an Islamic purification project (Tabligh Jamaat) that steers clear from ethnicity and politics. These developments all pivot around an increased awareness of identity. Indeed, it matters a whole lot more to be Muslim now that this category determines friend/foe relationships and security conditions. Islamic institutions are thus preoccupied with demarcating Muslim from non-Muslim, but they define “Muslimness” in rather different ways.

In the case of the Ambalattaru shrine, Muslimness airs ethnic genealogy and territorial belonging. For Tabligh Jamaat, on the other hand, the category Muslim denotes a sense of religious purity. It is a deliberate move away from territory and genealogy; Islamization is about shedding those un-Islamic influences. Tabligh Jamaat's discourse is about the global umma; divisions along lines of nationality and ethnicity are secondary. The mosque federation, on the other hand, has an explicitly local focus. It is about uniting the community and shutting out divisive influences and external threats.

One important distinction concerns the relation of these discourses and institutions to “the political”, which brings us to the next step in this paper: the paradoxical relations between religion and politics. The case of Akkaraipattu flags divergent connections. The mosque federation constitutes the expansion of a previously merely religious body into a broader scope of public action. It tries to address politically charged problems by uniting the community and by mobilizing political leaders. The Ambalattaru Sufi shrine, on the other hand, sheds a rather different light on the political connection. Politicians funded the refurbishment of the mosque to bolster their image as good Muslim patrons and thus reach out to relevant vote banks. Tabligh, thirdly, claims to be neither about patronage, nor about principled politics. They stay out of politics altogether, because it is a divisive factor that generates problems rather than resolving them, members of the movement would not get tired of saying. Somewhat schematically, we could thus distinguish principled politics (concerned with the rights and anxieties of a population group), pragmatic politics (pivoting around patronage) and anti-politics (attempts to turn away from politics at large). Whilst such firm discursive positions are important to understand, readers will not be surprised to find people circumventing principles or jockeying between discourses in the section below. As Akkaraipattu oscillates between escalations of violence and resumptions of relative peace, different outlooks and boundaries are activated and de-activated. The religious field conditions people's possibilities, but does not script their behavior altogether.

Religion, Politics and Patronage

Let me shortly change perspective here and look at the role of politicians.⁶ The unruliness of the electorate and the dialectic between principled and pragmatic politics are a challenge for them. To understand their position, we need to take the broader narrative of modern Muslim politics in Sri Lanka on board. Even a nutshell summary sketches the dynamics quite clearly. Pragmatism was the dominant feature of Muslim politics up to the mid-1980s. Muslim politicians tended to engage in what O'Sullivan (1999) has labeled “the politics of access”: they managed to deliver patronage to their electorates by adopting a non-militant stance within the two mainstream (Sinhala dominated) parties. In Akkaraipattu, Uthama Lebbe was a prominent example of this approach. During the 1980s, a more militant discourse of Muslim ethno-nationalism and minority rights emerged. This was manifest in the creation of the Muslim United Liberation Front (MULF), founded by Akkaraipattu's activist M.I.M. Mohideen, and its more successful parallel: the SLMC, run by M.H.M. Ashraff (from Kalmunai). Both parties propagated Muslim rights and a form of self-determination, thus partly mirroring Tamil nationalism. However, the MULF never became a powerful player and the SLMC only survived because its ethno-nationalist ambitions were trumped by realpolitik. Despite its Islamic outlook and presentation, it joined Sinhala-led administrations from opposite sides of the political spectrum to secure ministerial posts. Politics of access thus continued – be it with a shift of patronage towards east Sri Lanka. The ethno-political rhetoric of the SLMC could have potentially united the Muslim community under one banner, but its patronage politics had the opposite effect. Town-based patrimonial networks of individual “big men”, like A.L.M. Athaulla in Akkaraipattu, divided the party. After the death of Ashraff in 2000, these politicians competed for ministerial posts, either on an SLMC ticket or with their own electoral vehicles.

Politicians like Athaulla have to juggle. Being too militant would bar their access to key positions, but they cannot be unresponsive to the various pressures from their constituency either. To garner support, they must navigate the divisions within that constituency. How to engage with the ethnic geography of a Sufi shrine without antagonizing Tablighis? Overly outspoken ideologies – Islamism, ethno-nationalism – create difficulties for them, because they divide their vote banks or impede their ability to join government. Instead the tendency is to make oneself unassailable as the town patron by monopolizing patronage. The engagement with religion therefore tends to centre not on political issues, but on symbolism – funding mosques, constructing religious landmarks, and fraternizing with respected religious leaders. The message of the two big Islamic entry gates spanning the main road and the modern conference centre (see map 4) will not be lost on a visitor entering Akkaraipattu: this is the town of A.L.M. Athaulla (his name and picture are clearly posted), he is a good Muslim and he is making Akkaraipattu prosperous. At the 2008 Provincial Council elections, he assembled mosque leaders and the mosque federation and told them whom the people should vote for. The patrimonial logic thus connects religion and politics in a marriage of convenience. It does not engage in theological debate or setting religiously inspired political objectives. Big men engage with religion through symbols, people and institutions to connect to their clientele.

The political, politics and anti-politics in practice

The view from below highlights similar difficulties in balancing contradictory loyalties. The community and religious leadership in Akkaraipattu use both political and anti-political behavior in their everyday struggle for survival. On one hand they have to do politics, to deal with immediate crises, as the grenade incident at the Grand Mosque illustrated, but also on a personal level: to safeguard jobs and personal favors. On the other hand, people in Akkaraipattu have plenty of reason to turn away from politics. Not only because they see politics as abject and impure, but also because it tends to endanger everyday security and survival. As usual, there are important differences in Akkaraipattu between what people say they do and what they actually do, particularly in relation to something inherently controversial like politics. People air discourses about politics or anti-politics, but in fact they practice both. The paragraphs below discuss some examples of how people negotiate the various discursive contradictions and limitations.

Let me start with the testimony of a senior civil servant who has previously enjoyed influential positions in Akkaraipattu's religious establishment. Religious leaders will present themselves as pious and non-political, he explained, but at election time, their mosques and constituencies become political currency and they cannot resist the pressure. Politicians will use them to marshal support and if they object, they will face problems. “It may not be right, but we think like this. We need a job for our son, so we go to the minister. That's how elections work here. To get Samurdhi [government support for the poor], a loan, anything, you go through the minister. He calls me in the office ‘do this, do that’ and makes me contravene our own policy and break the law. But he says: ‘I'll look after it. You must do it. What are you telling.’” Political power thus trumps anti-political conviction when push comes to shove on a personal level. He continued: “Even people who follow Tabligh and Tawhid need personal favors. To the outside they are anti-political, but inside it's different. I have also called friends to tell them do this and that. Violence is also used. Even the police and army would come in for certain purposes.”

A second example concerns the personal struggle that a senior member of Tabligh Jamaat shared with me. He is one of the members of the movement who is particularly insistent with regard to piety. As an educated person, he will eloquently explain why politics will not lead to anything. One day, SLMC leader Ashraff sent two people to invite him to speak at a political meeting. He explained his position, but they suggested he should speak about the need for Muslim unity, a message he could not disagree with. He could not withstand their pressure, but, he added, “I spoke only about God. Not about politics.” While this Tabligh leader upheld his personal boundary, he became part of a political event and his participation will have boosted its political symbolism and legitimacy.

The Mosque Federation, a third example, is itself a manifestation of contradictions. It sees itself as a religious organization that contrasts itself with the polity (like the Tablighis), but it also constitutes a response to the ethnic violence in the 1980s and its narrative emanates locality, belonging and ethnic geography (like the Ambalattaru shrine). In addition, the federation engaged with formal politics as it tried to optimize the vote by uniting the Muslims in elections. This, however, proved to be beyond their capacity; other forces trying to influence voting behavior proved stronger. Then, the federation attempted with increasing vigor to influence the course of Muslim political leaders. As the war progressed, the federation joined hands with other mosque federations, first at district and later at provincial level, to advocate Muslim aspirations. With the 2002 peace process, Muslim anxieties peaked as their political future was debated in the absence of a Muslim delegation. Popular declarations were launched in Muthur and at the university campus in Oluvil (near Ninthavur) (map 2) to propagate the Muslim rights to ethnic self-determination and demand that Muslim political leaders defend such a position. The mosque federation was one of the forces behind these statements. In addition, they set up meetings with various Muslim politicians, calling on them to overcome their differences and reunite in one Muslim front, but this failed. The Muslim polity remains divided and no separate delegation was added to the peace talks (which had started to collapse anyway). The Mosque Federation thus spans principled, pragmatic and anti-politics. It tries to deal with the political in an anti-political way by keeping the distorting and divisive influence of politics out (for example during the mosque incident), but meanwhile it attempts to influence politics, by calling on politicians to engage with the political in a more responsible way (lobby during the peace process).⁷

Conclusion

This article contributes to the wider literature on contemporary Islam and politics that takes issue with dominant concepts like secularism and Islamism and the debate about Islam's relation to modernity and liberal democracy. These writings debunk the idea that any political system is essential to Islam and posit that more contextualized and nuanced approaches are needed to understand the variegated interface between Islam and politics. So far, this literature has remained largely disconnected from Sri Lankan studies. Whilst Sri Lankan Muslims have become a somewhat fashionable topic in recent years, there is a tendency to either deal with the religious and the political angle separately, or subsume one into the other. This article has shown that the dynamic between the religious and the political is much less straightforward and in fact harbors a number of paradoxes. Identity issues have assumed an omnipresent importance in course of the war in Sri Lanka: it matters a whole lot more to be or not to be Muslim now that it defines friend and foe, ethnic geographies and everyday security. However, the meanings attached to that category – Muslim – vary and this leads to paradoxical differences of emphasis: the ethnic dimension versus the religious one, local versus global, political versus anti-political and so on. Salient examples presented in this article included the discourse around the Ambalattaru Sufi shrine, which thrives on Muslim genealogy and the need to preserve the purity of a historical Muslim place under threat of Sinhala encroachment. Tabligh Jamaat followers on the other hand define purity in terms of Islamization and see ethnic sentiments and local politics as worldly digressions. Coming from a different angle, the Muslim ethno-nationalist discourse steers clear from religion, which in turn poses a challenge for Muslim politicians, who engage with Islamic leaders and symbols rather superficially and opportunistically. Meanwhile the mosque federation present themselves as apolitical, but when they try to orchestrate community behavior (by preventing pogroms or trying to unify the vote) they enter a highly political domain. In all these cases, the Muslim identity is pivotal, but the resulting religio-political connections or disconnections diverge.

This observation matters, because it helps explain some of the divisions within Sri Lanka's Muslim community at large. It helps resolve the puzzle of a group of people that often fails to unite around shared interests to jointly face common enemies, despite facing major existential threats: from the state and from the rebels, from Sinhalese and Tamils, from Christians and Buddhists. Furthermore, it counters implicit and explicit prophecies of imminent Muslim “jihadi” violence by wahabi-inspired groups. Instead, this article highlights the internal cleavages of identity politics: the war has united and divided the Muslims. It moves beyond a dichotomy of “radical” and “moderate” Muslims and the question of who is going to emerge on top. Rather than a struggle between dissatisfied armed youth falling pray to fundamentalism on one hand and liberal peaceful cosmopolitan Muslims on the other, we witness a much wider variety of discourses propagating divergent ethnic, religious and political sentiments. And as a result there are intra-Muslim fissures, contradictory political outlooks, paradoxical notions of purity and different ways of defining ingroup and outgroup. Rather than a dichotomy of two fronts (be they labeled as traditional/modern, radical/moderate, piety/politics, or fundamentalist/secular), people position themselves in different ways as they navigate their everyday life in a context that oscillates between violent skirmishes and periods of relative peace. They employ different discourses to engage or disengage with politics and if their principles do not suit the situation, they find pragmatic ways to circumvent them. What is deeply political at one instance may re-appear as pious the next, depending on the occasion.

As a result of these convolutions, the religious field cannot simply be singled out. Far from a privileged spiritual arena, it is structured by incidents of violence, patrimonial loyalties, political contestation, and ethnic identity issues. Muslim identities are not merely a function of religious dogma, but result from continuous reproduction through social relations, religious traditions and reform movements, economic activities, violence and political dynamics. This article thus supports the emphasis of recent studies on the ways contemporary Muslims produce themselves in their everyday life (Bayat 2007; Soares and Osella 2009). That does not mean, however that the religious field is entirely contingent, or that no overall trends can be identified. Important changes are taking place: Islamization, a hardening of ethnic fault lines, the soaring importance of locality and genealogy, and the context of protracted insecurity. These changes bear testimony to the fact that the religious field is influenced by various factors. Armed conflict, ethno-nationalism, and the proliferation of trans-national movements like Tabligh all had an impact on the way religion and politics interact, but this impact was neither homogeneous nor hegemonic. Violence and ethnic contestation have raised the stakes, but this applies both to political and to anti-political behavior. To return to the metaphor of Muslim prayer on a ferry cited at the inception of this article, religion does not remain unaffected when it enters new terrain. The current, wind, and the waves exert force, but their influences are not uniform; they may work with or against each other. War, ethnicity and Islamic revival movements raise questions of orientation, identity and boundaries, but rather than clear answers, we find multiple trajectories and divergent individual ways of navigating the prevalent conditions.