Until the end of last month, the peace accord signed between India and Sri Lanka on 29 July appeared, precariously and against the odds, to be holding firm. As I write this article, several incidents of major violence are threatening to destroy an agreement which, for two astonishing months, seemed to promise the country reconstruction and renewal. When, in May, the Sri Lankan Armed Forces launched an assault on the northern Jaffna peninsula, which had for two years been almost entirely controlled by the main Tamil militant organisation, the Liberation Tigers of Tamil Eelam (the LTTE), nobody could have foreseen imminent peace. So bleak and featureless was the political horizon that the BBC’s Colombo correspondent was on holiday when the accord was announced. The surprise, and the relief it brought to those sections of the community most devastated by the war, were remarkable. But Rajiv Gandhi has signed accords before. The Punjab accord was pulled like a rabbit out of a turban with much the same wizardry less than three years ago, and the Punjab is now virtually in civil war. Is the same happening to Sri Lanka?

One source of difficulty in answering the question lies in the fact that Sri Lanka is such an unlikely setting for an ethnically-based insurrectionist movement. Though one of the world’s poorest countries on a per capita basis, its remarkably equal distribution of income and historically high provision of education and welfare services have resulted in a low incidence of rural poverty and therefore one of the lowest rates of urbanisation in the world. Though its democracy has been manipulated in dubious ways, it has proved itself capable of regularly turning out incumbent governments. The Tamil minority community is socially conservative and inhospitable to both religious and political fundamentalism (many Tamils view the vaguely Marxist jargon of some of the guerrilla organisations as little more than a rhetorical badge, a claim to insurrectionist authenticity). Sri Lanka lacks the centuries-long history of ethnic tensions that continues to poison relations in Kampuchea or Northern Ireland: the first serious Sinhalese-Tamil violence took place in the late 1950s. And although the Sinhalese nationalism to which Tamil separatism is a response dates back to the 19th century, it was not until the election of a government determined to implement a ‘Sinhala-only’ language policy in 1956 that this nationalism took a specifically anti-Tamil form. Sociologists and historians continue to search for evidence of a deep social foundation for the ethnic crisis, but although hints and incidents can be found, they remain stubbornly sparse. If ever a country demonstrated the potential of myopic or cynical manipulation at the level of national politics to nurture the seed of ethnic conflict in an unpromising soil, it is Sri Lanka.

Sri Lanka has not one but two distinct communities of Tamils. One group, known as the Indian Tamils and comprising 5.6 per cent of the population, came from Southern India during the 19th century, to work the plantations (mainly of tea) established by the British. Though this group has suffered considerably at the hands of the majority Sinhalese, as a result both of violence and of the disenfranchisement inflicted upon it at Independence, there is little evidence of any militant activity in response. Most Indian Tamils live on the plantations in the hills at the centre of the island, separated physically and socially from the more numerous Sri Lankan Tamils concentrated in the north and east of the country.

The precise origins of this latter group, which comprises one-eighth of the population, are heatedly and irrelevantly discussed by both Sinhalese and Tamil historians, but it is clear that Tamils have been in the country for at least two millennia. The dominant myth of Sinhalese origin stresses inconsistently the importance both of the alleged priority of Sinhalese presence in the country and of the claim that the Sinhalese themselves are not aboriginal but descend from the superior Aryan races who conquered the island from Northern India sometime during prehistory. There is comedy as well as tragedy in these twistings of academic history, in which even the inoffensive discipline of archaeology is used and abused to buttress rival titles to territory. In a country in which significant portions of the population have been resettled on government land as a result of large-scale irrigation schemes, so that the ethnic composition of important areas has been changed, territory is a highly-charged issue.

Muslims comprise a third important ethnic minority, one with deep historic roots in the country. They speak Tamil, but are a quite distinct religious and social community from the Hindu Tamils. Although only 7 per cent of the population, they have considerable importance in the present crisis as a result of their disproportionate concentration in the crucial Eastern Province of the island, where Sinhalese, Tamils and Muslims comprise about a third of the population each. Other ethnic minorities make up less than 1 per cent of the country’s population, but some groups, such as the so-called Burghers (of mixed Asian-European descent), wield significant economic and social influence. None of the country’s minorities was singled out for particular opprobrium by the growth of Sinhalese Buddhist nationalism in the last century, a movement spearheaded by the Buddhist revivalist Dharmapala. If there was hostility it was to the impure outsider as such, a category that took in the Muslim and the Briton as much as the Tamil. The emphasis was rather on the special role accorded to the Sinhalese race as guardians of the purity of Buddhism. Though many Sinhalese children were brought up on stories of battles with barbarian Tamils at the dawn of history, the role these stories played in the popular consciousness was hardly any more sinister than that played by the Wars of the Roses in the demonology of the Yorkshire Cricket Club.

Geography had a part in this: specifically, the thick jungle that separated the Tamil communities of the north and east from the Sinhalese in the south, and the hills that divided both of these from the different Sinhalese communities around Kandy. Prior to the unification of the country under the British, there was little ethnic friction because there was little ethnic contact (even today there are many rural Sinhalese who have never met a Tamil). The Sri Lankan Tamils did well under the British, filling substantial numbers of administrative and other white-collar occupations. This was partly because the early conquest of the Jaffna peninsula by the Portuguese and then the Dutch had meant a greater spread of missionary education here than among the Sinhalese; and partly because agriculture was less rewarding in the drier regions of the north and east. By the time of independence in 1948, some British administrators (including Leonard Woolf) had expressed anxiety lest the political power concentrated in the hands of the majority Sinhalese should lead them to ignore the distinct needs and interests of the minorities: but none foresaw the abruptness with which the balance of educational and administrative advantages was to be redressed in favour of the Sinhalese.

The first government of independent Ceylon, that of the United National Party, was swept from power in 1956 by the Sri Lanka Freedom Party of S.W.R.D. Bandaranaike which promised to enshrine ‘Sinhala-only’ as the national language (in place of the trio of Sinhala, Tamil and English) within 24 hours of election. By requiring the children of each ethnic community to be educated in the medium of their own language it also ensured that the national language, instead of unifying the country, would act to exclude all but the Sinhalese from effective participation in public life. (When I visited Jaffna in 1985 I attended the office of a lawyer friend: he was explaining to a client the procedure for attending the inquest of his brother, who had been killed in what the Army termed ‘crossfire’. The invitation to the inquest was written entirely in Sinhala, a language the man – like most of his fellow Tamils – had had no opportunity of learning.) The idiocy of this policy (as well as the impossibility of delivering all the benefits which Sinhalese voters had been promised consequent upon its implementation) soon came home even to Mr Bandaranaike himself, who later on that year signed an accord – the first of several in this grim tale – with leading Tamil representatives. This promised to reverse the main points of the Sinhala-only policy and to implement safeguards for minority interests. But the United National Party, having once been outflanked for the nationalist vote, was determined upon revenge, and led massive protests against the accord, protests that spilt over in 1958 into the first large-scale violence against Tamils. The accord was torn up, both metaphorically by the Government and literally in the streets by a procession of UNP politicians and saffronrobed Buddhist monks. Since then the acreage of saffron at any political rally has been viewed by both political parties as an indicator of their nationalist credentials. The country’s Buddhist monks combine great authority over the ordinary population with considerable nationalist fervour and indignation at any proposal that smacks of ‘giving in to separatism’ or ‘dividing the homeland of the Sinhalese’. It is a stance made easier by the fact that most monks have never lived anywhere near the Tamil areas and have no conception of conditions within them or indeed (until recently) of the gravity of the ethnic conflict – a state of ignorance for which the Government’s strict control of the press is at least partly to blame.

The thirty years since these events have been punctuated by violent street politics whenever any party in opposition has felt it could exploit apparent government weakness on the ethnic issue. Politics itself has become increasingly violent as politicians have used gangs of hooligans to round up support at election time. Some of these gangs were almost certainly involved, in a co-ordinated way, in the anti-Tamil riots in Colombo in 1983. And successive governments, feeling themselves weak in their control of the Sinhalese majority, have resorted increasingly to repression. This has not been directed only against Tamils: an insurrection in 1971 by the revolutionary youth of the Sinhalese JVP Party was put down at the cost of over ten thousand lives. But governments have certainly used repression against the inconveniences of Tamil protest. The first Tamil demonstration against the 1956 Act, a pious Gandhian gathering on the Galle Face in Colombo, was violently broken up. Until the Seventies Gandhian non-violence was unquestioned as the Tamil community’s response to its perceived grievances. But violence by the security forces and by street mobs produced a counter-reaction, and in the Seventies small groups of Tamil youth began to take to the gun.

Some Tamil grievances might well have been defused by deft political management without the need for substantial concessions. For instance, in the early Seventies the Sri Lanka Freedom Party of Mrs Bandaranaike (the widow of its first prime minister) implemented a system of quotas for admission to higher education which discriminated against Tamils in favour of the relatively educationally-disadvantaged Sinhalese. It is certainly arguable that such affirmative action was not intrinsically unjust (India has implemented a far more draconian system of quotas in favour of disadvantaged castes). But the manner and the context of its implementation, which led to its being seen as a continuation of the programme to reduce minorities to the status of second-class citizens, were without question imprudent. Large numbers of articulate and ambitious Tamil youths, their aspirations to higher education and white-collar employment blocked, grew frustrated with what they saw as the patience and Gandhian sentimentality of their elders, and resolved on less meek forms of protest. At first, this took the form of petty vandalism, but the lazy brutality of the Sri Lankan Police in response provoked counter-violence. The first victims of political assassination by the Tamil militants were other Tamils, killed not for ‘collaboration’ with the majority (against whom, as a community, Tamils stressed that they had no complaint), but in their perceived role as agents of a repressive state.

The descent into violence has been accompanied by a progressive marginalising of the moderate forces in the Tamil community. In the mid-Seventies the Federal Party, which represented Tamils in national politics and was under pressure for its failure to ensure government concessions, resolved to campaign for an independent Tamil state (or ‘Eelam’), and was renamed the Tamil United Liberation Front (TULF). But it remains, as it always was, essentially moderate, middleclass and non-violent. Its Colombo spokesman is even a subscriber to the London Review of Books. But until the accord he was the only TULF MP resident in the country: two of his colleagues had been murdered by Tamil extremists, and the remainder were in exile in India. A constitutional amendment in 1983 had banned from Parliament any MP who would not take an oath of loyalty to the unitary constitution, which advocates of separatism – even democratic separatism – were unable to do. At one stroke the Government had done more to ensure the impotence of the forces of moderate Tamil opinion than the militant extremists had achieved in years. Now that armed violence has apparently shown itself capable of delivering government concessions, the credibility of the TULF in the Tamil areas to which power will be devolved under the accord has suffered a further blow.

Alongside the political deterioration, there has been a vertiginous acceleration of violence by all parties to the conflict. During the Seventies violence against Tamils came either from more or less organised mobs, or from the Police. Day-to-day violence was mainly directed against suspects in custody, especially after the passage in 1979 of the Prevention of Terrorism Act, which has been compared by the International Commission of Jurists to the 1967 Terrorism Act of South Africa, and which provides among its many objectionable features for the detention incommunicado for up to 18 months of any person suspected of terrorist involvement. But the Sri Lankan Army was still a small, largely ceremonial force with no record of violence against civilians before 1981. By 1985 it had become notorious all over the world for its indiscipline and its resort to indiscriminate reprisal killings of Tamil civilians, up to two or three hundred at a time in the worst cases. These killings were often particularly brutal, given the fact that in South Asia death by burning has a special symbolic importance, and that your enemy’s children are the most direct route to revenge against your enemy. Much of this violence was sporadic and unplanned, the reaction of scared and poorly-trained Sinhalese boys of 17 and 18 to the terror of an invisible enemy. But it was compounded by two developments: first, the creation in 1985 of an armed force of ‘home guards’ – a euphemism for handing out guns to Sinhalese villagers who wanted to join the battle; and second, the tendency through 1985 and 1986 for systematic and co-ordinated harassment, often from the air by helicopter gunships, to replace private initiatives by groups of soldiers.

Without question, the particular grievances of Tamils over language, territory, education or constitutional details have been dwarfed by concern for their physical security as the major issue of the crisis. The ranks of the Tamil militants are by now full of men (and some women) who could never have been recruited by slogans about educational quotas or linguistic parity, but whose families have been destroyed and who want revenge. And there has been, more recently but even faster than in the case of the Army, a descent into atrocity on the Tamil side. Until December 1984, the militants’ targets were confined to the security forces. In that month, a group of Sinhalese civilian settlers in an ethnic border area was murdered on the grounds that they had been armed and were allegedly terrorising local Tamils. In May 1985 over a hundred civilians were shot in a surprise attack on the town of Anuradhapura, a major centre of Buddhist pilgrimage. By mid-1986 killings of Sinhalese civilians, including children, had become routine. It was chilling to hear some Tamils of impeccably moderate and democratic credentials refer privately to the regrettable ‘necessity’ of revenge attacks on civilians.

Most recently, the violence has taken internecine form within the Tamil community itself, with battles between the main Liberation Tigers of Tamil Eelam and other militant groups, both before and since the signing of the accord. These battles have more to do with simple power than with any ideological difference. The Tigers themselves have comparatively little ideology: their leader Prabakharan made a reputation in the Seventies for his efficiency as a bank robber and an assassin. (Tamil moderates, though condemning his activities, were often prone to do so less than fortissimo, and with a sneaking admiration for the way ‘the boys’ – a significant term – were capable of outgunning greatly superior numbers of Sinhalese police.) Since everything depended upon this reputation for efficiency, challenges to his authority from rival bands were met uncompromisingly. A battle at the beginning of this year established the dominance of the LTTE in the Jaffna peninsula, but the situation in the eastern part of the, island has never been so clear-cut. There has also been an increase in summary executions of suspected informers: one tragic sign of this, as reported to me in June, was that the parents of young Tamil men in custody, who until recently had been petitioning the security forces for their release, had begun pleading for them to be kept in detention. Early release had come to be interpreted by the LTTE as evidence of informing. The difficulties currently faced by the Indian Army in keeping order in the Eastern Province are due partly to the struggle between the LTTE and its rivals for dominance in the future political administration of the province. And they are due partly also to the simultaneous struggle for such dominance between the Tamil groups as a whole and the Sinhalese and Muslim communities.

The accord provides (besides a restoration of Tamil and English to the status of official languages) for a devolution of power within the country to a system of provincial councils. The powers of these councils are not entirely precise, but would include a significant degree of control over the local Police Force, education, and the ethnic composition of communities settled under irrigation schemes. Such proposals have been discussed before, but a major stumbling-block has been the question whether the Northern and Eastern Provinces should be separate (as now) or merged to form a single unit. Tamils form the overwhelming majority in the Northern Province, but only about one-third of the population in the east. They would therefore dominate a merged province, but in a separate Eastern Province would be in a minority. The proposed solution involves a temporary merger, to be followed by a referendum to decide whether the people of the Eastern Province will remain in the merged unit or form a province on their own. The President has powers to postpone such a referendum, and few people believe the target date of the end of 1988 will be met. The Sinhalese are certain to vote against a continued merger, and the Muslims are likely to do so, given the existence of tensions between them and the Tamils, some of which have been fomented by the security forces. It is therefore highly probable that the requisite majority against a continued merger will be the outcome, but if it appeared that the Tamils had been outvoted en bloc the stability of the accord might be in serious doubt. The Government’s hopes must therefore lie in the possibility of significant Tamil votes against merger, a possibility increased by the recent intra-community violence. A merged province would be dominated not just by Tamils but by the LTTE, a fact which is not lost on those Tamils whose sympathies lie elsewhere. If administrative boundaries within the province could be redrawn to lie more neatly along ethnic lines, they might be willing to take their chance on a separate province. Until such a result seems assured the referendum may continue to be postponed.

In the meantime the peace settlement is under threat from several directions. As soon as it was announced that the Tamil militants would be required to surrender their arms, there was a run on supplies of polythene sheeting in Jaffna and Trincomalee. What proportion of these arms have been secretly buried instead of surrendered is anybody’s guess, but enough are still in circulation to pose serious problems for the peace-keeping contingent of the Indian Army. Unless the Indians are much more effective than they have been so far, the referendum campaign (if it even takes place) is likely to be a very bloody one. In addition, the LTTE has been protesting at alleged violations of the accord by the Sri Lankan Army and the continuation of Government-organised Sinhalese settlement in ethnically-sensitive areas. At the time of writing the rise in Tamil violence followed the suicide of 12 militants captured by the Sri Lankan Navy, who were flown to Colombo and feared they would be tortured. Recent violence by the LTTE against the Sinhalese probably combines two motives: for implacable opponents of the accord, the desire to provoke a Sinhalese backlash that would destroy it; and for those within the LTTE itself who realise that they may have to live with the accord, the hope that any waverers in the Tamil population will be reminded that the Sinhalese are a far more serious enemy. Among the Sinhalese, and even within the Government, there are also enemies of the accord who have been working to provoke the Tamil militants.

The problem which is at present facing the Indian Army is serious. If its offensive against the LTTE should be prolonged the consequences within India could be explosive. The fifty million Tamils who live across the water in Southern India are in many ways different from the Sri Lankan Tamils, and their affinity with and concern for the Sri Lankans have been exploited more often for domestic political purposes within the state of Tamil Nadu than as a constructive contribution to the conflict. But Rajiv Gandhi has many opponents who will make the most of the killing of Tamils by Indians, and his domestic political position is precarious enough for them to want to strike soon.

Sinhalese opposition to the accord was intense even before the recent attacks. The Inspector General of Police reported on 10 August a dramatic rise in the sale of saffron robes in Colombo, and the Government has recently implemented a system of identity cards for Buddhist monks to prevent impersonation. In particular, there are fears of a major revival of the JVP, which combines revolutionary Marxism with a strongly racist brand of nationalism. JVP sympathisers were almost certainly behind the attack on the Parliament building in August, which failed to harm the President but killed an MP and injured two of the most senior cabinet ministers, one seriously. The Provincial Councils Bill will be presented to Parliament sometime next month, and all MPs have been threatened with death if they vote for it. Only a threat by the President to dissolve Parliament, and the certainty in that event that most of his party’s MPs would be defeated, may bring about its passage. For once in a long history of vacillation and broken undertakings on the ethnic question, President J.R. Jayewardene has thrown his great political skills unequivocally into the search for a solution. But he is now 81, and it will be a long time before the settlement has enough internal momentum to survive without him.

Many Sinhalese have been deeply distressed to see the army of a foreign power take over a part of their country, even at the invitation of their President. For some it confirms a certain paranoia about the Tamil hordes using Tamil provinces as a stepping-stone to conquest of Sri Lanka, a paranoia which has been reinforced by the worsening conflict and has lain behind much Sinhalese opposition to previously-mooted accords – not for nothing have the Sinhalese been described as ‘a majority with a minority complex’. But many more moderate Sinhalese are also disturbed by the surrender of their sovereignty to the regional super-power. The Indian peace-keeping force numbers some eleven thousand. How long it will remain is unclear: the President said that it would leave within weeks, but it is hard to envisage its departure before the referendum has taken place. It long ago became clear that the Sri Lankan security forces, far from containing the civil war, were contributing to the crisis without any prospect of achieving decisive military victory over the insurgents. The May offensive in Jaffna led to a stalemate, and even the areas recaptured from the militants were quickly re-infiltrated. The Indian Army is considerably more disciplined and efficient, and has in theory some chance of being trusted by the Tamil population. But the recent killings in the Eastern Province cast serious doubt on its ability to bring lasting peace – and it is quite certain that without the presence of some kind of foreign force peace cannot be attained in the foreseeable future. It must be hoped that most Sinhalese will realise that the best means of ridding their island of the Indian Army lies in an early implementation of the accord.

It has become fashionable among cocktail-party sociologists, particularly since the Iran-Iraq war and the resurgence of Islamic fundamentalism, to attribute ethnic violence in the modern world to the infection of politics by fanaticism, and to talk as though decent doses of self-interested pragmatism, whatever their other shortcomings, at least help to inoculate the polity against this kind of evil. Sri Lanka is a reminder that self-interested pragmatism is not always so benign. Thirty years ago, the protests against the first accord between the Government of Ceylon and the Tamil community were orchestrated by a middle-aged opposition politician named J.R. Jayewardene, who led a massive protest march on the hill city of Kandy. By an interesting coincidence he is now President of Sri Lanka, and there has been talk of nominating him for a Nobel Peace Prize for his part in the recent accord. But crises of this magnitude are not so easily halted and any talk of the Nobel Prize may soon be forgotten.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN


Please include name, address, and a telephone number.

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences