Governance Practices and Symbolism: De facto sovereignty and public authority in ‘Tigerland’

Abstract

This article focuses on how the Liberation Tigers of Tamil Eelam (LTTE) insurgency performed de facto sovereignty and public authority in Northeastern Sri Lanka. It is situated within the wider academic debate on governance by state and non-state actors. We venture to unravel the complex linkages between the LTTE's governance practices and legitimation strategies by looking at narratives, performances, and inscriptions. While monopolizing the justice and policing sectors, in other sectors the LTTE operated pragmatically in conjunction with the state. The organization tried to generate and sustain public authority and legitimacy through a variety of violent and non-violent practices and symbols. It ‘mimicked’ statehood by deploying, among others, policing, uniforms, ceremonies, nationalist songs, commemorations of combatants, and the media. This not only consolidated its grip on the Northeast, but also engineered a level of support and compliance. We conclude that the LTTE's governance included practices that were created and carried out independently from the Sri Lankan state, while others took shape within a pre-existing political order and service provision by the state. The article elucidates the LTTE's mimicry of the state, as well as the operation of parallel structures and hybrid forms of state-LTTE collaboration. This facilitates a nuanced understanding of rebel governance beyond a simple state versus non-state binary.

Introduction

Journalistic and policy accounts of rebel groups are frequently dominated by one-sided images of warlords, organized crime, human rights abuses, child soldiers, and natural resource plunder. Scholarly research, however, increasingly demonstrates that in many cases armed groups perform substantial forms of governance, often in tandem with predatory practices.¹ It can be argued that they possess de facto sovereignty and execute public authority, as highlighted in this special issue. However, there is a limited understanding of the empirical manifestation and practical dynamics of rebels’ sovereignty and public authority, and the political legitimacy they derive from it. This reminds us of a similar observation made by Ferguson and Gupta with regard to states:

The metaphors through which states are imagined are important (. . .). But the understanding of the social practices through which these images are made effective and are experienced is less developed. This relative inattention to state practices seems peculiar, because states in fact invest a good deal of effort in developing procedures and practices to ensure that they are imagined in some ways rather than others.²

This applies, we would argue, to an even larger degree to non-state armed actors involved in governance. The Liberation Tigers of Tamil Eelam (LTTE) in Sri Lanka are a clear case in point: it accounted for severe predatory behaviour towards civilian populations, simultaneously paired with the performance of substantial governance practices, leading to a de facto sovereignty and public authority. Some of these governance practices, including those of the Tamil Eelam police force and judiciary, functioned independently from the Sri Lankan state, especially in areas under full LTTE control. Other practices were, however, shared with the Sri Lankan state to a larger or lesser degree, particularly in sectors such as health care and education, and in zones of contested territorial control.

Whereas it has long since been accepted in academic literature that various forms of governance or ‘governmentality’ can be exercised by a variety of actors at the same time, this article hopes to shed light on the specific ways and dynamics through which this has been done in LTTE-controlled areas in Sri Lanka, and how that has been perceived and experienced by the local population, thereby paying attention to processes of legitimation and power. First, we introduce our theoretical considerations and discuss the emergence of the LTTE. This will be followed by a discussion of the LTTE's sovereignty and law enforcement, its public service provision, the existence of hybrid rule and authority, and, finally, the LTTE's symbolic legitimation of its rule. Following Schröder and Schmidt we pay attention to narratives, performances, and inscriptions.³ We conclude that the empirical manifestation of LTTE governance is more complex than a simple state versus non-state binary would suggest, ranging from mimicry of statehood and parallelism to hybridity.

Theoretical considerations

De facto sovereignty and public authority

This article departs from the assumption that sovereignty is not the prerogative of the state, but can de facto be practised by a non-state armed actor. Hansen and Stepputat introduce the concept of ‘de facto sovereignty’ in opposition to the traditional understanding of the concept which portrays the sovereign state as the bedrock of a civilized international order.⁴ By detaching sovereignty from the state, they direct our attention to its practice. They define de facto sovereignty as ‘the ability to kill, punish, and discipline with impunity’,⁵ which, however, denotes a rather absolutist version of sovereignty. When control is practised predominantly by military or police forces through the use or threat of violence, it is sometimes called ‘coercive sovereignty’.⁶ As sovereignty has become increasingly circumscribed, strategies of legitimation and a level of consent among the ruled become an increasingly important part of the equation, too (in this form it has been conceptualized as ‘representative’ or ‘popular’ sovereignty).⁷ In fact, there is a top-down and bottom-up dimension to it. It is a question of using empirical analysis to establish how these two are manifested in practice.

We will investigate how the LTTE was able to legitimize and institutionalize its de facto sovereignty; this is where another key concept of this special issue—public authority—becomes useful. Following Weber's conception of authority as ‘legitimate domination’, Sikor and Lund define authority as:

an instance of power that is associated with at least a minimum of voluntary compliance [. . .It] characterizes the capacity of politico-legal institutions, such as states and their constituent institutions, village communities, religious groupings and other organizations, to influence other social actors.⁸

When using this conceptualization two aspects stand out. First, we see that public authority is essentially about legitimacy and that it requires a certain form of consent from its constituency beyond the sheer exercise of force by the power-holder. Here the concept shows a parallel with the bottom-up dimension of sovereignty. Secondly, public authority is inherently relational. It consists of vertical, mutual connections between those actors who possess authority and the constituency that complies with it.⁹ More recent scholarship has started to conceptualize ‘state’ and ‘non-state’ forms of public authority as overlapping and interdependent.¹⁰ Following Meagher's conclusion that a ‘shift in theory’ needs to be backed by insights in specific cases, this article contributes to this emerging body of knowledge by means of a case study on the LTTE.¹¹

Governance and governmentality

Contrary to more traditional conceptions of governance, political and social scientists increasingly recognize that the provision of security, welfare, and political representation is not necessarily the prerogative of the state.¹² Though the formal state may still be regarded as the dominant actor in providing public goods, non-state actors—including those that compete with the state—prove to be in possession of the means (i.e. armed forces, taxation, and an administrative system) and ambition to provide a constituency with security (by regulating the internal use of force and offering protection from external threats), welfare (by offering social and utility services), and political representation (through institutions for consultation and normative regulation), usually in a fairly demarcated territory.¹³ Therefore, governance should not only be seen through a state-centric lens, but, more broadly, as the ‘whole set of practices and norms that govern daily life in a specific territory’.¹⁴ As such, our analysis should move from a focus on state exclusivity towards a more comprehensive and agnostic study of governance—with or without a formal government.¹⁵ This position echoes the Foucauldian notion of ‘governmentality’ that—in the words of Sending and Neumann—is ‘an analytical concept aimed at grasping government as a form of power, as the techniques and procedures for directing human behaviour’.¹⁶ According to Sending and Neumann, governmentality is

aimed at investigating the specific practices and techniques of governing as an empirical phenomenon, thus seeking to replace a focus on institutions (. . .) with a focus on practices (. . .) [and] identifying the mentality or rationality that characterizes the systematic thinking and knowledge that is integral to and renders possible different modes of governing.¹⁷

Foucault extended governmentality beyond the domain of the state to include civil society, the family, and personal life. Ferguson and Gupta also discuss how market forces have taken over from government under neo-liberalism and that several forms of transnational governmentality have come into being.¹⁸

Dillon has pointed out that the concepts of governmentality and sovereignty are not necessarily competitive, nor oppositional, but that their relationship can be characterized as complementary.¹⁹ As Dillon explains with regard to sovereignty and governmentality:

That complex interdependence, the complicity of the one in the other, is exhibited in their mutual reliance upon each other and upon the discursive production, dissemination, and consumption of regimes of truth. The will to know, and the will to power, as of course Foucault continually insisted, share the same pedigree.²⁰

The consumption of these ‘regimes of truth’ is also relevant for our understanding of how an insurgency like the LTTE presents its ‘truth’ to its constituents. Despite the fact that the LTTE was never successful in gaining international recognition for an independent state, it undertook various strategies to legitimize its struggle and rule over the population.

Legitimacy and legitimation

The legitimacy of a rebel ruler's involvement in governance may be separated in both a juridical and an empirical dimension.²¹ Despite the absence of juridical legitimacy or international recognition, which is usually difficult to attain, rebel groups may enjoy varying levels of legitimacy among a particular constituency.²² One way to gain legitimacy can be found in ideology. Rebels usually try and produce a more-or-less coherent narrative legitimating their struggle and outlining their political goals. Another is found in the symbolic realm. The symbolic dimension of rebel rule may be essential to underpin and, hence, understand the de facto legitimacy of rebel groups and the way in which legitimacy feeds the subjective and individual identities of the civilian population that lives in a rebel-controlled territory.²³ Legitimation, and the pursuit of authority, may include processes of identity building, through a symbolic repertoire of commemorations, rallies, anthems, flags, and logos, for example.²⁴ These symbolic processes may involve the ‘mimicry’ of state practices, as some rebel groups seek to construct political authority in similar ways to nation states.²⁵ This also highlights the performative practices of legitimacy and authority in daily life and the interaction between the sovereign and the people. As observed by Lund:

One group's challenge of another's grip on governance may thus be staged in terms of claiming the symbols of public authority as well, and as much as exercising the practical tasks of governance. Symbols of public authority are not moored to specific institutions, just as the ‘same’ institution may exercise public authority at one point and be rather insignificant in this respect at another.²⁶

The question, however, remains how and why symbols of legitimation will be effective in civil war and during rebel rule specifically. According to Schröder and Schmidt ‘the most important code of the legitimation of war is its historicity’.²⁷ As they explain, the ‘symbolic meaning of prior wars is re-enacted and reinterpreted in the present, and present violence generates symbolic value to be employed in future confrontations’.²⁸ In other words, current conflicts need a certain discursive link to past events in order to be legitimated. The historicity of present-day confrontations is represented through narratives, performances, and inscriptions, and each of these representational strategies is open to manipulation.²⁹ Demmers points out that through participation in narratives, performances, and inscriptions, people may come to accept and support the (violent) course of action proposed by their leaders as legitimate and justified.³⁰ In other words, there is an interplay between elites and other layers of society where these representational strategies are collectively created, maintained, and changed. Hence, in this article we will investigate how the LTTE attempted to legitimize its rule through various symbols of, and claims to, legitimacy, focusing especially on historical LTTE narratives; performances of statehood, heroism, and martyrdom; and inscriptions.³¹

Methodology

Our research is based on primary, empirical data collected in 2014, 2015, and 2016 as well as secondary research into academic literature and policy documents. At the beginning of the first round of fieldwork in 2014 the Mahinda Rajapaksa administration restricted foreign nationals’ access to the Northern Province. Intelligence agencies and police were keeping a close eye on every foreigner who might openly be asking questions about the LTTE, as a result of the pressure exercised by the international community to hold an independent investigation into the alleged war crimes by the government during the last phase of the war in 2009. The suspicion directed towards foreigners made it difficult, if not impossible or outright dangerous, to carry out field research except through local Sri Lankan researchers.³²

We decided to cooperate with a small local NGO that had intimate, contextual knowledge of the risks and prevailing surveillance in the study area. Local interviewers could carry out interviews unobtrusively with inhabitants (fishermen, farmers, etc.) in areas previously under LTTE control. With every round of fieldwork an introductory workshop and a debriefing were held with the local researchers, after which research instruments and the data collection strategy were fine-tuned. A semi-structured interview guide was used to assess the governance structures and basic service provision in the territories previously under full or partial LTTE control. In total, 76 interviews, averaging two-and-a-half hours each, were held with community members in nine different locations in Trincomalee District (33 interviews), ten locations in Batticaloa District (25 interviews), and four locations in the Northern Province (18 interviews). Subsequently, professional translators translated the Tamil transcripts into English, which were then analysed by us.

After the Sri Lankan presidential elections on 8 January 2015 and the regime change that followed, it became possible for us to visit the Mullaitivu District, Kilinochchi, and Jaffna. In early 2015 we also carried out interviews in the homes and/or offices of key informants such as civil society leaders, community leaders, NGO workers, religious leaders, doctors, ex-LTTE cadres³³ and supporters, and local government officials. A total number of 20 key-informant interviews were held in the Trincomalee District (seven interviews), Batticaloa District (nine interviews), and Jaffna city (four interviews). In April 2016 we did another round of fieldwork comprising 62 interviews in total, focusing on attitudes to and popular support for the LTTE and how the population in the Vanni had experienced the symbolic legitimation of LTTE rule.³⁴ The collected data have in every instance possible been triangulated with secondary literature and other publicly available primary sources.

The rise of Tamil militancy and the LTTE

The framing of the Sri Lankan conflict by the protagonist parties is based on an active ‘reconstruction’ of history and assertion of Sinhalese, Tamil (and Muslim) ethnic identities. Kapferer speaks of the ‘invention of tradition’ and of Sinhalese nationalism as a process of ‘remythologization’ by which chronicles of the past³⁵ are converted into a hegemonic, state-nationalist, Sinhala-Buddhist ideology.³⁶

Portuguese, Dutch, and British colonialism contributed to ethnic, religious, and socio-economic differentiation and societal divisions which were further enhanced by anti-colonial, nationalist, revivalist movements during the nineteenth and twentieth centuries.³⁷

After independence in 1948 the Sri Lankan government disenfranchised one million Indian Tamils who had been brought from South India as plantation labourers by the British. This created serious anxiety among other minority groups, including the Ceylon or Jaffna Tamils. In 1956 the Sri Lanka Freedom Party came to power with a Sinhalese ethno-nationalist agenda. The government declared Sinhalese the sole official national language (‘Sinhala-only’) to the detriment of Tamil and English, thereby effectively excluding Tamils from government jobs.

Tamil voters were deeply alarmed and the (Tamil) Federal Party (FP) demanded a federal state comprising separate Tamil-speaking northern and eastern parts, and a southern Sinhalese part, and that both Sinhala and Tamil be recognized as official languages. It further demanded a stop to state-aided colonization of Tamil areas by Sinhalese farmers.³⁸ Non-violent demonstrations and protests by the FP against these new policies were targeted by Sinhalese mobs and anti-Tamil violence spread across the country. Tamil shops were attacked and looted, and an estimated 150 Tamils were killed.³⁹ In 1957 and 1958 there was again communal violence against Tamils. In the 1960s and 1970s controversial colonization schemes in Tamil areas were implemented, and so-called ‘educational standardization’ hampered Tamil students’ access to university. Finally, the 1972 Constitution awarded special protection to Buddhism and an earlier clause protecting ethnic and religious minorities was removed. In the meantime there were episodes of violence against Tamils in several parts of the country, often with the connivance or complicity of the state and the police, and impunity of the offenders.

Despite growing resentment and frustration among the Tamils, the FP proved unable to achieve any meaningful results in its subsequent negotiations with the government. In 1972 the various Tamil parties joined together to form the Tamil United Front, which was renamed the Tamil United Liberation Front (TULF). At its first convention in Vaddukoddai in 1974 the TULF resolved that it wanted to establish a free sovereign state of Tamil Eelam based on the right of self-determination in order to safeguard the very existence of the Tamil nation. From this moment on, Tamil politics took a fundamental and, with the emergence of a militant Tamil separatist nationalism led by dissatisfied youths, ultimately a violent turn. From the early 1970s onwards, young Tamils began to organize themselves in a variety of radical political groups.

In the early days of Tamil militancy, there were over 30 different groups, and on 5 May 1976 Vellupilai Prabhakaran founded the LTTE. The militant factions acquired arms and received military training from, among others, India and the Palestine Liberation Organization (PLO). These groups would not only attack the Sinhalese state in their attempts to establish Tamil Eelam, but also targeted each other in a search for exclusive leadership and ideological hegemony. The LTTE ultimately succeeded in eliminating its competitors and claimed to be the ‘sole representative of the Tamil speaking people in Sri Lanka’.

After the ambushing of 13 soldiers by the LTTE on 23 July 1983, anti-Tamil riots broke out in Colombo, killing hundreds, if not thousands, of Tamils (estimates go up to 3,000 casualties) and damaging the homes and livelihoods of probably 30,000. An estimated 100,000 Tamils were displaced and 175,000 fled abroad. There is evidence of the government's complicity in organizing those riots.⁴⁰ Soon hereafter the conflict escalated into a full-blown war that was to last for 26 years.

In 1987 the Indian Peace Keeping Forces (IPKF) were sent to the island to stabilize the situation but they were unsuccessful and had to leave in 1990. That year the LTTE forced all Muslims to leave the Jaffna peninsula and the North, so as to create an exclusively Tamil area. The LTTE waged four major ‘Eelam wars’ against the government, and several times also entered into negotiations with them in an attempt to reach a political solution. These failed—or perhaps were never intended to succeed. On 23 February 2002 a Cease-Fire Agreement (CFA) was brokered by the Norwegian government. During the CFA period, the boundaries between the areas controlled by the government and those by the LTTE were delineated and supervised by the Sri Lanka Monitoring Mission. In its area of control, the LTTE ran its own small de facto state. However, after a few years, violence increased and numerous violations of the CFA occurred on both sides. In January 2008 the government formally abrogated the CFA and intensified its campaign, with the aim of vanquishing the LTTE militarily. On 19 May 2009 it finally defeated the LTTE and extinguished its entire military and political leadership.

The LTTE's de facto sovereignty

During the 1990s, there were three regimes of military control in Sri Lanka's war zone. As Gaasbeek explains, these emerged both in formal and in colloquial language, with the English terms ‘cleared’, ‘uncleared’, and ‘grey’.⁴¹ The territory referred to as ‘cleared’ was controlled by the Sri Lankan military (and its paramilitary counterparts) during the day and—officially but not necessarily in practice—also at night.⁴² ‘Uncleared’ territory referred to that controlled by the LTTE both day and night, and territory referred to as ‘grey’ was generally controlled by the Sri Lankan military during the day and by the LTTE at night, but could also be visited by government-allied Tamil paramilitary groups.⁴³ In the eastern areas, the patchwork of fragmented sovereignty was the most complicated, but on the local level this became the ‘normal’ state of affairs.⁴⁴ This area had turned into a region ‘fragmented by frontlines, checkpoints, curfews and entrenched ethno-political boundaries’.⁴⁵ After the withdrawal of the IPKF in 1990 the Jaffna peninsula came under full control of the LTTE, thereby becoming the first locality to come under the de facto sovereignty of the LTTE.⁴⁶ Although the LTTE lost military control of the Jaffna peninsula in 1995, it expanded its control in the Vanni and eastern parts of the country well into 2004 when its predominance waned after the defection of the Karuna faction to the government.⁴⁷

Equally important, the nature and degree of the LTTE's de facto sovereignty varied across the different geographical regions. Its influence in the North was very different from that in the East. In the North, the LTTE was for a long time in control of a large territory known as the Vanni. Its military strength was concentrated here, and its grip on the population was also the strongest in this area.⁴⁸ One could posit that here the movement had gained autonomy from the Sri Lankan state, at least in the security and judicial sectors. In the East, territorial control was more fragmented and different spheres of influence clashed or intermingled.⁴⁹ As Korf and colleagues explain, at the end of the 1990s multiple, coexisting orders and systems of rule had emerged, mixing spaces of authority with the state apparatus: LTTE rule; rule of the Sri Lankan security forces; and the customary norms of religion, caste, and class.⁵⁰ Moreover, in some localities these systems were constantly oscillating, as front lines shifted back and forth over short time frames.⁵¹ There were also significant differences between the centre and periphery of the Northeast, as the government usually had more influence in provincial centres such as Trincomalee and Batticaloa, while the LTTE exerted more influence in the remote areas.⁵² More so than in the North, the East was ethnically very diverse and therefore also a site of ‘multicultural contestation’.⁵³

Law enforcement

The LTTE practised its de facto sovereignty initially through its own police force and judiciary. It began institutionalizing its own police stations in Jaffna and the Vanni in the early 1990s and continued to do so over the following decade.⁵⁴ Particularly from the second half of the 1990s up until the peace negotiations in 2002 the LTTE seems to have invested considerable effort into policing the areas under its control. In the perception of the respondents living under the auspices of the LTTE police forces, they functioned like the real thing. As one of the respondents puts it:

(. . .) they [the LTTE police forces] were 100 per cent policemen. They would wear a police uniform, you know like a nice uniform. And there would be no bribes! No corruption! If you wanted to pay them, it would not work, you would get punished.⁵⁵

Within the police force there were different sections: crime prevention, traffic, a technical division, transport, communications, camera/photography, intelligence, and the environmental police.⁵⁶ Similarly to the fragmented military control, the influence of the police force was only partial in most areas of the Eastern Province.

The demarcation between LTTE civilian police and the LTTE ‘military’ was blurred. As ‘Inspector General of Police of Tamil Eelam’ Balasingham Mahendran (alias Nadesan) said in an interview with the Sunday Times in 2002: ‘If there is an offensive military operation, our men and women [the Tamil Eelam Police] take part in it.’⁵⁷ One of the respondents of this study, similarly points out: ‘If needed, they [the police officers] would have to go to the Vanni to join the battle.’⁵⁸ The respondent further explains that usually ‘these policemen would be ex-cadres. (. . .) After some incident they would be referred to join the police force.’⁵⁹ Hence, cadres would become policemen and sometimes policemen would again become cadres involved in combat in the North.

At the end of the 1980s the LTTE did not yet control large territories and did not possess the capacity to set up a well-functioning, parallel court system.⁶⁰ Throughout the 1990s, however, alongside the expansion of the Tamil Eelam police, the LTTE began establishing a system of courts.⁶¹ As to why the LTTE did this, the chief of the legal and administration division at the time, Illayathamby Pararajasingham, highlighted several reasons during an interview.⁶² First, he mentioned that the system should help protect ‘the poor’:

Our leader felt that the legal system in the country was not helping the poor. Therefore he decided that the ‘Tamil Eelam’ areas should have a separate courts and legal system which could serve the poor. From the early 1990s we have been developing the legal system. We introduced the ‘law college’ in 1992 with the courses first being opened only to the LTTE armed cadres who had passed the Advanced Level examination.⁶³

Though this notion of ‘serving the poor’ may have been inserted for public consumption, it reveals nevertheless a desire to secure a level of legitimacy from below, or at least the perception thereof. Secondly, he stressed that another main goal was to run a smooth civil administration: ‘it is not to challenge the system in the South, but to run a smooth civil administration in the North. We should have a system suitable for the people.’⁶⁴ Thirdly, he mentioned: ‘the people in the North and East have lost faith in the legal system of the country. Therefore this system should continue.’⁶⁵ At its most developed moment in time, the Tamil Eelam system consisted of a hierarchical court structure, consisting of six District courts, two High courts, an Appeals Court, and a Special Bench (similar to a Supreme Court). As one of the respondents from Trincomalee town says:

There were courts, these were 100 per cent clean, not corrupt. They were 100 per cent operational and enforced by the LTTE law enforcement. (. . .) They studied law and they had their own law. There was in the North also a law college.⁶⁶

The track record of the LTTE judiciary is, however, mixed, particularly in the areas outside of the Vanni. Overall, respondents in this study state that the system of the LTTE judiciary was ‘suitable to the people’ or the ‘judicial unit of the people’,⁶⁷ but a number of respondents, particularly from the East, point out that judgments were far from impartial and that the system functioned poorly. As one of them puts it: ‘impartiality and justice was not seen in these [LTTE] inquiries. Whatever they said was the verdict. They forced people to accept it.’⁶⁸ In those areas where the LTTE only had a partial influence it would not have been able to set up functioning police forces or courts. In these areas it was either the government's justice system fulfilling those tasks or the military on either side of the conflict (the Sri Lankan military or the LTTE commanders/cadres) through more ad hoc forms of ‘justice’.⁶⁹ As one of the respondents explains: ‘the government's judiciary system was functioning, but the LTTE also called people to their territory to investigate. (. . .) Investigations and punishments were a speedy process and the hearing was limited, therefore there was no justice done to the people.’⁷⁰

Generally, the LTTE police and the judicial units were known for their harsh punishments, physically and also psychologically: the upholding of law and order is reported to have been very intrusive, both in areas under their control and in areas under the control of the government.⁷¹ Similar to the fear that civilians expressed towards the regular LTTE cadres, the LTTE police also ruled through fear. To illustrate, a fisherman from Pudukuduyirippu elaborates as follows: ‘the LTTE established regional police departments to maintain their law. The people adhered to the law of the LTTE out of fear.’⁷² A respondent who used to live in the Vanni explained how the law and order of the LTTE police was accepted and appreciated, but frustrating at the same time:

The people were okay with it. They kind of accepted it. Well I mean they adapted, you know. They had to follow the rules! I was not happy, not even me. We can't move like machines. Some punishments were very high! The boys said: ‘we could not see that, could not do that, etc.’⁷³

Several respondents refer to how ‘the one person per family rule’ in particular instilled fear among the Tamil population in the Vanni. This rule forced every family in the Vanni to contribute at least one cadre to the LTTE fighting force. Some people among our respondents did this willingly, but several expressed their distress. A respondent from Mullaitivu explains:

People were afraid of the LTTE as they sometimes recruited the youth to the movement and they had a rule that each family should provide one member to them. They also enrolled young boys. They caught the children forcibly. We were distressed to see this. We were also afraid. We were apprehensive that our children too would be placed in the same situation.⁷⁴

There are also several accounts of an ‘imagined’ LTTE influence in the government-controlled territories.⁷⁵ This does not imply that the influence was not ‘real’, rather it exemplifies the power of imagination and the fear of possible LTTE punishments. Bremner, who conducted fieldwork in 2005 in a ‘cleared’ village in the Northeast, puts it as follows:

The LTTE played a role in the day-to-day lives of the people even though this was a government-controlled area. The LTTE state was imagined through its court system with its feared and swift system of violence and punishment. This imagination was created through rumour, which circulated around, and about the impersonal moral justice, discipline and violence of the LTTE quasi state within the enclaves of its court system.⁷⁶

According to Bremner, several stories circulated about LTTE punishments.⁷⁷ For example, stories about drunken men who were made to empty tubs full of water with little soda bottletops, stories about severe beatings, and stories about the rough physical work one had to do in order to be ‘rehabilitated’. Similarly, a respondent of this study explains:

A person accused of drinking illicit arrack was asked to pay 365 LKR as a fine. But he could pay only one rupee a day at a particular LTTE camp far away from his house. Due to such punishments some men totally gave up drinking arrack.⁷⁸

In relation to the LTTE environmental police another respondent mentions:

Or they would tell you, if you did something against the environment, (. . .) that 60 KM away you have a coconut tree that you have to take care of. That is now your tree, so every day you would have to travel there and water the coconut tree.⁷⁹

The same respondent elaborates on the LTTE rules that had to be followed:

For example no prostitution, no homosexuals, it was not allowed to cheat on your man or wife. Movies were not allowed. No sexy movies you know, from India, they were not allowed. (. . .) You know like sexy movies with women sexy dressed. (. . .) That also, but just regular movies, they were not allowed because women would not be appropriately covered, clothes too sexy, things like that. So people would be punished if you would watch that. (. . .) Also there were dress codes; you know people were not allowed to wear sexy clothes. (. . .) Men also, but particularly for women, they could not wear like short skirts or things like that. It would have to be covered, long. If not, they would warn you, and you would get punished.⁸⁰

A common perception among Tamils in Jaffna also seems to have been that it was completely safe for women to go out on their own at any time of the day in the LTTE-controlled Vanni, whereas in Jaffna town, which was under military control post-1995, women would not go out after seven o'clock at night. There, women, including visitors from Western countries, were allegedly subject to sexual harassment and physical attacks.⁸¹

Hence, the observation can be made that the LTTE was able to exercise de facto sovereignty in the areas under its military control. It indeed possessed the ‘ability to kill, punish, and discipline with impunity’, which it also attempted to institutionalize into a state-like system of police departments and a Tamil Eelam court system. Additionally, in this regard Thiranagama highlights that groups like the LTTE continuously attempt to ‘institute themselves as sovereign through the organization of life and death’, a point we will return to later on when discussing LTTE martyrdom.⁸² Moving beyond the observation that in the territories under its control the LTTE had the coercive ability to discipline the population, it also attempted to legitimize its rule by what it deemed to be a righteous order, which included, for example, ‘decent’ and disciplined behaviour among those who were being ‘ruled’. The LTTE thus tried to foster compliance with the norms it set out.

Service delivery

Apart from securing law and order, the LTTE performed sovereignty through the provision of other services and, more generally, through its ability to carry out various functions usually identified with modern statehood. As elaborated by Rotberg, such functions can amount to a long list of components, but we shall focus here on the two that are deemed essential in Sri Lanka, with its fairly long history of state welfare provision: service delivery in the health-care and educational sectors.⁸³

Health care

As Mampilly indicates, for a long time the LTTE ‘capital’ of Kilinochchi had a large government-run hospital staffed by a crew of doctors and nurses paid for by the government, but under the rule of the LTTE.⁸⁴ The data of this study also indicate that the medical personnel working in uncleared areas received a salary from the government, but that the hospitals were administered and ruled by the LTTE leadership. As one of the respondents puts it: ‘They [the LTTE] were controlling most of the basic public services. But the health care, so the medical staff, was paid by the government.’⁸⁵ So despite the fact that the Sri Lankan Ministry of Health was officially providing health-care services through government hospitals, the LTTE made the final decisions about implementation.⁸⁶ Hence, there was a complex interaction through which the government and the LTTE were both attempting to influence the health-care sector. As one Tamil respondent from Jaffna puts it: ‘You know, we were fighting against the government army, and if you would get wounded, you would go to a government hospital. Food and medicine for the population were also coming from the government.’⁸⁷ In the next section—‘Hybrid orders of rule and authority’—we will address this complex phenomenon in more detail and across different sectors.

Apart from the government hospitals in the uncleared areas, the LTTE itself was also directly involved in health care through the provision of mobile medical units.⁸⁸ With regard to these, a medical doctor from Jaffna explains:

In the name of LTTE commander Thileepan they [the LTTE] started a mobile medical unit. So with the mobile medical unit the LTTE was able to reach the people that were at the time all scattered around the Vanni. The LTTE also developed a medical college for the cadres and for the regular civilian people. These LTTE-trained doctors were able to do operations with very limited supplies. I heard from government-trained doctors that they were impressed by how these doctors did the operations with so little supplies. The LTTE-trained doctors were particularly good at that. They were able to do complicated operations with limited technology, that's what they are known for. So the LTTE was providing the health care where they could.⁸⁹

Providing necessary health care to the civilian populations in the Vanni and the LTTE-controlled areas of the East was particularly difficult due to the checkpoints and the strict embargoes on medical supplies.⁹⁰ Particularly in the last years before the defeat of the LTTE (2006–2009) the embargoes became stricter and the shortages, worse. The doctor from Jaffna explains:

Up until the very final battle the government allowed very little medical supplies into the LTTE-controlled areas. So everything was too little in that final phase. I know that because one of my friends was there until the end as a doctor (. . .). They had to do everything with their own clothes as they did not even have simple supplies like bandages to use for the people. It was a terrible time, they had from everything too little. But even up until the day of the final defeat they were able to help people to give birth, they were able to do medical operations and surgeries. My friend told me that one time he had to do an amputation of a leg without sedatives, simply because it was the only way to save this person's life.⁹¹

The uncleared areas were also provided with medical support by NGOs and humanitarian organizations.⁹² The International Committee of the Red Cross (ICRC), for example, was allowed by the government to enter the uncleared areas, but they would have to register what supplies they were taking inside.⁹³

In terms of personnel, there was an enduring shortage throughout the different phases of the war.⁹⁴ Working in these areas obviously entailed security risks, which understandably created hesitation among the Tamil doctors and nurses about whether or not to go and work there.⁹⁵ Different respondents moreover point out that there were known cases when the LTTE coerced medical personnel into helping it. One respondent mentions: ‘The LTTE kidnapped many doctors and released them after fulfilling their medical needs (. . .)’, and ‘medicines were stolen by the LTTE’.⁹⁶ As a result ‘civilians suffered a lot’ and ‘the doctors were scared to work here’.⁹⁷ Because of its reliance upon highly skilled personnel and the timely provision of supplies, the health sector in the uncleared areas was undermined to a much greater degree than other sectors like education.⁹⁸

In the East the situation was somewhat different from the North. In the areas investigated in Trincomalee District, for example, nearly all respondents accessed their health-care services from the government hospitals, which continued to be led and administered by the government. Only respondents from Sampur mention that they received health care from LTTE medical teams, and one referred to an LTTE hospital in Mutur.⁹⁹ In places like Kuchaveli and Kinniya respondents state that the LTTE had some influence over the medical facilities.¹⁰⁰ Apart from that, Trincomalee and Batticaloa District were mainly provided for by the government and (I)NGOs.

Education

The educational sector was also affected by the raging war. A former member of the ICRC who worked in the Trincomalee District and the Vanni for several years says:

(. . .) education was disrupted in these areas because many of the warring parties would occupy school buildings for their own purposes. Due to that there are now still a lot of ‘slow-learners’. A lot of these children would not be able to go to school because the building was occupied or it was too insecure to travel there.¹⁰¹

Despite the difficult circumstances, the educational sector in the uncleared areas functioned relatively well during the various phases of the war. For example, a local NGO worker points out:

(. . .) education was OK in the uncleared areas. People tried to read and educate themselves [. . .] that was one of the few things that went on very well. Also the LTTE would allow us to do education projects.¹⁰²

Similarly, Mampilly states that the education system in the uncleared areas was remarkable in terms of its ability to provide a continuity of schooling, given the various disruptions the war imposed on daily life.¹⁰³ At those times when the regular school system was interrupted by a lack of teachers or materials, members of the so-called Tamil Eelam Education Council¹⁰⁴ were able to supplement the government's education system in some areas of the North.¹⁰⁵ This provided schools with a sort of ‘response mechanism’ to deal with the disruptions caused by the war. As a respondent of this study points out, in the Vanni the LTTE ‘had education centres. They had their own university for medicine, for computer things, electronics, and the mechanical field.’¹⁰⁶ Based on his life in the Vanni in the early 1990s he explains further:

People tried to get educated. They study, because they have a good drive. They, for example, tried to be a doctor. They studied well in this period, even though the war raged on. When I was there [in the Vanni], I studied and I passed all my exams, because there you study.¹⁰⁷

A respondent from Jaffna states:

Sometimes people think that the LTTE didn't allow people to get educated, but that is not true! I myself sat in my medical exams at the time in Jaffna. I could do the government or the parallel LTTE exam. That was all there in the early 1990s.¹⁰⁸

The educational sector was influenced by both the government and the LTTE. As one respondent puts it: ‘In Kilinochchi and the uncleared areas in the East (. . .) school principals were appointed by the government, but controlled by the LTTE.’¹⁰⁹ In this complex interface, school principals and teachers had to follow LTTE rules in daily life, while officially working for the government. A respondent from Jaffna says: ‘(. . .) my wife, she was working as a teacher, so a government job, so she got paid by the government, but we were living under the instructions from the LTTE. So that was the special situation.’¹¹⁰ Given that the educational sector was less reliant on the immediate supply of goods (unlike the health-care sector), our respondents point out that it was better equipped to deal with the disruptions of the war.

Having discussed various governance sectors in which the LTTE was active, we may conclude that in some sectors the government was not allowed to function in the uncleared areas, in particular the police and justice sectors. These were taken over completely by the LTTE and functioned independently of any external control or influence by the Sri Lankan government. The government did, however, provide governance services in other sectors such as health care and education. Apart from the fact that the LTTE aimed to impose law and order in the areas under its control, we may also assume that it provided governance services in order to exercise that control, to serve the civilian population, and to engender collaboration. Through these governance provisions it was able to cement its de facto sovereignty and public authority throughout the uncleared areas, particularly in the Vanni, and normalize the situation there.

Hybrid orders of rule and authority

As several authors have highlighted, empirical examples of de facto sovereignty and public authority by non-state actors often show overlapping ‘networks’ with the state, other non-state actors, and international actors, coexisting in the same territorial and institutional space.¹¹¹ As Suykens describes in his study on Naxalite and state governance in the Telangana state of India, regimes of rebels and the government build on each other and do not necessarily contradict one another in particular dimensions of governance. He analyses these governance dynamics in Telangana as diffuse authority whereby both sides of the conflict benefit from the shared influence on a commodity chain in the local economy.¹¹² In the grey and uncleared areas of Sri Lanka one could also observe overlapping networks of governance between the LTTE and the Sri Lankan government. Stokke refers to this as a de facto dual-state structure whereby the LTTE exercised influence on state institutions in government-controlled territory and local governments continued to function in LTTE territories.¹¹³

It is challenging to understand precisely the complex, overlapping networks that existed during the different phases of the war in Sri Lanka. As explained earlier, the police and judiciary functions in the uncleared areas were under the complete control of the LTTE and were fully and independently carried out by it. The other public services were largely provided and/or paid for by the government, while simultaneously being regulated and/or controlled by the LTTE. As Mampilly notes, the civilian population of Sri Lanka had grown accustomed to the continuous provision of public services after independence¹¹⁴ and it was in the interests of the LTTE to continue providing these services during its rule.¹¹⁵ In order to do so, it had to work with the pre-existing institutions of the Sri Lankan state, especially as it would have been difficult for the LTTE to foot the bill associated with the provision of these services. Hence, government hospitals and schools were incorporated into the LTTE's administration. As Mampilly points out, insurgent leaders approached their counterparts in the government after the IPKF's withdrawal in 1990 to ask them if they would continue their provision of services in the Northeast.¹¹⁶ Hence, the distinction between contestation and cooperation by the LTTE and the government is not always as clear-cut as it may seem. Although they were fighting a war, there was also a kind of coexistence in the provision of public goods, partly reminding us of the notion of ‘cooperative conflict’ coined by David Keen.¹¹⁷

Despite the LTTE's territorial control and the enduring warfare between the parties, our study indicates that the Sri Lankan government deliberately—and strategically—ensured some welfare provision to the civilian population in the uncleared areas of the Northeast. As one of the respondents in this study puts it:

From lower levels to higher levels in the administration the executive officers in the uncleared areas were paid for by the government. It led to a very unique situation. The reasoning from the government was clear; they wanted to keep their connection to the civil populations in the Vanni. This was their connection, and their way of showing that the Sri Lankan state was still functioning in that area, for its citizens.¹¹⁸

Several respondents explain the rationale of the government in a similar fashion, pointing to the fact that it wanted to keep in contact with the civilian population in the Northeast. By continuing to be involved in the provision of public goods, the government was showing that, despite the presence and influence of the LTTE, it was still able to provide public services. Hence, the government saw this as a way to maintain both its claim to sovereignty and the integrity of the country as a whole. As one of the respondents explains, a lack of service provision by the government ‘would have strategically played into the hands of the LTTE with their claim on the establishment of a separate state’.¹¹⁹ The LTTE managed to build a sort of state-within-the-state, but one that was not internationally recognized; this could have changed if the government had handed over all service provision to the insurgency.¹²⁰ Or as Mampilly puts it: ‘the government preferred to negotiate directly with the rebel leaders about service provision because they feared that the insurgents might set up a comprehensive parallel administration as a testament to their secessionist credentials’.¹²¹

For the LTTE, the complex interface of mutual dependency also served its interests as these arrangements enabled it to keep the Tamil population provided with basic public services, without having to put too much of its own resources into them.¹²² In terms of public service provision to the population, the LTTE depended partly on the government.¹²³ For the LTTE, for the time being, it seemed sufficient to monopolize the security and justice sectors, and have a significant influence on other forms of service provision facilitated through existing state structures. Anton Balasingham, LTTE's key theorist, gave a similar explanation in an interview in 2002:

Don't forget that government institutions are still functioning in areas controlled by the LTTE. We do not interfere with those. We have only taken over the enforcement of law since our armed cadres are confined to barracks. And there we are expanding civil administration. Some day you have to accept a Tamil regional police force and we have to discuss how it would harmonise with the national system.¹²⁴

Here Balasingham does not express the desire to establish a complete, parallel system of service provision to circumnavigate the existing state structures. Rather his aim seems to have been to further incorporate those structures into the LTTE rule.

Despite official condemnation of the LTTE and its appearance on terrorist listings, its governance structures also existed in collaboration with international actors, such as humanitarian organizations and NGOs, with a spike in interaction and resources in the aftermath of the 2004 tsunami.¹²⁵ These interactions were partly mediated through organizations such as the Tamils Rehabilitation Organization (TRO).¹²⁶ Both foreign and national aid and development workers coordinated their projects with LTTE officials, not only in the Vanni, but also in the uncleared areas in the East.¹²⁷ In the cleared areas, foreign and national NGOs played a supplementary role in supplying those basic services that the government was unable to provide; in the uncleared areas, the work of NGOs was a lot more difficult. They always required the go-ahead from the LTTE, and at the same time needed permission from the government to bring material resources into uncleared territories.

A pertinent question here is how to conceptualize the LTTE's governance efforts in relation to the state. In approaches that focus on anti-state sovereignty, rebel groups are sometimes perceived as ineffective and as having transplanted ‘state’ institutions that cannot fulfil the basic requirements of a sovereign system. However, this perspective may not be relevant to the case of Sri Lanka in several ways. The LTTE did not fully reject or replace the state institutions in areas under its control, but only took over the most strategic ones, while at the same time assuring the continuation of other services by the state under its own regulations. Scholars have discussed and documented various interactions, partnerships, and alliances between non-state (armed) actors and state institutions, and conceptualized these phenomena using various models.¹²⁸ In the case of Sri Lanka it makes sense to speak of ‘hybrid political orders’ as described by Boege et al.:

In hybrid political orders, diverse and competing authority structures, sets of rules, logics of order, and claims to power co-exist, overlap, interact, and intertwine (. . .). In this environment, the “state” has no privileged monopolistic position as the only agency providing security, welfare, and representation; it has to share authority, legitimacy, and capacity with other institutions.¹²⁹

If we look at the situation in the uncleared areas of the Northeast post-1990, we could argue that the ‘diverse and competing authority structures, sets of rules, logics of order, and claims to power’ in these areas indeed ‘co-exist[ed], overlap[ped] . . . and intertwine[d]’. As one of the respondents puts it:

The bureaucratic and organizational system was already there. I mean the government system. This however led to a very unique situation in which these offices were paid by the government, but they for example were not allowed to put up a Sri Lankan flag. The LTTE only allowed them to put up the LTTE flag!¹³⁰

This excerpt indicates that the LTTE was able to exert its power to such an extent that it could force these ‘government offices’ to fly the LTTE flag, while the government attempted to show its authority by formally appointing and paying officials prepared to work in these areas. The data for this study indicate that all local government offices in the Vanni, such as the Divisional Secretariat, the Department of Agriculture, and the Department of Education functioned according to the instructions of the LTTE.¹³¹ Although the people received government services, the LTTE were the ones that took the real decisions related to the functioning of the officials. And even though the government servants ran the government offices, the LTTE directed and monitored them.¹³²

Also outside the LTTE-controlled territories in the Northeast, the concept of hybrid political orders may apply. According to the respondents of our study, the LTTE was also able to exert its influence in the local governance of the grey, and even the cleared, areas throughout the Northeast.¹³³ As one of the respondents in a former grey area in Trincomalee puts it: ‘The government offices functioned during the war, but the LTTE had an indirect influence.’¹³⁴ Gerharz similarly explains that post-1995 Jaffna was under the military control of the government, and law enforcement was the purview of the Sri Lankan security forces.¹³⁵ At the time, however, a common perception among the people was that the LTTE was, in fact, a more forceful and effective institution than the government in enforcing law and order.¹³⁶ Some people also thought their injustices would be more effectively redressed through the LTTE than the armed forces. Jaffna town was a typical example of two overlapping and intertwined systems of governance.¹³⁷ As Klem explains, in towns like Jaffna, Trincomalee, and Batticaloa, which were formally and militarily under control of the government, the LTTE was able to exert its power through its invisible presence in people's everyday lives, and through its influence on state bureaucracies.¹³⁸

This echoes Sarah Byrne's notion of the ‘absent presence’ of the Nepalese state in the lives of its citizens.¹³⁹ However, in her view, this is part of a tactical and responsive government practice, while we would argue that in the case of the LTTE their influence may have been of a more structural nature. It was based on an extensive system of surveillance by what people called ‘spies’ (among other strategies). Several respondents in our study stated that ‘they [the LTTE] knew everything, even if you could not see them’.¹⁴⁰ This same idea of rather firm control combined with limited visibility is also confirmed in Klem and Maunaguru's description of the LTTE's role in running the rural development society (RDS) in Adivasipuram.¹⁴¹ This case shows the often-complicated forms of entanglement between the state and the LTTE, as the RDS, while largely under LTTE control, maintained relations with the larger state apparatus and derived resources from that link. Klem and Maunaguru conclude that ‘The institutional boundaries between the spheres of government and the LTTE were thus rather more blurred than a simple categorization of sovereign state versus insurgency would suggest’, underlining again the hybridity of governance and authority.¹⁴²

Symbolic legitimation: narratives, performances, and inscriptions

According to Barker, legitimacy forms ‘the master question of politics’.¹⁴³ Barker defines legitimation (i.e. the attempt to acquire legitimacy) as ‘an action or series of actions—speech, writing, ritual, display—whereby people justify to themselves or others the actions they are taking and the identities they are expressing or claiming’.¹⁴⁴ A number of authors point out that legitimation should be seen as a subjective, changing, and context-dependent process.¹⁴⁵ In this political process, symbols are a means of influence and control, whereby the interpretation of a symbol is not intrinsic to it, but collectively created, maintained, and changed through time and place.¹⁴⁶ As Stone explains, a symbol can be seen as ‘anything that stands for something else’.¹⁴⁷ Symbols can be words, songs, pictures, logos, or events, for example. These can symbolize a set of ideas, such as a political party or political movement.¹⁴⁸ As one of our respondents puts it: ‘If you would find a photograph of our leader Prabhakaran or a Tiger flag in a house, you would know that he or she is a Tiger supporter. These were important to get a separate identity.’¹⁴⁹ Symbols of the LTTE as a rebel movement, and of the struggle for an independent state with a distinct national identity, continued to evolve over time. This effort was projected towards internal audiences such as the Tamil population living in the Vanni, but also to international audiences and the diaspora, as part of the LTTE's pursuit of international recognition for its proposed Tamil Eelam. Following Schröder and Schmidt we will discuss below the main narratives, performances, and inscriptions that can be identified in the legitimation processes of LTTE rule.

Narratives

The narrative of the liberation of the Tamil motherland arguably resonated among much of the Tamil community. The lead narrative consisted of Tamil nationalism, resistance against oppression of the Tamil minority by the Sri Lankan state, the existence of a historical homeland of the Tamils in Sri Lanka, and the demand for Tamil Eelam, based on the right to self-determination.¹⁵⁰ Though these elements all pre-date the birth of the LTTE, they were reproduced, magnified, and extended in the LTTE's representation of its struggle. Its political programme showed a peculiar mix of historical, nationalist, socialist, secular, and transformative storylines.¹⁵¹ It not only fought for an independent Eelam, but also wanted to fundamentally change the conservative, traditional nature of Tamil society by a process of socialist transformation:

The struggle for self-determination of the Eelam Tamils has an evolutionary history extending to 40 years. It was a historical struggle characterised by state repression and resistance by the Tamils. The Tamil freedom movement was peaceful and non-violent at the early stages and later developed and advanced into an armed revolutionary struggle as state repression intensified and assumed the character of genocide. (. . .) The only alternative left to the Tamil nation under the conditions of mounting national oppression (. . .) was none other than popular armed resistance directed towards the goals of national liberation and socialist social transformation.¹⁵²

As one of our respondents puts it:

The small struggle transformed into a big one. Then it got the history of a national struggle. The most important entity of a race is a national flag and a national anthem. It was found here. People liked it a lot. The people used it in public events with honour.¹⁵³

Our studies show that to this very day various elements in the LTTE's ‘master narrative’ are present and adhered to, even though the armed struggle is over. Although this narrative resonates more deeply in the North than in the East, and less so among Tamil-speaking Muslims, many people still refer to the widespread support the LTTE and their political goals enjoyed. Though they sometimes feared the movement, they feared and hated the Sri Lankan state and army far more. Because of the recognition of their rights as equal citizens and their sense of belonging to the Tamil minority community, they opted to support the LTTE rather than the central government. Or as one of the respondents in Jaffna states: ‘there were several negative things, but we respected that they were fighting for us, we respected that they were sacrificing their lives for us. Not everything was right, but we accepted that.’¹⁵⁴ Following this narrative the LTTE performed state-like functions, which will be discussed below.

Performances of statehood

From an insurgent's perspective, the performance of statehood can serve to portray authority and to cement claims to legitimacy. As Mampilly points out: ‘In essence, deploying a symbolic repertoire is an attempt by a rebel government to performatively legitimate its sovereign claim.’¹⁵⁵ In several South Asian cases where insurgents govern their strongholds, symbols function to validate rebel rule. Sundar, for example, shows how memorials, flags, and commemoration days serve as key ingredients to the sovereignty practised by Maoist insurgents in central India.¹⁵⁶ In this connection, she posits the existence of ‘mimetic sovereignties’, where both state and rebels start to imitate and resemble one another in their performance of sovereignty: ‘the Indian state impersonates guerrilla tactics in order to fight the Maoists, while the Maoists mimic state practices of governmentality’.¹⁵⁷ In other words, what we may observe is a type of mimicry of the state by insurgent groups to legitimate their rebel rule.

In the case of the LTTE, Prabhakaran created a logo, a central committee, and a constitution for the LTTE in 1976. He reportedly also personally designed the Tiger uniforms. The public space in Jaffna (before 1995) and in the Vanni was dominated by the LTTE through symbols such as posters, flags, and monuments.¹⁵⁸ After the LTTE's move from Jaffna in 1995, the newly established LTTE capital Kilinochchi became an illustrious place of state-like symbolism. Buildings with flags and signposts to the various Tamil Eelam institutions, for example to indicate the police offices and the LTTE courts, covered the outline of the city.¹⁵⁹ The Tamil Eelam police force used its own salutes, and a national anthem was sung. There was also a national bird, a national tree, and a national flower to represent the separate nation.¹⁶⁰ The LTTE further used symbolic measures such as national days and hymns to support its cause.¹⁶¹ Based on the responses in this study, the perception of the Tamil communities regarding the LTTE was indeed that it was similar to a state. In both the North and in the East a majority of the respondents believe that the LTTE had ‘everything a government should have’. Moreover, as one of our respondents from Kinniya puts it, there was ‘absolute royal respect for the national Tiger flag and the National flower, the November flower. We also saluted and worshipped these. We willingly did so by ourselves.’¹⁶²

The LTTE also clearly invested in projecting its legitimacy among particular audiences in the international community. Foreign researchers, journalists, aid workers, and diplomats were invited into rebel-controlled territory to see the LTTE's organizational capacities.¹⁶³ During this ‘tour’¹⁶⁴ the foreigners would usually pass the LTTE checkpoint at Omanthai, which was constructed like a national border, with flags, signs, and armed officers who controlled the vehicles. There would be customs staff in LTTE uniforms to check passports and formal procedures of tax collection would be carried out on vehicles transporting commodities. In the LTTE capital of Kilinochchi the foreign observers were able to see the different offices and institutions that the LTTE had established. Their visit would be very well organized and there was a special guesthouse where these LTTE guests would be hosted. Various films, other media, and documentation would be shown to present the sophistication of the institution-building efforts of the LTTE. For example, it distributed an organizational chart to one of the authors in 2004 to show the multiple departments within the LTTE's Peace Secretariat and Political Wing.

At the international level the LTTE has been at the centre of a number of paradoxes in prevailing international norms.¹⁶⁵ On the one hand, there existed hostile international attitudes towards armed struggles against internationally recognized states. Insurgencies like the LTTE might thus become incorporated within a terrorist framing.¹⁶⁶ As such, the LTTE was banned under anti-terrorism legislation by the USA (in 1997), the UK (in 2001), and India (in 1991). Countries like Australia and Canada prohibited the LTTE's fundraising activities in 2002.¹⁶⁷ At the same time, however, countries like the USA and the UK also actively promoted negotiations between the LTTE and the Sri Lankan government.¹⁶⁸ By doing so they implicitly recognized the LTTE as a legitimate political actor with which to negotiate. As part of this effort, several LTTE delegations visited Western countries, including the Netherlands, France, Spain, Switzerland, and South Africa, to examine constitutional models and governance arrangements.¹⁶⁹ Particularly after the CFA in 2002 several Western countries established varying forms of contact with the LTTE's leadership.¹⁷⁰ This, taken together with the various terrorist listings, made relations between the LTTE and these countries both ambiguous and paradoxical.

Performances of heroism and martyrdom

In terms of performance, Schlichte and Schneckener state that the ‘respect and credibility that leaders and fighters might earn for their readiness to sacrifice their lives for a common cause might eventually lead to latent forms of legitimacy, at least within the targeted constituency’.¹⁷¹ As elaborated by Hellmann-Rajanayagam, the death of a martyr may function to reaffirm the cohesion of a particular group, to legitimize its convictions, and to strengthen its self-respect.¹⁷² Suykens shows in his article on the Naxalites in India that the martyrs of the insurgency help followers to imagine the future and to shape both their prospects and those of the movement.¹⁷³

A number of authors consider the sacrifices made by its cadres as an important element in the LTTE's symbolic portrayal of its struggle.¹⁷⁴ As Roberts explains: ‘martyrdom was a critical factor in drawing popular support among the Sri Lankan Tamil people. [The. . .] devotional commitment indexed by the suicidal act was evaluated highly (. . .) by many a Tamil person.’¹⁷⁵ Bavinck notes that a kind of ‘mourning industry’ emerged in the North during the 1990s.¹⁷⁶ There were countless commemorations of fallen cadres where these heroes or martyrs were accorded the status of ‘sainthood’.¹⁷⁷ Or, as Trawick identifies, regarding the interpretation of death: ‘a person [LTTE cadre] who kills is just doing his job. A person [LTTE cadre] who dies is a hero.’¹⁷⁸

The LTTE was indeed keen on commemorations of fallen cadres. As one of our respondents points out: ‘The fallen cadres were commemorated every year, and we would also go there every year to commemorate them.’¹⁷⁹ The respondents of this study from the Vanni and in the uncleared areas of the East all maintain that most of the population attended these kinds of ceremonies. As one says: ‘People went and participated in these ceremonies and they willingly did it. They wanted to pay their respects’,¹⁸⁰ and another respondent mentions with regard to these commemorations: ‘They [the civilian Tamil population] considered it as their tradition and as a part of their duties towards their motherland.’¹⁸¹

On the surface these commemorations only express sentiments of mourning and remembrance of the fallen. More essentially, however, these ideas help to construct and maintain a political community, which can be seen as a form of nation-building.¹⁸² As one of our respondents puts it:

During public events, tiger flags were raised and symbols were worshipped. People were proud of this. All of us should salute the national flag of our soil. It is our duty and responsibility. Isn't it one of our special duties to perform poojas to the pictures and memorials of our heroes who sacrificed their lives for our race? People got involved in this activity with a lot of interest and enthusiasm.¹⁸³

Other respondents mention how these events were related to ‘our’ leader, ‘our’ soil, and the sacrifices made, and that they should be respected and supported wholeheartedly:

Our flag, the tiger flag and the symbols of every division are worshipped. That is our special quality. That is our debt of gratitude to our leader. We should have done a lot of virtuous deeds for getting on our soil such a brave Tamil son who has sacrificed his life for us.¹⁸⁴

The sense of community was also emphasized by several respondents with regard to the achievements of the LTTE ‘heroes’:

All the heroes are our brothers. So, the people and the Liberation tigers together celebrated it impressively displaying pictures and lighting lamps in the houses. The skills, abilities, and achievements of the heroes were revealed on the heroes’ day. It is a day to be proud of.¹⁸⁵

The respect of the LTTE and the civilian population for the ‘fallen heroes’ extended beyond commemorating them into supporting their remaining family members. As one of the respondents explains:

The LTTE gave much respect to the fallen heroes. They even had a separate department to take care of their family members. (. . .) It was called the ‘Heroes welfare society’. The families would get help from the LTTE, in terms of medical or financial support.¹⁸⁶

In other words, not only the LTTE cadres themselves, but also their family members continued to play a symbolic role in the larger struggle of the LTTE.¹⁸⁷ These families were often referred to as ‘heroes families’ and given particular advantages in its governance system.¹⁸⁸

Inscriptions

In the LTTE's case it needs to be pointed out, finally, that the different forms of legitimation were enhanced through the LTTE's propaganda channels. As one of the respondents in this study says:

The LTTE had their radio channel in Jaffna, Mullaitivu and Kilinochchi. (. . .) They were doing documentaries, films, and short films. They would release good films that would describe what was being achieved. That was the propaganda. These films were also taken on the battlefield. (. . .) They would show it very fiercely you know, they take all the Hindu songs/music, and they would get very vibrant.¹⁸⁹

When asked about the songs that were used one of the respondents states:

Things about the fight like ‘we want our land back’, or ‘we have to fight’. (. . .) There are plenty of songs. It is like, they wanted to boost [their image], and they wanted to get that feeling in the community. They did it very deliberately.¹⁹⁰

There was only limited media available at the time in the Vanni, and the LTTE decided what people were allowed to watch and what was forbidden.¹⁹¹ As one respondent says: ‘There was hardly any access to other media, we had to watch it. But like I said, most people also liked watching it.’¹⁹² According to Brun, many of the propaganda films produced by the LTTE portrayed the soldiers of the Sri Lankan Army as alcohol abusers who danced while the Tamil people suffered.¹⁹³ Most films included a portrayal of the battle sites and the history of the movement, and showed the atrocities against Tamils and the LTTE's achievements in the struggle.¹⁹⁴ Many films cited the Black July pogrom of 1983 as justification for the movement's fight against the Sri Lankan government.¹⁹⁵ Through these propaganda channels the LTTE gained what Demmers calls the ‘power to define’: determining the ‘legitimate’ course of action.¹⁹⁶

The LTTE invested many of its resources into maintaining cemeteries of the fallen, and more generally it made the administration of death central in its governance.¹⁹⁷ As Sangarasivan explains, the ‘laying of bodies [of the LTTE heroes . . .] and the building of tombstones inscribe the presence of the honoured dead into the land [and] their physical substance coalesces with the soil of the land to create a culturally circumscribed sacred space’.¹⁹⁸ In other words, both spatially and physically the LTTE inscribed the death and sacrifice of the heroes for Tamil Eelam into the soil of the Vanni.

Finally, the question arises: how were these narratives, performances, and inscriptions maintained? According to Demmers, some political actors have greater ‘powers to define’ than others.¹⁹⁹ Gramsci coined the notion of ‘hegemonic culture’, in which the values of the dominant classes have become the ‘common-sense’ values of all.²⁰⁰ One could question whether the hegemony of the LTTE's rule in the Vanni in fact imposed compliance on the common man. Particularly in the Vanni the LTTE possessed a hegemonic position in terms of disseminating the daily representations of the war and deploying practices of governance and statehood. The legitimation strategies of the LTTE were effective in the sense that it was able to define the dominant narrative and was able to perform and represent statehood and sacrifice, and that the population participated in it. The population was thus involved in recreating the narratives surrounding the liberation struggle—it participated in the institutions and the commemorations, and presented a variety of inscriptions. In this dynamic process these narratives, performances, and inscriptions were collectively created and maintained.

Conclusion

The article has shown how the LTTE exercised de facto sovereignty and public authority in the areas under their direct or indirect control. Though this implied—in the words of Hansen and Stepputat—‘the ability to kill, punish, and discipline with impunity’,²⁰¹ there was also a bottom-up aspect to it whereby the LTTE attempted to acquire popular consent and compliance through mechanisms other than coercion. This constellation around the LTTE's de facto sovereignty and public authority was the first central argument of our article.

The LTTE's governance varied per sector. Whereas it monopolized justice and policing in the uncleared areas, in other sectors and geographical areas governance services were fulfilled more pragmatically. The LTTE allowed other actors—both state and non-state—to work to fulfil the basic needs of the population as long as this did not interfere with their military strategies and helped boost their legitimacy. Despite the fact that the military and political struggles of the LTTE were distinctively anti-state, concrete governance practices displayed various forms of political hybridity in which the governance practices of the LTTE and the Sri Lankan state overlapped, intertwined, and sometimes collaborated. Our study shows clearly that a focus on governance practices and mechanisms, as evinced in the Foucauldian notion of governmentality, is a better heuristic device than only looking at institutions per se. This has enabled us to discern the multiple interrelationships and manifestations of governance between the LTTE and the Sri Lankan state as well as international non-governmental agencies. This forms the second central argument of this article.

Our study further indicates that performing sovereignty and public authority by the LTTE was not confined to their ability to maintain law and order, and the instrumental delivery of public services, but also required a broader legitimation to assure compliance, if not popular consent. Next to the use of ideology and coercion, this involved several symbolic dimensions in which the rebel group legitimized both its struggle against the enemy and its rule over the population. Following Schröder and Schmidt, we discussed the role of narratives, performances, and inscriptions to understand how this came about in practice. A first element used extensively was a politico-historical narrative on the oppression of the Tamil people by the Sinhalese (state) and the necessity of a liberated Tamil Eelam. It also involved a political programme outlining the transformations envisaged under Eelam statehood. Through state ‘mimicry’ and state-like performance, the LTTE could show the people that it was able to effectively deliver a series of governance services. Another important element in the symbolic dimension of LTTE rule entailed the elevation of the struggle and the heroic status of LTTE cadres fighting against the government—and dying in the fight. Cemeteries, commemorations of the ‘heroes’, flags, and other national symbols were the visible inscriptions of the struggle and the sacrifices made, and were widely respected within the Tamil community. The various symbols were not only physically present in the public space, but were also disseminated by the media and through propaganda. In using symbolism, the LTTE implicitly and explicitly attempted to legitimize its rule over the Tamil population and the nascent statehood of its projected Tamil Eelam. Integral to the LTTE's efforts to exercise sovereignty and public authority were attempts to create, gain, and maintain a level of consent and legitimacy, both internally and externally, alongside the use of coercion, to ensure compliance. This forms the third central argument of our article.

The conclusion of our analysis is that a simple binary of state versus non-state actors is not suitable to grasp the empirical manifestation of LTTE governance. We distinguished a more differentiated picture in which state mimicry and state-like performance, parallel structures, hybrid forms of governance and co-optation, and different forms of legitimation were combined in a multiplex pattern that varied over time and place. We conclude that the LTTE's de facto sovereignty and public authority were based on a number of governance practices that were operated by the movement independently from the Sri Lankan state, while other practices took shape within the pre-existing political order and service provisioning by the state.