ISSN (Online) - 2349-8846
-A A +A

Learning to Live in the Colonies and Camps

Repatriates and Refugees in Tamil Nadu

Frank Heidemann ( teaches at the Institute of Social and Cultural Anthropology, LMU University of Munich, Germany. Abhijit Dasgupta ( teaches at the Department of Sociology, Delhi School of Economics, University of Delhi.

Involuntary migration of Tamil repatriates and refugees from Sri Lanka to Tamil Nadu began in the late 1960s and continued for several decades. The relief and rehabilitation offered to them by the Government of India was far from adequate, and life in the camps and colonies was hard and often unbearable. The unsuitable living conditions forced the migrants to learn how to deal with adversity and to assert agency in the midst of despair and hopelessness. Although life in the camps and colonies was difficult, migrants managed to carve out a space for themselves.

Sri Lankan Tamil repatriates and refugees came to Tamil Nadu involuntarily between 1968 and 2009. Each family has its own complex migration history, and their motives and sufferings cannot be studied in the context of a uniform narrative. They struggled for survival, and in the best cases, tried to build a new life in unfamiliar lands and under extremely adverse conditions. The Indian state offered funds, but state actors involved in relief and rehabilitation were largely absent from relief and assistance for these groups. Although repatriates and refugees constitute two distinct categories, they were confounded in coffee-table conversations, in the press, and even in the district collectors’ accounts and similar official reports.

This paper is based on the fieldwork conducted by the authors at intervals over three decades. Frank Heidemann conducted anthropological fieldwork in Tamil Nadu (participant observations, narrative interviews, and visual documentation) from 1982 to 1984, followed by restudies in the Nilgiris district in the 1990s, and in Madurai and the Andaman Islands in 2013 (Heidemann 1989, 1992, 1997, 2006, 2016). Abhijit Dasgupta (2016) conducted fieldwork with Sri Lankan refugees in four different sites in Tamil Nadu. Inferences made from this combined body of fieldwork have been signposted throughout the paper with “field observation.”

Sri Lankan Tamil refugees began to arrive in Tamil Nadu from the early 1980s due to an outbreak of communal riots. The influx of Tamil refugees continued intermittently for nearly three decades thereafter. Repatriates, on the other hand, arrived in India after the signing of international treaties between India and Sri Lanka for resettlement of stateless persons in South India. The two groups are different because of the historical circumstances that led to their migration and the legal mechanisms of their entry into India. The migration of the two groups was, by and large, involuntary. In the case of the repatriates, the two Indo–Sri Lankan agreements (the Sirima–Shastri Pact of 1964 and the Sirima–Gandhi Pact of 1974) regulated the transfer of resident Tamils from the Sri Lankan upcountry (earlier called Ceylon) to India. These stateless Tamils (denied citizenship of Sri Lanka) were descendants of a labour force recruited from India in the colonial period. They were fighting for Sri Lankan citizenship, and their goal was integration. Between 1968 and 1987, a total of 4,59,000 stateless Tamils were made Indian citizens.1 They were uprooted and brought to India. Today, most of them live in Tamil Nadu or neighbouring districts in other South Indian states. They are different from the refugees who moved to India as a result of the Sri Lankan civil war (1983–2009), especially after the 1983 riots.

Refugees were descendants of precolonial Sri Lankan citizens residing in North Sri Lanka (mainly in the Jaffna Peninsula) and in the East, or those who moved to Colombo and other parts of the island country. During the communal riots and civil war between the Liberation Tigers of Tamil Eelam (LTTE) and the Sri Lankan army, they were allowed by the Indian administration to enter India and were accommodated in camps or lived with friends/relatives in towns and villages. In contrast to repatriates, refugees came as Sri Lankan citizens. Further, they wanted to return to Sri Lanka, where their effective leadership was fighting politically and physically. They reached India in various phases as a result of the dynamics of the ethnic violence on the Island. Even after the end of warfare in Sri Lanka, many of them continue to live in government-run shelters in Tamil Nadu. A third category of migrants from Sri Lanka was that of Tamil asylum seekers who came for short stays and with Sri Lankan passports but could not return to their country for various reasons.

What all these different groups have in common is that they speak Tamil as a mother tongue and share a history of involuntary migration, with its origin in Sri Lanka, and the experience of arrival and settlement in new and unfamiliar places in South India. The relief and rehabilitation benefits offered to them were inadequate, camp life was hard, and middlemen and moneylenders made day-to-day existence unbearable (field observation).

However, these adverse conditions had unintended consequences. The refugees and repatriates learned lessons in survival and asserted agency in the midst of despair and hopelessness. Unlike repatriates—who were scattered all over South India and many of whom were not yet fully (or finally) settled— refugees were sent to camps located in different parts of Tamil Nadu and were not allowed to travel outside their camps without the permission of local officials. The dole they received was their only source of sustenance. Men, women, and children had to share small spaces and were under constant surveillance. They were accommodated in huts in villages, in unused official buildings, and in cyclone shelters in coastal areas. The Ministry of Home Affairs of the central government in Delhi offered financial assistance to Tamil Nadu for relief and rehabilitation, but the funds hardly reached the target groups. Insufficient living space and lack of employment opportunities made the situation worse for refugees, but like repatriates, they too discovered new strategies for survival in this extremely adverse situation.

One of the objectives of this paper is to gain clarity in this confounded field and to show why the state rehabilitation programme did so little to offer shelter and economic security to uprooted Tamil people. In doing so, this paper describes how repatriates and refugees exercised agency and began to rebuild their lives in their own ways.

Stateless in Sri Lanka and the Bilateral Pacts

The history of involuntary migration between India and Sri Lanka began more than 100 years before Lal Bahadur Shastri and Sirimavo Bandaranaike signed the first of two Indo–Sri Lankan repatriation agreements in 1964. The forebears of upcountry Sri Lankan Tamils in Sri Lanka were recruited as indentured labourers from drought-ridden coastal districts of South India in the 19th and 20th centuries to work as agricultural labourers (Heidemann 1992). In the 1830s–40s, they came as seasonal labourers to coffee plantations; many died while travelling from India to Sri Lanka. Later, in the 1880s, migrants were brought to tea and rubber plantations as a permanent labour force. After independence, these upcountry Indian Tamils were not granted citizenship because the political leadership of Ceylon wanted to keep the size of the Tamil population as small as possible. Therefore, their status remained unclear. Without the rights afforded by citizenship, they could not participate in elections and effectively had no political representation. For a long time, India rejected calls to take back the resident Tamil plantation workers from this neighbouring state and grant them Indian citizenship (Coelo 1976).

In 1964, an estimated 9,75,000 individuals of Indian origin lived stateless in Ceylon. The first Indo–Sri Lankan Peace Accord, usually referred to as the Sirima–Shastri Pact, determined their destiny according to the following compromise: in total, 5,25,000 persons would be granted Indian citizenship and be repatriated, while 3,00,000 people would stay in Sri Lanka and become citizens of the state in which they were born (Phadnis and Kumar 1975). Ten years later, in 1974, a second agreement was reached between Indian Prime Minister Indira Gandhi and Sri Lankan Prime Minister Sirimavo Bandaranaike that divided the remaining 1,50,000 persons (who were not covered by the first pact) equally: 75,000 individuals were to be granted Indian citizenship, and 75,000 were identified for Sri Lankan citizenship (Muni 1984). Thus, 6,00,000 Tamils in total were to be repatriated to India, and 3,75,000 were to remain in Sri Lanka.

The physical process of repatriation began in 1968 but came to a standstill after the outbreak of civil war in Sri Lanka in 1983. As of 2015, an estimated 5,04,000 people (plus “natural increase,” that is, children born to the group after 1964) were repatriated to India. The status of the remaining 96,000 people remained unclear until 1986, when the Sri Lankan Parliament passed the Grant of Citizenship to Stateless Persons Act (No 5) (field notes). One clause of the act mentioned that those 96,000 applicants identified for repatriation would now be granted Sri Lankan citizenship. With the enactment of this legislation, the number of persons who were to receive Sri Lankan citizenship rose to 4,69,000. It should be emphasised that the pacts brought under their purview some 1.5 million Tamils, but many were unaccounted for, and their fate was uncertain at best (field notes).


Many of those upcountry Tamils affected by the Indo–Sri Lankan Peace Accord considered the Island their homeland and applied for Sri Lankan citizenship. Most applications, however, were rejected. As a result, these Tamils had to apply for Indian citizenship and leave Sri Lanka. Otherwise, their work on the state-owned plantations would be terminated and a male member of the family arrested, sent to jail in Colombo, and then taken to the ferry, where his relatives would join him to make the journey to India.

There were protests against this forcible repatriation, but the Sri Lankan government took a firm stand on implementing the repatriation schemes. However, there were also many “voluntary” migrants who, later in India, claimed that they had been misinformed about their prospects in India. They were promised employment or land for cultivation and shelter, but they had scant information about the new land, little education, and no skills except for experience in plantation work. They did not know about the heat of the South Indian plains, and they were not aware of social conduct in a multi-caste village. Many had arrived thinking that Tamil was the only language spoken all over India.

In 1983, ethnic riots broke out in Sri Lanka at a time when 1,50,000 Tamils were awaiting repatriation due to procedural delays. The riots changed the political equation and the entire landscape of North and North-East Sri Lanka. The stateless upcountry Tamils became incredibly vulnerable during the ethnic crisis. It was too dangerous to stay and it was time for them to leave at any cost. Many of the stateless upcountry Tamils took it upon themselves to travel by boat with the displaced Tamils from Sri Lanka’s North and North-East. Some went to Jaffna and used it as a departure point. Life in Jaffna too was not safe, as bombing and shelling forced many to flee the city. Three different categories of Tamils—repatriates with Indian passports, refugees, and stateless asylum seekers—all began their journey to India, at times all three groups represented in one boat. After arrival in Tamil Nadu, they were treated uniformly as Sri Lankan Tamil refugees. Each person received an identity slip, which entitled them to relief and rehabilitation.

Before the riots in Sri Lanka broke out in the 1980s, the Government of India organised the migration process and provided several rehabilitation schemes for Sri Lankan repatriates. The largest and least successful scheme was called a “business loan” and consisted of a train ticket for all members of a family to their destination and a “loan” of ₹5,000. In total, 75,738 families opted for this scheme, often on the grounds that no other scheme was available at that time. In Mandapam camp—ironically, the former quarantine camp of their forebears on their way to Ceylon—they were given ₹500 in cash and a cheque for ₹2,500. After proving that they had made a successful investment in whatever business they chose—tailoring, bicycle repair, and so on—they were eligible to apply for the balance (₹2,000), but this rarely happened. In addition, a housing scheme was set up to provide shelter. More than 40,000 families were listed as receivers for this scheme alone, but contractors made such applications on behalf of Sri Lankan repatriates and collected the funds; only in rare cases did repatriate families actually move into a house built on their behalf (Center for Research on New International Economic Order [CRNIEO] 1983; Heidemann 1989).

There were a few other schemes that offered, at least in the beginning, real prospects to the new arrivals. In one scheme, 2,326 families received employment in plantations in the Nilgiri hills. This was an ideal rehabilitation scheme as the repatriates had been displaced from the tea estates in Sri Lanka and the climate in the Nilgiris suited them. In another scheme, 2,484 families were granted agricultural land in dry remote areas. In this way, repatriates were used as a part of the plan for land colonisation (CRNIEO 1983). The slack season in agriculture caused unemployment, poverty, and starvation. Repatriates responded by moving to small towns and larger cities and engaging in casual labour in places like Madurai, Tiruchirappalli, Tirunelveli, and even faraway Madras. The repatriates deserted the land due to harsh conditions. A substantial number believed in the promises of labour recruiters and builders or road constructors. However, unlike their forebears, who migrated in groups and were guided by professional recruiters-cum-foremen called kanganies (Heidemann 1992), these repatriates moved in small groups or as individuals. Families were torn apart by the pressing need to secure employment, while some individuals joined remote family branches, if work was available (Fries and Bibbin 1984; Sivanandam and Nathan 1989).

Another rehabilitation project, involving employment in the industrial sector, looked promising. In all, 3,871 families received jobs in the industrial sector, and 4,918 families were accommodated through the Repatriates Cooperative Finance and Development Bank. The rest of the repatriate families, 5,031 in total, were sent out of Tamil Nadu to places in Kerala, Karnataka, or Andhra Pradesh or to the union territories of Pondicherry and the Andaman and Nicobar Islands (Suryanarayan et al 1986; Heidemann 2016).

The day of departure from Mandapam camp marked a caesura in the lives of repatriate families. Up to that point, the state machinery was omnipresent: state agreements made them move to India; government agents directed them to the ferry and into the camps; repatriates lived under state surveillance; and clerks registered their names in files, granted them access to a rehabilitation scheme, issued the railway ticket, and allocated them shelter and some relief. Their agency was subsumed by the state machinery and hardly visible. This changed drastically on the day of departure. With little and vague information about their destinations, those who opted for the business loan scheme boarded the passenger coaches at Rameswaram and headed into a new world. In most cases, they collected information about life in India from fellow repatriates, relatives, or neighbours from their former plantations. Despite difficult circumstances and lack of communication technology, the repatriates managed to travel long distances and settle in an unknown terrain. They were on their own, without any assistance from the state. Their information on the conditions in the new terrain was based on letters from relatives or friends who wrote about housing, employment, weather conditions, and the whereabouts of other repatriates (Heidemann 1989).

Fifty Years Later—Madurai

In order to better understand experiences with the various state rehabilitation projects, two areas were selected in 2014 in which to study the life stories of Tamil repatriates. Half a century after the signing of the Sirima–Shastri Pact, it is not easy to locate repatriates in most towns and in the countryside of Tamil Nadu. Most places known from fieldwork and visits in the 1980s were deserted, without any traces of the migrants. In some of the rehabilitation sites and housing colonies, the term “repatriate” is still in use. It was hard to collect accounts of their experiences, their trials and tribulations. What, then, happened to the repatriates? Two locations near Madurai showed two different processes at work: first was the assimilation into the local community and development of a new identity; the second was pauperisation and desertion of the houses they had received from the state.

In spite of the desire to avoid the term “repatriate” and to make uprooted Tamils from Sri Lanka in all respects Indian citizens, it was found that the second- and third-generation migrants that were interviewed were still identified as “Ceylon Repatriates,” or “Ceylon People.” While in the 1980s–90s, the dialect and appearance of the older generations were different from other Tamils, their children and grandchildren became South Indian Tamils in all respects. In contrast to those families who settled singly and who intermarried with other groups in South India, the repatriates in and around the sites of former rehabilitation schemes constitute clearly demarcated groups. Their distinction is based not on caste, religion, language, profession, or political association but on their shared heritage as part of an administrative project. In the beginning, people in the repatriate category were sponsored by the government. Later, they became a target group for non-governmental organisations (NGOs), who played an important role in promoting the nomenclature. Looking back after half a century, the term “repatriate” appears as a state-made category.2 Unlike the “ethnographic state,” which invented and created groups by means of enumeration and description (Dirks 2001: 125–227), the postcolonial state was involved in a massive migration across national boundaries and named all those who arrived in India, repatriates. At a time when countries were objecting to any attempts to refranchise a group of people in a state, India went ahead with a new experiment at the cost of too many lives.

A few locations around Madurai were used exclusively for the repatriates. In Pudur, the Directorate of Rehabilitation built 350 houses, 325 for Sri Lankan repatriates and 25 for Burmese repatriates. After a decade, most of the repatriates deserted this area. In 1976, a local Tamil academician was heading an NGO, but during a visit in March 2013 he found his former office uninhabited; workers had deserted the area. Likewise, Ramachandran, a Sri Lankan-born spokesperson revealed that the All India United Front Federation for Repatriates—an NGO he managed—ran out of funds several years ago.

In the 1970s and 1980s there was—to a certain extent—an awareness about our suffering, but today NGOs take no interest in us. The rehabilitation schemes have run out; the state has forgotten us. Here, in Pudur, most of the families sold their houses and moved to other regions; only 5% entered state government jobs, a further 5% ran small businesses like tailoring or a petty shop, and the rest worked as construction workers. The violence done to us is forgotten. We have no family networks in Tamil Nadu like other day labourers. (personal interview, 2013)

As with many other settlements, where one or two repatriate families achieved greater success than others, one repatriate in Pudur became a local leader, owned properties and businesses, sent his children abroad for higher education, and acted as a spokesperson for both repatriates and other Tamils. One thing is common among these local leaders: they come from the families of village heads, trade union leaders, office clerks, or shop owners. The common belief that economic success is often reproduced in a particular social milieu can be confirmed in such case studies.

Melur, located about 30 km north of Madurai, was the other resettlement colony that repatriates left. In all, 190 families lived in the Mill Gate Ceylon Colony of Melur. In 1999, about 160 families left after the spinning mill, which was inaugurated as a repatriate rehabilitation scheme in 1976, was closed down (group interview with trade union members, 2013). Compared to Pudur and other colonies, repatriates were unemployed and frustrated in Melur. They were active members in local sangams (associations) and sought improvement by means of government help. They organised weekly meetings, maintained lists of active members, wrote petitions, and maintained contacts with politicians, administrators, and journalists. They were well organised but lacked the numbers to effectively put pressure on the local administration to improve conditions in the colony. Raju, a spokesperson, said,

Housing colonies and factories were built for the benefit of the builders, not for us. Those who earned a profit went away. We are left behind without work and prospects. (personal interview, 2013)

Tamil Nadu Tea Plantations

As stated earlier, the most successful and sustainable scheme was the one that provided employment in tea plantations in the Nilgiris district. Here the repatriates managed to rebuild their lives all over again, putting to good use their skills as plantation workers and mingling with the local population and evolving their own culture and lifestyle (CRNIEO 1983; Heidemann 1989). By 2014, when this research was conducted, more than 2,000 families had found work in the fields of the Tamil Nadu Tea Plantation Corporation (TANTEA), located in the western part of the Nilgiris district in Gudalur taluk, neighbouring the state of Kerala. Encouraged by the upcountry climate and the hope of work in the tea industries, a massive inward migration of the relatives and friends of these families from Sri Lanka took place.

In the mid-1980s, more than 15,000 families from Sri Lanka reached Gudalur and settled near the TANTEA plantations on their own initiative. Some repatriates purchased land near TANTEA property, built huts on barren land, and started cultivation. They earned a reputation of industrious and motivated workers across the Kerala border where they were then able to earn higher wages as construction workers. Subsequently, the more successful families bought small plots of land, their sons married Malayalam-speaking women, and they spent a good deal on their grandchildren’s education. However, this good life in Gudalur was short lived. The drop in the demand for tea in national and international markets had an adverse effect on tea production. After a long period of economic depression in the tea sector, repatriates began to move away, the families leaving for nearby towns in search of work. They retained their homes and left some of their children in Gudalur with relatives to continue their schooling, in the hope that the good days would return in the near future (Vamadevan 1989). One of the local trade unionists, an upcountry Tamil who came with a Sri Lankan passport after the violence in Sri Lanka reached the plantation area, explained:

We learned from letters that this was the best place for Ceylon People to settle. Tea could be grown, empty land was available at an affordable rate, and good schools were close by. But the ownership status of the land was unclear; you could buy but without patta [land title]. Elephants came and destroyed rice fields, and the drop in the auction rate for tea increased unemployment. In the 1990s, leaders from the LTTE wanted to recruit supporters here, but most Tamils were reluctant. We ran away from the war and did not want to fight. Today everything is calm—too calm—[there is] less work, and many locked their houses and went to other places for better-paid daily work. (personal interview 2013)

Colony Building and Regaining Agency

The repatriates showed their tenacity on a number of occasions. In various parts of the Nilgiris district, they showed extraordinary determination to reconstruct their lives by building tenements and huts with their own labour. In the eastern part of the Nilgiris district, in Kotagiri taluk, they built 21 colonies between 1972 and 1984 (Heidemann 1989: 184) and a few more in the years that followed. The first colony, Queenshola, is located next to a TANTEA plantation, and most of the others were built along main roads or on revenue land at some distance from the villages of the Badagas, the local farming and landowning community. All houses built on land set aside for a highway project were demolished by the local administration (Heidemann 1989: 183–87).

The new colonies began to change the landscape of the hilly areas of the Nilgiris. Unlike in other areas of Tamil Nadu, where repatriates ended up in slums and labourer accommodation or settled among other migrants, the repatriates in Kotagiri coalesced into a distinct group that retained agency in an adverse situation and made full use of it. They exercised this agency not only in the building and rebuilding of colonies but also in developing new ways to coexist with local landlords. To Badagas and other local residents, the move to construct colonies appeared as both a challenge and a puzzle. The labourers left behind proper stone houses in the Badaga fields and instead began investing their energy and resources in constructing mud huts without land title, running water, or electricity. They risked loss of employment and eviction from the land they had occupied.

In an incident in Alakarai in 2002, repatriates settled on what was supposedly revenue land but actually turned out to be unused property belonging to two Badaga families who were engaged in a legal conflict. The owners of the disputed property evicted the settlers and demolished all huts. In other cases, the police or the forest department evicted the settlers. Conflicts with Badagas arose often, especially when the colony was too close to the land-owned by the villagers (Heidemann 2006). When tensions arose, Badagas felt that they had treated repatriates appropriately as dependent labourers. They considered themselves as landlords with paternalistic duties and the right to command.

Decades before repatriation, the Tamils in Sri Lanka had acted as a unified labour force as they did not like individual dependence on the plantation owners. In the colonies of the Nilgiris, they gained what they once had in the Sri Lankan plantations, and they were proud of what they had achieved. They made their place and became their own agents—visible to all and manifested in material form (Heidemann 1997). Today, repatriate colonies are part of the social landscape. Houses have tiled roofs, electricity, and water pipes within reach. In most colonies, they have temples. They are a community of repatriates living together, permanently, with faith in their future.3

In recent years, Badagas and repatriates began to practise a rather pragmatic way of coexistence and cooperation. In many cases, a long-term relationship between landowner and labourer continued or emerged after colonies were built, but this kind of relationship was less paternalistic; the labourers were free to move if they liked. This is why many repatriates moved out of the district in the 1990s for job opportunities at construction sites, on government projects, or with contractors repairing roads. They went as individuals or in small groups of men or as young couples, leaving their children behind with parents. Some repatriate families were even able to buy or lease Badaga tea land. Occasionally, Badaga landowners complained that no labourers could be found to harvest vegetables and pick tea. Repatriates were paid less on the plantations compared to wages on construction sites and thus, at times, stayed away from agricultural work. This gave the repatriate community better bargaining power. As a result, those who were employed on rather uniform day wages in the 1990s now receive wages based on the tea harvest (in kilograms), or work on contract. A system of overtime was introduced, and young energetic people worked day jobs and additionally harvested tea in the fields in the evening or early morning hours. The work relationship became more formal, structured, and complex. These repatriates reject working for lower wages as a matter of self-respect and as a kind of economic strategy (Heidemann 2006; Vedavalli 1994).

To sum up the repatriate case, in the first decades after their arrival in India, Sri Lankan repatriates had little knowledge about social and economic practices in Tamil Nadu. It took almost a generation to build new networks and to emerge as a mobile, hardworking labour force, organised in trade unions or other associations. Today, the repatriates are scattered all over Tamil Nadu and beyond. They marry partners from their ancestors’ villages or from other repatriate communities, often within the same caste, but marriages beyond the limits of their own community, religion, and language group are not considered a violation of norms. Rather, what they object to are patronising relationships and old forms of bondage. They consider themselves modern and class-conscious labourers. However, many of them live in precisely the state of economic dependency they sought to avoid. What makes them different from other day labourers is their deep rejection of what they call “a state of slavery.” They negotiate for wages and are willing to follow work opportunities even to remote places. They build their own homes but have become a mobile working force. This double strategy of simultaneous place-making and mobility—of rooting yet routing themselves—enabled them to survive in an unfamiliar terrain.

Refugees and Journeys to West Asia

Unlike repatriates, Sri Lankan Tamil refugees began to arrive in the southern Indian state of Tamil Nadu in 1983, and this exodus continued until 2009. On two occasions, attempts were made to send the refugees back: the first was in 1987 after the signing of the Indo–Sri Lanka Peace Accord; the second was in 1991 after the assassination of former Indian Prime Minister Rajiv Gandhi. According to official reports, about 1,50,000 refugees arrived in Tamil Nadu between 1983 and 1987, trying to escape ethnic violence in Sri Lanka. Of these refugees, 45,000 were repatriated between 1987 and 1989. The next exodus took place in 1989 after the withdrawal of the Indian Peace Keeping Force (IPKF) from Sri Lanka (Dasgupta 2003). Many Tamil civilians left Sri Lanka because they were caught in the crossfire between the LTTE and the Sri Lankan army. Most of these refugees were repatriated in 1991 and 1992 after the assassination of Rajiv Gandhi.4 The next influx began in April 1995 and coincided with the declaration of Eelam War III.5 The refugee influx continued until the end of the war between the Sri Lankan army and the LTTE in 2009. The story of Sri Lankan Tamil refugees, therefore, revolves around three large-scale displacements—or three phases of displacement—and repatriations on two occasions (Table 1).

A simple calculation of the official data in Table 1 shows that out of a total of 2,78,481 only 1,93,604 refugees left Tamil Nadu either during periods of state-sponsored repatriation or through private channels. If one considers the total time span as mentioned in the table, then the calculations show that since the early 1980s, 2,78,481 refugees arrived (24 July 1983 to 25 February 2002) out of which about 1,00,000 refugees were sent back in two phases (1987–89 and 1992–95).6 Another 1,00,000 left India for Sri Lanka or a third country on their own with valid travel documents. By the end of 2013, India had about 1,00,000 Tamil refugees, 70,000 of them living in refugee camps, and the rest with friends and relatives or in rented flats (Dasgupta 2016).

Camps were located in different districts in Tamil Nadu. Some were constructed on the outskirts of villages, and some were situated in abandoned houses and schools. Most had only basic amenities and were in a deplorable condition. The worst camps were those set up in cyclone shelters. Thousands of refugees were asked to move into cyclone shelters, located near the seashore. Adele Balasingham (2001) gives a vivid account of these camps.

The cyclone shelters are circular structures, dotted along the eastern coastal belt of Tamil Nadu. They are located here for shelter for the surrounding villages if cyclone lashes the coastal area … The roomless shelter had become a maze of ragged and colourful saris. It is one of the ironies of human life that, despite the commonality of situations of collective living, the people always revert to their basic social cell—the family. So in a desperate attempt for privacy, each family had cordoned off a small area—sometimes a small area of a few square feet—by tying their least-needed saris together and hanging them as a de facto wall, separating themselves from their neighbour. Behind these veiled walls, families of five, six, seven, eight, perhaps even more, the very old, the newly married, and the newborn would stake their claim to survive. Smoke and fumes from kerosene cookers or makeshift wood stoves made the place eerie as well as unhealthy and dangerous. (p 94)

Refugees who were sent to coastal districts ended up in these cyclone shelters; the rest of them had to move to other parts of the state (field observation). Almost all the districts had camps. Some, such as Tiruvannamalai (13), Dharmapuri (10), Coimbatore (9), and Salem (8) had several camps, while some, such as Kanchipuram (1) and Tiruvallur (2) had only a few (Dasgupta 2003). Small huts were constructed in the rural areas to accommodate the refugees. In addition, abandoned houses, school buildings, and so on were also used for the refugees.

In cyclone shelters, village huts, and abandoned buildings, the asylum seekers had extremely limited access to basic necessities. Those who received shelter in the huts in villages were the lucky ones as they managed to get open space and fresh air. Some of them managed to earn money as daily agricultural labourers. Tamil refugees were kept away not only from local Tamils but also from NGOs (field observation). In many parts of the world, NGOs work to support and assist refugees. Timely intervention by these organisations has saved many lives in Africa, and more recently, in the Balkans and West Asia. Even in India, voluntary organisations played an important role in the relief and rehabilitation of refugees from Bangladesh, Tibet, Afghanistan, and more recently, from Myanmar. However, NGOs—including the United Nations High Commissioner for Refugees (UNHCR)—had been barred from entering camps in Tamil Nadu or carrying out any programmes for the refugees. Many refugees had approached UNHCR Chennai to obtain visas, permission to remain as refugees, or for help in locating untraceable family members. In the absence of official permission to work among the refugees, the UNHCR could do little but refer all these cases to the refugee rehabilitation department of the Government of Tamil Nadu.7 Poor living conditions, inadequate dole, and restriction of movement made camp life unbearable for the refugees.

Economic hardships caused by the lack of assistance received from the government, scarce employment opportunities in and around refugee camps, and poor living conditions forced many men and women to take drastic steps. Women between the ages of 18 and 30 found a unique way of meeting their needs. Some individuals in this group received offers from agents in Sri Lanka (who had counterparts in India) to go to Saudi Arabia or to other countries in West Asia to work as housemaids. In order to secure a job, the woman had to pay money to a broker. Those who were healthy enough to undertake the journey and willing to change their religious identity from Hindu to Muslim were offered these positions. They also had to forego their right to stay as refugees in camps as their absence from the camps automatically led to the deletion of their names from the list of refugees eligible for doles. In financial terms, the opportunity to work as a housemaid in West Asia was a lucrative one as maids earned a minimum of ₹4,000–₹5,000 a month (and sometimes more, depending on the ability of the worker and the generosity of the employer).

To learn about the working experience of women who took up these positions, including their journey to West Asia and earnings, refugee women were interviewed in four different locations: in the districts of Tiruvallur (Gummidipoondi camp), Vellore (Abdullapuram camp), Pudukkottai (Thoppukollai camp), and Thoothukudi (Thapathy camp). The narratives collected in these camps highlight on the one hand the complications involved in working as housemaids in West Asia, and on the other, how the conditions helped them to exercise agency. Women from different age groups were identified for interview so that a broad spectrum of views could be obtained. They were interviewed inside the camp, sometimes in the presence of their husbands. Their stories shifted from one theme to another, often marked by emotion. From these narratives, let us look at only those sections that are relevant to the study of women’s agency.

Let us consider the experiences of Lakshmi and Asai. Lakshmi arrived in India in 1990 from war-torn Jaffna and found shelter in Thoppukollai camp. She spent three years in Saudi Arabia as a housemaid, for which she earned ₹5,000 per month. She had to borrow money to meet the expenses for the journey to Riyadh but repaid her loan once she started earning. At the end of the third year, she obtained a tourist visa to return to India directly from Riyadh. She reached Chennai and then went to the camp to join her husband, daughter, and son. Since she had obtained a tourist visa, remaining in India after the expiry of her visa was illegal. She continued to live in the camp with her family, used her savings during her stay, and paid money to the “Q” branch official and the revenue inspector to allow her to stay within the camp.8 In this way, she managed to carry on her camp life after her visa expired.

But not everyone is as lucky as Lakshmi. Asai, an inmate of Abdullapuram camp, left for Saudi Arabia in 1996 and returned in 1998 when her two-year contract came to an end. Like Lakshmi, she entered India with a tourist visa valid for six months. Before her visa expired, she left for Jeddah, where she remained for three years. She finally returned to Abdullapuram camp in 2001, once again on a six-month tourist visa, and then continued to live in the camp after her visa expired. She was taken into police custody for overstay and unlawful entry into the camp, but she managed to obtain release as the local “Q” branch officials gave her permission to stay as a special case. Both Lakshmi and Asai were aware of the fact that their entry into, and residence in, the camps was in violation of the law. However, they had hardly any options except to break the rules and do their best in an adverse situation. In these cases, one also sees how refugees exercised agency to explore alternative sources of earning and (re)establish family life.

Repatriation of Refugees to Sri Lanka

The assertion of agency and taking action in the context of an overwhelmingly unfavourable situation was noted also at the time of repatriation on two occasions: the first in 1987, at the time of intervention of the IPKF in Sri Lanka, and the second after the assassination of Rajiv Gandhi in 1991. If one examines anti-repatriation protests, one again finds the presence and relevance of refugee agency. Following the assassination in May 1991, there were strong demands from the Indian government and by the All India Anna Dravida Munnetra Kazhagam (AIADMK) party in Tamil Nadu for the repatriation of Sri Lankan refugees (field observation). The AIADMK spearheaded the demand for repatriation because all the four individuals accused of the assassination of Rajiv Gandhi and those who received death sentences from the Supreme Court of India were all refugees: Perurvalam was from upcountry Sri Lanka and had studied electronic engineering there; Murugan arrived in India in 1991 and was studying engineering; Nalini, a friend of Murugan, arrived in 1991; Santha came to India in 1991, just before the assassination, and stayed with the photographer Haribabu, who was killed on the spot after the assassination. The Indian government reconsidered suitable amendments to its earlier policy in order to send the refugees back as early as possible. But the situation in Sri Lanka was far from normal; the war between the Sri Lankan army and the LTTE had paralysed normal life on the island.

In order to capture the mood of the hosts in the wake of Rajiv Gandhi’s death, let us consider the views expressed in reports published at that time. An interview with a Tamil government official revealed that the Indian government wanted refugees to vacate camps in Tamil Nadu and to go to Madhu in Sri Lanka, where temporary shelters were available for them. The official expressed concern that there were over 2,00,000 refugees in Tamil Nadu and that many of them might not return to Sri Lanka at all. He said that given India’s long-standing policy against forcible repatriation, it would be difficult to send the refugees back at a later stage. The central government finally announced that the first group of Sri Lankan refugees staying in Tamil Nadu would be returning to the Island in the first half of December 1991 (Suryanarayan and Sudarsen 2000).

In 1991, the decision to repatriate camp refugees to Sri Lanka was taken in spite of humanitarian concerns. The UNHCR carried out a survey in order to find out whether refugees were ready to give consent to plans for repatriation (Dasgupta 2003). Of the 53,661 refugees who filled out the questionnaire, 16,663 expressed their willingness to return to Sri Lanka, while the remaining 36,998 refugees were not in favour of repatriation. The state went ahead with the plan to deport as many refugees as possible, including those housed in special camps. The officers from Tamil Nadu decided to complete this process by the end of November 1991. As far as 208 Sri Lankan government servants were concerned, it was confirmed that the Sri Lankan government was willing to receive them, and arrangements would be made to send them back by air. It was decided that there would be no repatriation of 88 “hard-core militants” and 365 “hard-core supporters” of the LTTE, who were reportedly in the custody of the special investigation team probing the assassination of Rajiv Gandhi (field notes).

The arrangement for repatriation moved at a fast pace, and this triggered action on the part of the refugees. Refugees came out in protest against repatriation (field observation). The central government decided to postpone refugee repatriation for the time being, but it began again in 2002. During the peace process, the refugees were asked one more time to return to Sri Lanka.9 This time, the refugees insisted that there must be an unambiguous agreement between the Sri Lankan army and the LTTE to resolve the ethnic question, eschew violence, and restore democracy and the rule of law in North and North-East Sri Lanka before any such return took place. The refugees raised the following concerns:(i) representation in the peace talks had to be broadened, deepened, and made more transparent, so that all citizens could genuinely participate in their own political destiny;(ii) the agreement reached between the two parties (the Sri Lankan army and the LTTE) had to guarantee all Sri Lankans and persons of recent Indian origin domiciled there human rights, including the right to life, citizenship, property,universal suffrage, and so on, as well as the right to self-determination, as described by the Universal Declaration of Human Rights and upheld by international law and practice; (iii) the agreement had to make specific provision for the return of refugees through the inclusion of written accords that guaranteed the rights, safety, and security of those returning and their social and material wellbeing; (iv) the refugees had to be involved in the negotiation of such accords and in validating the social and material conditions that would make possible their return. Refugees had to play an active and informed part in the return process itself. Repatriation could only take place on a voluntary basis, upholding international standards of safety and dignity.10

The state repatriation programme failed to achieve its objectives in the face of stiff resistance from the refugees. Refugees raised their voice at a time when the Indian government and almost all the political parties in Tamil Nadu were demanding stern action against them. This reconfirms the point made earlier that refugees do exercise their agency at a critical time and against all odds.

The return to their homeland after exile is a special occasion in the life of refugees because of the possibility of reuniting with family members and friends who stayed back, returning to possessions and property, and more importantly, the opportunity to live once more in one’s homeland in peace. One can cite the example of the return of refugees to Bangladesh after the Bangladesh Liberation War or return of the displaced to their homeland in many European countries after World War II (1939–45). However, for Tamils, the return home was far from a happy experience. Those who returned to the Island after a long stay in a foreign country developed a different perception of their homeland. Bitter memories of war shaped this perception, and the uncertainty of life after return created fear and insecurity. Their neighbours were not keen on welcoming the returnees. Some had occupied the land or taken over the property of those who had left Sri Lanka, and they were not prepared to relinquish either. Those who stayed back felt that their claim and ownership was justified because they had stayed and weathered the rough days (field observation). Thousands of refugees are now living in camps in Tamil Nadu and waiting to return to Sri Lanka.


Repatriates and refugees found their feet on Indian soil with some limited assistance from the state, but largely through their own efforts to rebuild their lives in extremely difficult conditions. The rehabilitation programme failed to offer them something akin to their setting in Sri Lanka. The rehabilitation spaces were new, and they had to come to terms with an unfamiliar and sometimes unproductive terrain. Rehabilitation in the Nilgiris was an exception, but successful settlement in colonies was the result of individual efforts and not of the state-run repatriation schemes. Refugees too developed ingenious ways to cope with and adapt to adverse conditions in the camps. The dole they received was far from adequate, so women refugees undertook journeys all the way to West Asia where they worked as housemaids to earn money for their distant families. In camps, when attempts were made to repatriate refugees to Sri Lanka, residents openly resisted the programme. These are just two examples that show how at a critical time Sri Lankan refugees exercised agency and worked not only to survive but to build a better life in India.


1 Statistical data on repatriates are based on information from the Director of Rehabilitation and published in Heidemann (1989: 303).

2 It must be noted that this term was a misnomer right from the start. Upcountry Tamils were not brought back to their home country. Rather, they were transferred to the larger region from which their forebears had historically originated. Moreover, many sites of repatriate settlement, such as the Nilgiris district, were as far away from their ancestral places as from upcountry Sri Lanka.

3 We are aware that demography is a highly contested field and that population numbers can constitute political statements. Census data on repatriate settlement colonies do not exist. Therefore, we chose to reproduce a census conducted by a local NGO, Makkal Maruvazhoo Mandram. From their count, the following figures indicate the numbers of families living in repatriate housing colonies in the Nilgiris: Kotagiri taluk—11,047; Coonoor taluk—7,382; Gudalur taluk—24,375. There is thus a total of 42,804 repatriate families in the Nilgiris. These numbers, we must add, may include Tamil (and other) day labourers. However, these housing colonies were initiated and are shaped by Sri Lankan repatriates.

4 See Dasgupta (2003) for an overview of the controversial repatriation undertaken by the Government of India in 1991.

5 Eelam War I began after the spread of riots against Tamils in 1983. Eelam War II started in 1990. In April 1995, the LTTE broke off talks with the Sri Lankan government and declared Eelam War III. For more information, see Adele Balasingham’s (2001) book on the Tamil resistance.

6 Repatriation phases are drawn separately from influx phases and therefore will not correspond with Table 1.

7 The Commissionerate of Rehabilitation and Welfare of Non-resident Tamils of the Tamil Nadu government looked after the camps, administered doles, and kept track of the whereabouts of the refugees.

8 The “Q” branch of the government of Tamil Nadu now looks after refugee-related matters. It is basically an intelligence wing of the government and performs the tasks of surveillance and policing.

9 A peace process was initiated by the European Union in 2002 in order to solve ethnic crises in the Island.

10 This protest was organised under the leadership of the Organisation for Eelam Refugee Rehabilitation, an organisation of displaced Tamils.


Balasingham, Adele (2001): The Will to Freedom: An Inside View of Tamil Resistance, Mitcham: Fairmax Publishing.

CRNIEO (1983): A Report on the Survey of Repatriates from Sri Lanka, second ed, Center for Research on New International Economic Order, Madras.

Coelo, Vincent Herbert (1976): Across the Palk Straits: India–Sri Lanka Relations, Delhi: Palit and Palit.

Dasgupta, Abhijit (2003): “Repatriation of Sri Lankan Refugees: Unfinished Tasks,” Economic & Political Weekly, Vol 38, No 24, pp 2365–67.

— (2016): Displacement and Exile: The State-Refugee Relations in India, New Delhi: Oxford University Press.

Dirks, Nicholas B (2001): Castes of Mind: Colonialism and the Making of Modern India, Princeton: Princeton University Press.

Fries, Yvonne and Thomas Bibbin (1984): The Undesirables: The Expatriation of Tamil People of “Recent Indian Origin” from the Plantations in Sri Lanka to India, Calcutta: K P Bagchi.

Heidemann, Frank (1989): Die Hochland-Tamilen in Sri Lanka und ihre Repatriierung nach Indien (Up-country Tamils in Sri Lanka and the Repatriation to India), Göttingen: Edition Re.

— (1992): Kanganies in Sri Lanka and Malaysia, München: Anacon.

— (1997): “Immigrant Labourers and Local Networks in the Nilgiri,” Blue Mountains Revisited, Paul Hockings (ed), Delhi: Oxford University Press, pp 148–63.

— (2006): “Konsensus nur in der Vielstimmigkeit, Ethik, Ethos, Ethnos,” Annette Hornbacher (ed), Bielefeld: Transkript, pp 313–32.

— (2016): “Spatial Dimensions in Narratives of History: Sri Lankan Repatriates on Katchal and Little Andaman,” Manifestation of History: Time, Space and Community in the Andaman Islands, Frank Heidemann and Philipp Zehmisch (eds), New Delhi: Primus, pp 96–121.

Ministry of Home Affairs, Government of India (2002): Annual Reports, New Delhi: Government of India Press.

Muni, Anuradha (1984): “The Implementation of Sirimavo–Shastri Agreement on the Persons of Indian Origin in Sri Lanka,” Indians in South Asia, I J Bahadur Singh (ed), New Delhi: Sterling, pp 69–91.

Phadnis, Urmila and Lalit Kumar (1975): “The Sirimavo–Shastri Pact of 1964: Problems and Prospects of Implementation,” India Quarterly, Vol 31, No 3, pp 249–69.

Sivanandam, A and P S Nathan (eds) (1989): Alienated Everywhere: A Study on the Conditions of the Sri Lanka Tamils of Indian Origin and the Sri Lanka Repatriates in India, Ceylon Repatriates Association, Kodaikanal.

Suryanarayan, V and V Sudarsen (2000): Between Fear and Hope: Sri Lankan Refugees in Tamil Nadu, Chennai: T R Publications.

Suryanarayan, V, M Vamadevan and R R Sivalingam (eds) (1986): Rehabilitation of Sri Lankan Repatriates: A Critical Appraisal, Madras: Madras University Press.

Vamadevan, M (1989): Sri Lankan Repatriates in Tamil Nadu: Rehabilitation and Integration, Madras: Zen Publisher.

Vedavalli, L (1994): A Socio-economic Profile of Sri Lankan Repatriates in Kotagiri, Chennai: South Asia Books.

Updated On : 22nd Feb, 2018


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top