The Material Basis for Separatism:
Tamil Eelam Movement in Sri Lanka 

Amita Shastri *
Assistant Professor of Political Science,
San Francisco State University - February 1990
Courtesy - The Journal of Asian Studies 49, no. 1 (February 1990):56-77.
© 1990 by the Association for Asian Studies, Inc.

Introduction
 Background
 Population Growth and State-Sponsored Resettlement
 Changing Ethnic Ratios and the Sharpening Ethnic Divide
 Development and Non development of the Jaffna Peninsula
 Emergence of Eelam as an Economically Viable Concept
 Notes
  References

[see also
Territorial Identity of Tamil Eelam: Tamil  & Sinhalese Perspectives - Professor S.Nagarajan, 1988;
Sinhalese - Tamil Relations & the Politics of Space - C.Manogaran, 1997
An Identifiable Homeland - Nadesan Satyendra, Excerpt from Address at Thimpu Talks, August 1985 and
Sinhala Colonisation of Tamil Homeland ]

---

Introduction

The ethnic conflict in Sri Lanka has aroused increasing international attention. The demands of the Sri Lankan Tamil ethno regional movement for greater independence from the Sinhalese dominated center developed through various stages into a call for a separate stare in the mid-1970s. This was followed by an increase in the use of organized violence by both sides in the conflict. Most recently, the continued resistance of core Tamil militants to Indian attempts to institute a solution within a united Sri Lanka has emphasized the independent indigenous roots and powerful motivating vision of the call for a separate state.

Most analyses of the growth of ethnic nationalist movements focus on the chro­nology and format of conflictual political interactions, the deterioration of interethnic perceptions. and the changes in the relative economic positions of ethnic groups within a society because of asymmetrical and disadvantageous distributions of power. Dis­cussions of the most advanced subset of these movements, the separatist ethno regional movements, have failed to clarify the material basis on which the opposite process of a positive counter-identity formation takes place and the way this basis orders the specific elements in the separatist demand.

In an incisive schematic discussion of secessionist movements, Donald Horowitz points out that “advanced” groups from ‘backward” regions (like the Tamils in Sri Lanka) are reluctant and late secessionists (1986:243-49). Coming from a backward region, they need access to jobs and economic opportunities in the whole national market or territory. As members of an economically “advanced” group, they uphold criteria based on “efficiency” and “merit” in economic and employment decisions. It is only when ethnic discrimination and violence force them from positions in the rest of the country that the option of a separate state through secession gains adherents. Since their region is a “backward” one, they recognize that a separate state would offer poorer prospects than the previous unified stare territory. Yet in the deteriorating situation, the “forward” group becomes willing to settle for the ‘expanded oppor­tunities” and greater security offered by a smaller, sovereign state of its own (Horowitz 1986:249). Anger at the injury done to the group “honor” works to unify and mobilize it behind the separatist option.

This scheme is both illuminating and parsimonious. But as the latter, it neglects to throw light on intra group differences and motivations. Why should the majority of the population in the “backward” region identify with the aspirations of what are clearly its more “advanced” (and elitist) sections? Further, why should members of the “advanced” group (or sections of it) be willing to settle for a region so “backward” that it is unable to absorb their skills and capital-both of which are most appropriate and remunerative in a more modern or “advanced” environment? What are the “ex­panded opportunities” offered by such a region?

The model of “internal colonialism” developed by Michael Hechter (1977) makes it possible to explain the dynamics of the emergence of a reactive ethno nationalist regional identity from a more holistic structural perspective. It focuses on the rela­tionships of political and economic dominance that the center seeks to establish and extend over the periphery within societies that are developing economically and, par­ticularly, industrializing. Hechter thus shifts the crucial focus to the concerns, per­ceptions, and sensitivities of the periphery and the established asymmetrical distri­bution of power that is exercised to its disadvantage. His model implicitly recognizes the material potential of the peripheral region that the center seeks to control.

In this article I argue that the material basis for the separatist vision is provided by a conjunction of factors: sections of the group encounter barriers during their “for­ward” migration; in addition, a broad section of the regional population perceives that its relative “backwardness” and marginal status is being deepened by the penetrative, discriminatory thrust of the center’s development policy in the region. And yet, dia­lectically, this very process of development highlights the region’s economic potential and proves important for the emergence of a positive separatist vision-especially if, as in the Sri Lankan case, it is a “ forward” group from a “backward” region that is secessionist. The region will have to hold the potential of becoming “forward” if it is free from unfavorable distributions of power. The lack of such a prospect severely weakens the motivation and broader bases of support that a movement fur separatism can draw on and renders it more easily amenable to suppression.

The next section outlines the social and political factors that had a bearing on the evolution of ethnic group relations in Sri Lanka. I then focus on the long-term de­velopmental trends in the north and east, showing how these trends came to a head in the mid- 1970s and created particular pressures within the minority population living in the region. The subsequent section focuses on the distinctive attributes and processes of socioeconomic change in the Jaffna Peninsula, which cumulated in the tensions and attitudes that made it the epicenter of the separatist movement. I show how the positive economic potential demonstrated by the developmental process was intellectually appropriated by Tamils antagonized by the unitary political framework to create a cohesive argument for the viability of a separate Tamil state. Finally, I weigh the implications this argument has for the current situation in Sri Lanka.

It is important to examine local aspects of the Tamil situation if we are to un­derstand the reason for the separatist movement’s local support as anything but a simple primordial ethnic identification or a malevolent antinational terrorist intent. Such an analysis sheds empirical light on the complex issue of equitable group entitlement under universal rules when groups are differentially endowed with material and political resources. It also allows us to understand the otherwise inexplicable discord between the Sinhalese-dominated center and the Sri Lanka Tamils over who controls these areas, especially the Eastern Province (EP). This contentiousness has been sharply manifest in the failure of governments in Colombo to devolve power to the regional level despite repeated demands by the Tamil. More recently, it has been dramatized in the insistence of Tamil militants on the merger of the EP with the Northern Province (NP) as a necessary condition for any structure of decentralization. Broadly, the analysis serves to reaffirm the importance of political institutions and state organizations in the struc­turing of a society’s political incentives and patterns of political conflict.

In view of the complex, diffuse, and shifting nature of secessionist movements, I adopt an inclusive conception of the term “separatism” as the goal of all those who formally support the struggle for a separate and independent state. Sections of this support base might, in practice, settle for regional autonomy within the existing state. The demands and public postures of groups within such movements shift from au­tonomy to independence and back again depending on a variety of factors: on the state of negotiations between the central government and the separatists; on competitive tactical reasons between separatist groups themselves, or on concealed differences on the subject between them (Horowitz 1986:232). In this case study, such a definition includes the various Tamil militant groups and their precursor, the Tamil United Liberation Front (TULF), but excludes the Federal party (FP).

The Background

Various explanations have been offered for the rise of the Sri Lankan Tamil ethno ­regional movement. Some trace the roots of the conflict to the historic heritage of the island-to memories of past conflict between Sinhalese and Tamils and to divergent ethnic traditions and symbols that continue to be valued today. Others prefer to in­terpret this historical memory as having been shaped by contingencies in the late nineteenth and early twentieth centuries. As various elite groups jockeyed for power, they deployed ethnic symbols to mobilize popular support; this option was obviously most advantageous for elite members of the majority group. Legal, constitutional, and political analysts point to the impact that formats for political participation and representation introduced in the pre- and post-independence periods had for relationships between the two largest communities on the island. In an electoral democracy (after 1924), it was a virtually foregone conclusion that the Sinhalese, constituting more than 70 percent of the population, would dominate the national legislative and ex­ecutive institutions. After independence, the single-member constituency system of elections encouraged the emergence of a highly competitive party system dominated by two pro-Sinhalese parties, the United National party (UNP) and the Sri Lanka Freedom party (SLFP). The Sri Lanka Tamils, a minority community constituting 11 percent of the population, grew increasingly marginalized politically. Yet their ter­ritorial concentration in the north and east strengthened their representation in Par­liament and their identity as an ethnic group. (1)

Contemporary discussions of the subject emphasize the fact that the aspirant Sin­halese Buddhist middle and lower-middle classes asserted themselves in 1956 in not only a political but also an economic direction. They used the majoritarian principle to justify an expansion in the opportunity structure for the Sinhalese. The controversial Official Language Act of 1956 proclaiming Sinhala as the sole official language was the most overt attempt in this regard. Both the major parties used the ethnically imbalanced patterns of representation, patronage, and power aggressively to distribute jobs to members of the majority community and development benefits to their regions. Urban middle-class Tamils from Jaffna in the NP, who had traditionally relied on employment in the professions and public services in the southwest of the island, were hard hit by this turn of events. The proportion of Tamils in public services and the rapidly expanding state sector fell sharply thereafter.

As the representative of Tamil aspirations, the Federal party sponsored agitations to control the erosion in Tamil prospects and made repeated efforts to have a decen­tralized power structure institutionalized. These attempts failed in the face of strident opposition from the Sinhalese Buddhist clergy, pressure groups, and opposition parties. The ethnic conflict was exacerbated by incidents of violence between the two com­munities. Successive governments made a growing use of the increasingly Sinhalese police and army to quell disturbances.

Yet attempts in 1960 and 1965 to mobilize Tamil discontent into a demand for a separate state aroused little support (JKearney 1967:112-14, Wilson 1988:85). The FP continued to uphold the earliest and most significant of the agreements between the two communities, the Bandaranaike-Chelvanayagam Pact of 1957 (the BC Pact). The agreement ceded regional autonomy to the minority provinces, the NP and EP. Under it, they would have had powers over matters relating to local agriculture, land settlement, regional development, law and order, local revenue allocations, use of the Tamil language, and propagation of Tamil culture. Subject to the approval of Parlia­ment, the two provinces or sections of them could cooperate with each other, divide themselves, or merge with each other. From the Tamil perspective the agreement offered the structure of power most compatible with the compulsive realities of a majoritarian democracy at the national level while protecting their distinct needs and identity at the regional level.

The demand for a separate state, however, emerged as a powerful motive force among Tamils by the mid 1970s. Three events were catalytic. At the political level, the keen competition between the two major parties resulted in a cohesive swing vote of about 2 percent nationally, which returned the SLFP-led United Front (UF) with a landslide victory of more than two-thirds of the parliamentary sears (Wilson 1975:178). This landslide deprived the FP of the limited political leverage it had attempted to exercise as “balancer of power” between the two parties in the previous decade. Second and more important, for the first time, the ruling coalition in Parliament had a large enough majority to change the constitution. 

The new constitution of 1972 declared Sinhala the official language and provided Buddhism the foremost place. It concentrated political power in the representative organs of the unitary state - the legislature and the prime-ministerial executive - organs over which Tamils had been able to exercise little influence.

 The Tamils, demanding a federal state and a constitutional status for their language, found themselves powerless to influence the proceedings (Phadnis and Jacob 1972). From their perspective, the new constitution institutionalized and legitimized the growing dominance of the majority community in the state and polity. 

Third and perhaps most important, the controversial policies for university admission instituted by the UF alienated all ranks of Tamil opinion and sharply radicalized Tamil lower middle-class youths (de Silva 1979:486-97). The admission policies struck at the very basis of the economic mobility of Tamils from the urban Jaffna region by limiting their access to higher education and were instrumental in arousing deep and widespread cynicism about just treatment from the Sinhalese-dominated center.

Tamil politics of this period were characterized by sharp dissensions, disillusion­ment, and despair. Despite resistance by the established leadership, youth cadres in­sisted that the goal of a separate stare be adopted by various political forums after 1973. They ultimately succeeded in 1976 at the Vaddukodai Convention of the newly formed TULF (a coalition of various Tamil political groups of which the FP was the leading component). 

The demand of the Tamils thereby formally changed from one for regional autonomy to protect individual and minority rights within the existing state to one fur the right of self-determination by an ethnic group, self-consciously identifying itself as a “nation” and seeking to establish a separate territorial state identity (Thiruchelvam 1977). The results of the 1977 general election, fought by the TULF as a mandate for Eelam (their prospective state), showed that it enjoyed the greatest degree of support in the NP, especially on the Jaffna Peninsula. Support was less clear and more uneven in the EP.(2)

Although the electoral results and the separatist mandate have remained open to varying interpretations, there is no possibility of mistaking the intense degree of alien­ation that existed in the minority areas. What was the broad material basis for the support provided to the TULF and the militant youth cadres in their demand for separatism in the north and east? How did it link up to a particular positive vision of a separate or autonomous State?

Population Growth and State-Sponsored Resettlement

Sinhalese have frequently expressed hostility at Tamil claims of discrimination and neglect in spite of the major investments the central government has undertaken in their regions. The support provided to the Tamil separatist cause in the rural areas of the north and east has to be understood in terms of the impact that the unitary structures of the state, dominated by the majority community, had on the distribution of benefits from development in this region over the preceding decades. These trends reached a critical point in the early 1970s and were accentuated by the implications of the massive Mahaweli Development Project (MDP) if they continued.

Traditionally, paddy cultivation in the Dry Zone had been carried on in minority settlements close to the coast in the four rice-surplus districts of Batticaloa, Amparai, Mannar, and Vavuniya in the EP and NP. The interior expanses of the zone had been characterized by malaria, sparse population, and lack of water for irrigation. Subsistence agriculture was carried on at low levels of input and productivity. Levels of commu­nication and political mobilization were low.

Government health, welfare, and development programs in the region began to change this situation after the 1930s. With a time lag, the region experienced some of the highest rates of population growth-a trend that has been obscured by the demographic and political weight of the heavily populated southwestern core of the country. The high average of population growth was 2 percent and more per annum nationally from 1946 to the early l970s; with a sharp decline in mortality, most of the highest rates of natural increase were in districts occupied by the minorities. 

In the post independence period Vavuniya remained among the top 4 of the island’s 20- 24 districts in its rate of natural increase. Trincomalee, which had the second lowest rate of increase in the 1946-5 3 period, doubled its rate of increase in the subsequent decade along with Batticaloa. Although their increases slowed in the following decade (1963-71), in relative terms they remained the highest in the island. Likewise, Mannar accelerated from having one of the lowest rates of increase in 1946-53 to having one of the highest in 1963-71 (Abeysekera 1981:table 5).

The data available suggests that these trends continued over the 1970s. The Dry Zone districts registered some of the highest population-growth rates (including both natural increase and net migration) in the decade-each was over 2 percent per annum. Vavuniya led with a rate of 8.2 percent. The density of Colombo’s population far surpassed that of all other districts; nevertheless, the Dry Zone districts, which started from low bases, registered some of the highest rates of increase in density over the period 1971-81 (Sri Lanka, Statistical Abstract. 1979. (Sri Lanka, Census of Population, 1981. no. 2). Consequently, in the 1970s and after, there was an increasing need for local economic growth and opportunities for employment to accommodate the rising population as it came of age. This formed the context in which other developments took place.

After the 1930s, successive governments dominated by the Sinhalese elite also sought to alleviate the rural poverty and unemployment in the Sinhalese regions by irrigation expansion, land development, and peasant resettlement. Government ini­tiatives supported a significant movement of the Sinhalese population from overcrowded parts of the Wet Zone to newly developed locations in the Dry Zone. It seems reasonable to assert that very little migration from Sinhalese areas to these previously dry, pe­ripheral, and inhospitable areas would have taken place otherwise. In carrying out this program, the institution of electoral democracy, the goals of economic progress and self-sufficiency, and concern for the welfare of the indigent rural population formed a pool of powerful legitimating instruments and symbols for the central elite.

 As Table I shows, the rural districts in the NP, NCP (North-Central Province), and EP were the leading net immigrant districts on the island between 1946 and 1971. Not surprisingly, in view of the government-sponsored settlement program, the dominant proportion of the in-migrant population was Sinhalese, most often from the Kandyan districts. A significant proportion, however, was made up of Tamils from the Jaffna district (Abeysekera 1981:10-32), apparently migrating on a voluntary, individual basis. Along with establishing the likelihood of competition for the same resources, these trends tended to change ethnic ratios in areas of settlement.

An idea of the magnitude of the resettlement program and its differential impact can be gained through some figures. By the end of 1980 some 247,111. acres of paddy land and 135,007 acres of highland had been developed under the Major Colonisation Schemes and allotted to some 100,611 family units all over the island. According to official records, the proportion of Sinhalese among the allottees settled was about 81 percent - a figure above the proportion of Sinhalese in the total population. 

Among the island’s nine provinces, the predominantly Sinhalese NCP accounted for 42 percent of the total paddy land developed through irrigation schemes and 36 percent of the highland developed. About 35 percent of the total allottees were resettled in these regions, which were, by all accounts, overwhelmingly Sinhalese. 

The EP accounted for 25 percent of the paddy land, 19 percent of the highland, and 20 percent of the allottees. The NP accounted for 10 percent of the paddy land developed, 16 percent of the highland, and 10 percent of the allottees. Within the two minority provinces, Sinhalese formed 89 percent of the allottees in the district of Amparai (excluding the River Division of the Gal Oya scheme, in which Sinhalese formed 2 percent), 64 percent of those in Trincomalee, and 7 percent of those in Vavuniya - proportions well above the ones in the total population of these districts (Peiris 1985:28-30). According to Peiris’s study of the land commissioner’s records, there were no Sinhalese allottees in the districts of Jaffna, Mullaitivu, Batticaloa, and Mannar-something at variance with Tamil allegations (see TULF 14- 15).

Successive regimes at the center consciously tried to make the prospect of migration and resettlement both possible and attractive for the indigent beneficiaries. The settlers were not required to contribute to the cost. No effective charges for land and irrigation were made. In the early schemes, even houses and infrastructure were taken care of.

Local administrative officials also tolerated the illegal occupation and cultivation of state-owned lands because of weaknesses in the implementation machinery, the threat of a loss of productivity, or politically sensitive problems. According to a 1979 survey by the land commissioner’s office, around 4% of the land area in the island was illegally encroached on (Peiris 1985:31-33). Despite minority sensitivities and objections, how­ever, the center made no attempt to maintain local ratios by settling underprivileged persons belonging to the minority groups (especially from the overpopulated Jaffna region) or Tamil estate labor displaced because of land-reform policies.

The investments in irrigation and land development made possible an increase in the net harvested area and an intensification of agricultural production through a greater application of labor and chemical inputs. As a consequence, a shift took place in the major rice-producing regions of the island from the southwest to the rural Dry Zone in the northeast. Although the national production of paddy increased by 121 percent between 1947 and l975, the contribution of the rural Dry Zone rose from less than 35 percent of the total output to nearly half in the same period. The net harvested area in the Dry Zone increased by 86 percent and the number of paddy workers by 152 percent against the national averages of 43 percent and 112 percent, respectively (Moore 1985:tables 7.2, 7.3). As a group, the seven rural Dry Zone districts had above-average-size holdings of land, they produced at least twice as much paddy as was required to feed their rural populations, and three-quarters of their paddy farmers were surplus paddy producers (Moore 1985:145).

The unevenness of the benefits of this development can be gauged from the uneven spread of irrigation facilities, which overwhelmingly favored Sinhalese dominated areas in the region. By the late 1970s irrigation schemes covered 94 percent of the total irrigated area in the NCP and 76 percent in Amparai district. In the predominantly Tamil areas 88 percent of the irrigated area in Mannar was covered by such schemes, and less than 59 percent of the paddy land in Vavuniya and Trincomalee. However, the lack of local water resources, high evaporation, and poor soil conditions in the NP modify the latter figures in that less than 10 percent of the land area in Mannar and Vavuniya districts was utilized for agriculture against the Dry Zone average of 25 percent (Sri Lanka, Census of Agriculture. 1973.’ General Report.’ 24). The main centers of the Sri Lankan Tamil population, the Jaffna and Batticaloa districts, had less than 33 percent and 40 percent of their land irrigated by these schemes. Indeed, these two districts were distinguished in having the lowest percentage of sown land irrigated by any means-around 30 percent.

Overall the government’s policies increased the control of land by members of the majority group and provided broader bases of support to its elite. The political lead­ership of the newly settled areas devolved to Sinhalese traders and former public ser­vants. This elite was originally from the low-country Sinhalese areas and remained oriented to the issues generated in the southwest. Typically its constituents-the Kan­dyan peasant allottees-were less affluent and less experienced in dealing with the state machinery. They remained extremely dependent on the state and its functionaries.

Consequently, policies such as input subsidies and consumer subsidies, which were politically more useful to the regime in Colombo, were also the ones favored by the Sinhalese elite in the colonization areas (Moore 1985:198-201). Such policies benefited the numerous small-holding deficit paddy producers who were dominant in the Wet Zone and in the colonization areas. They went against the interests of the surplus-producing larger farmers of the Dry Zone who would have benefited more From high producer prices. The traditional landowning minority elites among the Tamils and Moors formed a disproportionate number of the surplus-producing group (Moore 1985:193). They consistently advocated greater regional powers over agriculture.

The distance from Colombo and the inefficiencies of Colombo-centered bureau­cracies added urgency to the demand of the minorities for a more decentralized structure of administration. Although the Dry Zone minority areas had some of the highest rates of use of fertilizer, they complained chronically about its local unavailability (Moore 1985:156-57). Bureaucratic inefficiencies were evident in the management of the paddy procurement schemes, tractor pools, and irrigation schemes (Moore 1985:156-57). Whereas the Sinhalese sections of the population could rely on political officials to overcome administrative shortcomings, the minority communities could not.

In political debates, the center consistently attempted to delink the Muslims of the EP from the ‘Tamil” autonomist demand and co-opt them by concrete material enticements to pro-centrist positions. Members of the Muslim elite received positions in the cabinet; they obtained state support in education and in the consolidation and recognition of their personal law (de Silva 1986:228-34). The imposition of district quotas in university admission in the mid- 1970s directly benefited Tamil speaking candidates from EP districts at the expense of those from the Jaffna district.

The massive Mahaweli Development Project negotiated in the early 1970s threat­ened to escalate these developments. The master plan of the MDP envisaged that on its completion in thirty years it would irrigate 900,000 acres, of which 654,000 acres would be formerly uncultivated land. Under the Accelerated MDP adopted in 1978, five major reservoirs were scheduled to be constructed over six years to provide irrigation for some 350,000 acres. Of this area, 320,000 acres would he newly developed land, much of which would lie in the Amparai and Trincomalee districts. Although the original master plan provided for some water to be carried to the NP, this part of the project was later abandoned because of a lack of water in the catchment area. The earlier Major Colonization Schemes, by contrast, had developed merely 380,000 acres of land over a fifty-year period.

Thus, although the rural Dry Zone developed as a whole, the Sinhalese areas and components of the population benefited disproportionately. The Sinhalese dominated unitary state both explicitly and implicitly promoted their interests. These facts in­tensified the perception among the Tamils that they were being inexorably marginalized even within their own areas. They were being left to concentrate in the poorer, more arid areas and to depend on inherited or private material resources. The terms of trade discriminated against them in the interests of social peace in the Sinhalese areas. The situation provided a powerful rationale to both the Tamil elite and the Tamil populace to favor regional autonomy.

What were the extent and the specifics of the demographic changes in ethnic terms in the minority areas? Since they have been a matter of acute controversy, we now examine them in some detail.

Changing Ethnic Ratios and the Sharpening Ethnic Divide

Table 2 shows the significant changes that took place in the ethnic composition of Sri Lanka’s population by administrative districts in the post independence period. 

Group I districts are prominent plantation districts in the Kandyan highlands in which there was a decline in the estate Tamil population, especially in the 1970s, because of repatriation and out-migration to other parts of the island. As a result the proportion of Sinhalese increased significantly. 

Group 2 districts, located in the NP were recipients of estate Tamil population and saw a proportionate decrease of the other ethnic groups, the Sri Lanka Tamils and the Muslims. 

In the Group 3 districts of Colombo and Nuwara Eliya, there was an increase in the Sri Lanka Tamil proportion and a decrease of the Sinhalese one. However, I suspect an overestimate of the former because of self-enumeration of estate Tamils, who had been awarded citizenship in the previous decade, as Sri Lanka Tamils in 1981. 

Group 4 districts consist of those located in the NCP and EP that saw significant increases in the Sinhalese ratio in the population as a consequence of government-sponsored colonization and a decline in the Sri Lankan Tamil and Muslim ratios. However, since minorities had previously formed a very small proportion of the population in the NCP, the decline in their proportion did riot become ‘a major political issue except in the urban areas where Tamils had previously resided in greater concentrations.

The drop in the proportion of Sri Lanka Tamils in the EP districts has formed the crux of the controversy regarding settlement under central auspices, its impact on ethnic ratios, and eroding political influence. In Trincomalee the relative weight of the Sinhalese grew from half that of the Tamils (21 percent and 40 percent, respectively) in 1946 to an equal weight (34 percent each) thirty-five years later. 

The division of Batticaloa in 1962 into two districts had the result of creating, from what had earlier been an undivided Tamil majority district (51 percent Tamil), the large new district of Amparai in which the Tamils were reduced to a minority (24 percent of the pop­ulation) and the Muslims and Sinhalese became the substantial component of the pop­ulation (42 percent and 38 percent, respectively). The Tamils were left to concentrate in the smaller district of Batticaloa (as 71 percent of the population) with the Muslims as a minority (24 percent).

These changing ethnic ratios had significant political implications that worried the Tamils (see TULF 1977). The settlement of Sinhalese in the interior of the EP districts increased their weight both in absolute terms and relative to the minority population settled along the coast and thereby weakened the latter’s claim to be the representatives and spokespersons for constituencies within the province. The first delimitation commission, in 1946, had drawn electoral district lines extending from the more heavily populated minority-dominated zones along the coast to the interior western provincial boundary of the EP. In the wake of the Sinhalese influx in the 1950s, the 1959 delimitation commission created the massive Amparai electoral dis­trict with a 91 percent Sinhalese population, incorporating much of the northwestern and central interior section of the old Batticaloa district. Further institutional rec­ognition of this population and its distinct interests was provided through the creation of the Amparai administrative district in 1963 (3)

As significant to the Tamils were the changes in ethnic ratios effected within the Trincomalee district. Whereas changes in the town of Trincomalee largely took place because of the growth of harbor-related activity during World War II, changes in the interior rural areas stemmed mainly from direct government sponsorship. By 1981 this had reduced the Sri Lankan Tamil population to near or below majority proportions in the population of several crucial areas. 

The proportion of Sinhalese in the three subdistrict or Assistant Government Agent (AGA) divisions of Seruvila, Kantalai, and Tambalagama rose from 13 percent, 4 percent, and 0.3 percent in 1921 to 58 percent, 83 percent, and 58 percent, respectively, by 1981. The settlement in the AGA division of Seruvila, south of Trincomalee, created a disturbing break in continuity in minority-dominated areas of the NP southward through the EP-a break to which the 1976 delimitation commission gave controversial institutional recognition by creating the Seruvila electoral district. This unit had a relatively low population and extended from the coast south of the town of Trincomalee westward and northward to cover the other interior AGA divisions of Kantalai, Moraweva, and Gomarankadawala (see figure 1).

Figure 1 

The Sri Lankan Tamil population also went down significantly in the four remaining divisions of the Trincomalee district. It was reduced from a majority of 57 percent in Mutur, around the harbor, and 59 percent in Kachchuveli, along the coast north of Trincomalee town to 48 percent each. The proportion of Tamils fell in the town of Trincomalee from 75 percent to 54 percent and in Kinniya, another division around the habour, from 36 percent to 7 percent. The proportion of Sinhalese rose significantly in Trincomalee town and Kachchuveli and that of Muslims rose in Muttur and Kinniya.

The abolition of provincial boundaries as recognized administrative units and the shift to districts in 1963 as the desirable sub national units for administration and decentralization has escaped the attention of observers. Doubtless this shift facilitated the increasingly dense nature of state activity at the local level after 1956. However, it fragmented the units of Tamil claims and demands-the provinces (units the BC Pact had recognized)-into smaller units that could be made more limited in their functions and responsibilities. This strategy is apparent in the center’s attempts there­after to retain all policy-making responsibility and delegate only administrative functions to district-level bodies. With the Tamils losing majority status in Amparai and Trincomalee districts, central spokespersons could also argue that the Tamils’ claim to the EP as their “homeland” was an exaggerated one.(4)

The erosion of the autonomist political position was discernible in the election results of several key constituencies in the EP in this period. The large majority of constituencies that voted for the FP or the less well-defined “Independent” label in 1956 with the rise of the language issue had by 1977 joined the national trend in voting for a major Sinhalese political party, especially the UNP. 

Clearly the growing proportion of Sinhalese in the population and the developmental advantages gained by sections of the Muslim and Tamil minority communities in some of the rural agri­cultural constituencies helped to elect candidates upholding more moderate positions and belonging to the prospective winning party at the center. Muslims, in particular, drew benefits from such cooperation with the major parties and sought to protect their weaker numbers and position vis-à-vis the assertiveness of the regional majority community, the Tamils. Important segments of Muslims in the Muslim-majority constituencies of Pottuvil, Kalmunai, Mutur, and Nintavur/Sammanturai who had moved from support from pro-UNP to pro-FP candidates or independents in 1956 had clearly swung back by 1977 to support the two major parties.

 The TULF maintained its support in the Tamil-majority constituencies of Trincomalee and Padiruppu and one of the two-member constituency seats each in Batticaloa and Pottuvil. The constituency of Kalkudah in Barticaloa was the only constituency that had a Sri Lankan Tamil majority (64 percent) but elected a UNP candidate, albeit Tamil, for the 1965, 1970, and 1977 elections. The support base of the UNP grew consistently from 4 percent of the electorate in the March 1960 election to 33 percent by 1977; that of the SLFP grew from 5 percent to 22 percent over the same period (Shasrri 1987:46).

In contrast to this erosion of support in the EP, the FP/TULF grew from strength to strength in the NP between 1952, the first election after its formation, and 1977, when it made a complete sweep of all fourteen seats there. The results clearly dem­onstrated the real and continued support base for the autonomist-separatist demand.

Overall, the effort by the center to undertake an expansionary role and to respond to, protect, and project majority interests into minority areas was clearly part of the process oil centralization of power that was taking place at the national level and the logical obverse of its failure to carry out promises of decentralization to the Tamils. The state used the instruments of central investment in infrastructure, irrigation, resettlement, and agricultural development to penetrate and consolidate its control over previously peripheral areas to the advantage of the majority community supporting it.

 The center and the majority-dominated parties ruling there could not have undertaken the same actions with a similar degree of freedom from regional control under the provisions of the BC Pact; for those very reasons they had a vested interest in delegitimizing and eroding what the pact had recognized. Not surprisingly, no agreement since between the center and Tamil representatives has been willing to concede powers even approaching those conceded by the BC Pact. 

Under the 1972 constitution and later under the 1978 constitution, this process of diluting minority control continued uncontrolled through the mechanics of the MDP. Indeed, the central government might have been successful in its quest to integrate and assimilate the minority areas since it had successfully eroded Tamil bases in the rural EP by the mid- 1970s, had not the conflagrant crisis in the Jaffna Peninsula provided powerful leadership for regional rural discontent.

What were the specifics of this crisis? It is a matter on which there has been a lack of understanding.

Development and Non development of the Jaffna Peninsula

The Jaffna district was (and continues to be) unlike the rest of the Dry Zone in important respects. It was a densely populated, low population-growth urbanized district like the southwestern littoral. Only limited incomes could be derived from its intensively cultivated land. It closely resembled Colombo and Batticaloa in that as many as half its smallholders obtained most of their income from nonagricultural sources (Census of Agriculture 1962, vol. l: table 1). Like the heavily populated Wet-Zone districts, Jaffna demonstrated a very small average size of paddy holdings, a low annual average of paddy production per head of rural population, and one of the lowest average yields of paddy per net harvested acre (Moore 1985: tables 7.1, 7.3). Three-quarters of these paddy farmers were deficit producers, leading to a need to import paddy from the other Dry Zone districts to feed its population. Much of this had previously been imported from the undivided Batticaloa district, the traditional rice bowl of the island.

However, a larger increase in paddy output was achieved in Jaffna than in the urbanized Wet Zone during the post independence period. Although the increase in yield per acre was only 50 percent greater, the increase in net harvested area was four times that of the core and the increase in output per worker around fifteen times greater (Moore 1985 tables 7.1, 7.3). This was achieved through a greater application of labor, the sinking of private tube wells, and an avid adoption of intensive production tech­niques, the last two made possible by increased credit from government banks. These provide some indication of the relatively greater degree of effort that was expended by individual Jaffna cultivators to increase productivity and income from local resources.

A combination of fortuitous circumstances and increased use of credit also made possible a dramatic increase in the production of subsidiary food crops in the early 1970s. An import substitutionist agricultural policy pursued by the center and a serious foreign-exchange situation in the early 1970s led to a sharp rise in the domestic prices of chillies and onions. The high prices made it economically viable for the peasants in Jaffna to increase production dramatically through credit, investment, and the adop­tion of intensive production techniques. The production of chillies increased at an average rate of more than 29 percent per annum and that of onions at more than 14 percent per annum between 1970 and 1977 (computed from RE and NASL 1971- 79). There was a brief period of remarkable prosperity for peasants in the north.

This development and prosperity did not require and was nor preceded by major government investments, irrigation, on land development schemes such as the ones that favored Sinhalese peasants. Indeed, the NP had been neglected as a recipient of such investment. Despite the center’s attempts to gain political mileage from the fortuitous developments (see Jayawickrama 1976:53), the loans had not been gifts like the land in the colonization schemes. Rather, credit was the only element in the input package for agricultural modernization that was available throughout the island on an equal basis (Moore 1985:147).

Consequently, there was little reason for northern peasants to feel obligated to the government at Colombo. There was far more fertile ground for the development of an independent, assertive, bargaining outlook in interactions with the Sinhalese­ dominated state and bureaucracy. Had northern peasants not received their windfall prosperity at this time, the intensity of the crisis and alienation of Tamils in the north might well have been more intense.

And yet their new found confidence was tempered by a knowledge of the fragile base on which their prosperity rested-unusually high prices for the subsidiary crops and limited reservoirs of usable ground water. Were the central government to change its policies and lower the import barrier, as it did in 1977, it would be disastrous for the subsidiary-crop economy in the north. Likewise, extensive pumping of water threat­ened to lower the water table and increase the brackishness of the water. There were reports this was already starting to happen by 1976.

Indeed, the importance placed by central spokespersons on agricultural loans to the north served to emphasize the low rates of public capital expenditure that had been the norm there. Most state expenditures were devoted to the plantation, industrial, and service sectors, predominantly located in the southwest of the island, and to the agricultural sector where they benefited the Sinhalese population most.

The limited alternative economic opportunities available in the minority areas of the north and east can be gauged from the fact that as late as 1975 almost 90 percent of industry in the island continued to be located in just the Western Province (containing Colombo) (AR 1974:74-75). The NP and the EP constituted the next two most industrialized provinces only in nominal terms, having a mere 6 percent and 2 percent of the island’s industry. Of the approximately forty major government-sponsored industrial units, only five were located in predominantly Tamil areas. 

Of these, four had been established in the 1950s and one in the 1960s (Manogaran 1987:130-34, 139). These units were capital intensive and offered limited employment opportunities. The only labor-intensive units in the north and east, the Gal Oya and Kantalai sugar plantations and factories, benefited Sinhalese settled in each area under official auspices. In the Tami] view, the industrial units were exploitative since they had been located in the region primarily for their proximity to particular raw materials. There had been little or no investment by the central government to upgrade their productive capabilities or expand their employment potential. There had also been no efCort by the government to invest in other projects in the north even when productive possibilities for such projects existed, such as in fishing.

The balance of sociopolitical power favoring the Sinhalese and the related unitary structures of government centering on Colombo frustrated attempts of Tamil youths to get jobs or benefit from even the few opportunities available locally. Special mention may be made of central projects like the Pesalai oil prospecting in the north and Prima Flour Mills in the east, which were supposed to enhance the development of these areas. At Pesalai jobs were given on an electoral quota based on the whole island. In contrast to 80 Tamils and Muslims, there were 120-odd Sinhalese employed there (Tribune [Colombo], 20, no. 4, May 51, 1975, p. 16). Of the 451 persons selected to work for the Prima Flour Mill in Trincomalee, more than 84 percent of them were Sinhalese (Manogaran 1987:134).

It is in the context of negligible central investment and limited opportunities for employment regionally that the exclusionary impact of central language, employment, and university admission policies on the urban lower-middle class youths of Jaffna in the early 1970s can be appreciated fully. A few figures indicate the predicament of the Jaffna middle class. As can be seen in table 1, Jaffna had been a net out-migrant district throughout the period 1946-71 like the other southwestern littoral districts of Matara, Kegalle, and Galle. 

Yet as the second-largest urban center on the island with a well-educated work force, it approximated Colombo most closely. But unlike the aspirant population of Colombo, its urban population had to look to other urban centers, es­pecially Colombo, for employment. (5) And unlike the other littoral districts and urban centers in the Sinhalese areas, Jaffna’s Tamil educated middle class found itself hand­icapped in competing for the most secure and prestigious employment in the state sector because of the center’s language and employment policies.

Despite this Jaffna experienced a marked increase in its net out-migration rate from 2.3 to 6.0 per 1,000 in the period 1963-71 and was the only district on the island to do so (Abeysekera 1981:16-17). The 1971 census data reveal that it was also alone in losing more than one-half (65 percent) of its migrants to other urban areas, with 36 percent of its out-migrants going to Colombo alone (Abeysekera 1981:31). These figures emphasize the importance that access to outside employment, especially urban employment, had for Jaffna Tamils. 

In view of the adverse language and employment policies of the central government after 1956, it appears that, as advocated by their leaders (see Kearney 1967:96-97), urban middle-class Sri Lankan Tarnils increasingly entered professions and businesses requiring higher degrees of literacy, skill, technical expertise, self-help, and self-employment but were compelled to do so outside Jaffna in areas where better opportunities were present. This situation dramatically underlines the low capacity that the Jaffna economy had to absorb much of its new swabasha (literally ‘one’s own language”-in this case, Tamil) educated lower-middle class in this period. It also explains the acuteness with which university ad­mission policies struck Jaffna Tamil youths in the early l970s by sharply reducing access to professional and technical programs like medicine, engineering, and physical sciences.

On the other hand, fully 25 percent of those leaving Jaffna chose to go to the rural Dry Zone and another 20 percent to Dry Zone urban areas (Abeysekera 198 1:tables 11, 13). These figures are surprising in view of the established literature’s emphasis on the Tamil stream migrating to Colombo, but not so surprising if we remember the regional contiguity and economic ties that historically existed between the pop­ulation of the two provinces, the NP and ER It serves to reemphasize the importance that migration to the EP had for Jaffna's Tamil-speaking population.

Population data also reveal that Jaffna registered an uncharacteristically high pop­ulation growth rate of close to 1.9 percent in the 1970s. This is significantly above the island’s average of 1.7 percent per annum for that decade (Statistical Abstract, 1979; Census of Population, 1981. no. 2) and strikingly different from the district’s position as one of the low population-growth districts in the preceding decades (Abeysekera 1981:table 5). 

Jaffna also registered a much higher rate of increase in population density than those recorded by the most densely populated districts of the southwest. Its density increased at the rate of 4.1 percent per annum while those of the comparable urban districts of Colombo and Galle increased by 1.2 percent per annum (Statistical Abstract, 1979,’ Census of Population. 1981, no. 2). 

These figures seem to reflect: (1) the consequences of the center’s hostile employment and admission policies, which forced in­creasing numbers of Tamils to remain at home; (2) the in-migration of estate Tamil workers to the NP; and (3) the in-migration of Tamil refugees from other parts of the island in response to various rounds of ethnic violence, especially in the late 1970s. The convergence of all these Tamil groups created a festering hostility toward a system that sought to isolate and exclude various groups of Tamils economically and politically.

Had there been alternative urban centers of growth within the Tamil region, it is quite possible that separatist sentiments would not have developed in the proportions they did. Thus, the question of developing Trincomalee, which was raised in the early 1970s, acquired a new and crucial significance and continues to influence discussions of the development of the north and east today. In the background of the severe foreign exchange crisis and industrial stagnation owing to the limited domestic market, the UF government tabled a white paper in 1972 on the question of the desirability of developing an industrial processing zone. 

The overcrowding, high real-estate values, and overburdened public utilities in the southwest (especially of Colombo district) made it seem desirable to locate such an industrial processing zone a Trincomalee, where it would be free from such problems and could utilize the fine natural harbor (Sun (Colombo), August 8, 1972; June 27, 1972; Tribune 17, no. 43, October 10, 1972, pp. 21-24). After discussion, however, the project was dropped as being too capital-intensive and extensive, since it would require the development of much in­frastructure. The government was facing a serious resource crisis (Shastri 1985:241- 43) and did not want to create another white elephant in the public sector (Tribune 17, no. 43, October 10, 1972, pp. 21-24).

For the Tamils the proposal raised the acute question of who would exercise control over this major new project and its substantial affiliated benefits. Administration by the center in a unitary state would doubtless encourage a new and qualitatively different influx of Sinhalese population to the region; administrators, financiers, industrialists, businessmen, contractors, workers, and their families. New infrastructural and functional links would be forged with the Sinhalese-dominated center, and the Trincomalee area would acquire a new strategic importance as a national center of industry and trade. The ethnic ratios in Trincomalee district were already sensitively balanced at parity between the Sinhalese and Tamil communities as a result Of state-sponsored settlement. The proposed project would without doubt tip the balance in favor of Sinhalese control of the district and its growing economic opportunities.

More than that, it would tip the balance in favor of Sinhalese and central control of the EP as a whole. The large rural tracts of the Amparai and Trincomalee district had already become non-Tamil majority areas, and Tamil autonomists had been able to do little to stem the in-migration of Sinhalese. The 1972 constitution had established the dominance of the unitary center and the Sinhalese Buddhist majority while offering no protection to minorities and no powers to their regions. With a chance in the urban ethnic ratios of Trincomalee, the EP would cease to be an area in which Tamils were the dominant plurality. As the Tamil elites based in Jaffna, Trincomalee and Batticaloa realized, their current economic and political power, their future economic prospects, and the future of their Tamil-speaking support base were inextricably bound together and endangered by this expansion. The historically evolved complex of factors that served to reproduce the structure of regional Tamil society as one of a distinct na­tionality group would be critically and irreversibly damaged (see TULF).

The discussion regarding industrial processing zones emerging in Sri Lanka at the same time proved important. It offered an alternative model for the industrialization of small countries to the import-substitutionist one which had been part of the reigning wisdom since 1956. The new model offered a critical but heretofore missing element in the Tamil argument for the viability of a separate state.

The Emergence of Eelam as an Economically Viable Concept

Previously, the greatest weakness in the Tamil argument for a separate state had been its lack of a viable economic base. The north and east were “peripheral” or “backward” compared with the “forward” southwest of the island. This shortfall seems to have been overcome by the early 1970s. The rural areas of the north and particularly the east had emerged as important paddy-producing regions. The smallholders in the Jaffna region had emerged as important producers of chillies and onions. Indeed, the locus of development in agriculture had shifted to the Dry Zone, and by the beginning of the 1970s Trincomalee was recognized as holding the key to the next stage of industrialization, which would be export based.

As the leading and most public proponent of the separate state of Eelam at the time, the secretary-general of the TULF, A. Amirthalingam, argued in an interview with the author in 1981, the Tamil areas had paddy, fish, and-given suitable poli­cies-various subsidiary crops for consumption and for trade. He asserted that the traditional lack of modern industry in the north and east was no longer an insuperable hurdle to modern statehood. It could be overcome by developing an industrial pro­cessing zone at Trincomalee, which would offer avenues for investment to local capital derived from agriculture and existing business and trade. 

The Tamils were already highly skilled and well entrenched in business and trade, and their talents and capital would be available and able to develop in the new state. Their highly skilled and educated youth would find lucrative employment not only in managing the new in­dustries but also in administering the new state. Foreign capita], foreign technology, and expatriate capital from the substantial Sri Lankan Tamil diaspora in the advanced industrial countries would help to set up not only the industries in the processing zone but also the advanced infrastructure like telecommunications and airports that were required to run a modern state. Most important, the new government would be free to mobilize revenue and direct expenditure for the benefit of the local region and its population. Equally important, it would be free to negotiate with and procure funding from foreign sources as an independent state.(6)

Thus, derived from material conditions, the concept of Eelam had developed until it seemed economically viable to its proponents. The scope of support for the estab­lishment of Eelam, however, varied within the region and its population. The NP, especially the Jaffna region, and the Batticaloa district in the EP, which had the largest concentration of Tamils and had seen the least expenditure of government funds, formed strongholds of support for the separatist cause. Its most adamant and vehement sup­porters, symbolized by the militants, were drawn from Jaffna lower-middle-class youths, who had the least to lose in economic and political terms in a struggle for the proposed state. 

They also advocated a more “socialist” equalitarian model for the new state (LTTE 1983 and 1984). Backing them with some ambivalence were the Tamil upper and middle classes in the north and east: lawyers, landlords, businessmen, and professionals who backed the TULF. They had to be pushed by the radical-youth wing to demand a separate state and consistently sought to effect a compromise short of separation - in effect, they sought to use the threat as a bargaining lever with the central government. Despite their differences, however, they continued to speak publicly for and be identified with the separatist demand and did not disown their association with the militants. A significant segment of the Tamil lower classes, "minority” castes, and estate Tamils in the NP also supported Eelam in 1977. As in all nationalist movements, popular discourse and rhetoric welded the various groups by emphasizing the discrimination they experienced as Tamils and the “honor” that would be theirs in a separate state of their own.

To a nonpartisan but knowledgeable observer, the conception of Eelam involved various economic and political problems, and it is not being propagated or justified here. 

The lack of a clear geographical boundary between the proposed states would make the establishment and defense of an international boundary a contentious issue. 

Another problem was the location of the river heads and reservoirs required to supply the irrigation needs of the proposed state: they would lie outside it in the central highlands. 

The intermixture of ethnic populations posed another formidable problem. A quarter of the Sri Lankan Tamil population lived in Sinhalese areas, and the EP had a significant population of Sinhalese and Muslims. The estate Tamils were concentrated in the central highlands and in Colombo. The migration and exchange of populations between the two states, once Eelam was formed, would carry a heavy price tag in the extant environment of ethnic hostility and violence. 

Whether the minority populations that chose to remain in each of the states could attain justice was another open question. Muslims in the FP were sympathetic to the demand for greater regional autonomy but not sanguine about acquiring it under the hegemony of Tamils, despite the latter’s assurances. 

Above all, as events have proved, the success of efforts to translate Eelam from ideology to reality would depend on a different and wider set of factors: the relative strengths of the proponents of the concept and their opponent (the Sinhalese-dominated government at Colombo); regional geopolitics; and the larger play of power in the international arena. 

The Tamils have had strong support from the neighboring Indian state of Tamil Nadu and from the government of India for their attempts to achieve greater justice and autonomy for the Tamils in Sri Lanka. However, belonging to the international community and sensitive to fissiparous movements within its own borders, India has consistently stopped short of support for a separate state. Under the Indo-Sri Lanka accord in 1987, India attempted to implement just such a policy and posed a mortal threat to the most militant of the separatist groups (the LTTE) in the process.

What needs to be noted, however, is that the concept of Eelam was intimately rooted in both the development and the discrimination experienced by the region and its population. The development of the region demonstrated to Tamils its potential for economic growth and opportunity. The manner in which that development was being effected, through the instrumentalities of the unitary state and majority rule consolidated in ethnic terms, made them acutely aware that they would not be share­holders in this development but would be its casualties Although their declining stake in the system explains their alienation from it and their acceptance of the separatist option, the region’s perceived potential 1or development provided a strong, intense motivation to struggle for the goal of a separate state.

The otherwise inexplicable insistence of Tamils on a unification of the NP and EP in any scheme becomes understandable once the topography, population trends, and productive potential of these regions are understood, intertwined as they have been until now with the concentration of power in the unitary center. 

From the Tamil point of view (as articulated primarily by middle-class Jaffna Tamils), much of the EP would have to be administered jointly with the NP if any regionalization of power and status were to be economically viable and safe from tampering by the Sinhalese majority. Such a unit would assure a viable economic space for the continuation of Tamil language and culture and be significant enough to influence the government at Colombo. Consequently, proposals that offered a devolution of authority to the Tamils in the NP but left its long-term relationship with the EP undecided, as the Indo-Sri Lanka accord did, fell short in the view of core Tamil militants. Any lasting settlement on the issue will have to take this view into account.

To return to theoretical matters, although the region itself remained agricultural and “backward” in relation to the more industrialized, economically active, and “for­ward” southwest, the characteristics and dynamics governing this “backwardness” were no longer the same. It was suffused with a quickening pace of change and development. The situation provided the potential for this region to become a “forward” one-in which members of the “forward” elite of Jaffna could find avenues for employment and mobility commensurate to their potential and aspirations. This demonstrated potential for a “backward” region to become a “forward” one if freed from unfavorable asym­metries in the instituted structure of power formed a critical element in the process of positive counter-identity formation in the secessionist movement.

Notes

(1) The Sri Lanka Tamils, who are the subject of this article and backed the demand for a separate state, constitute a substantial majority in two out of the island’s nine provinces (they are referred to as “Tamils” hereafter, unless otherwise specified). Until as late as the 1981 census, more than half of all Sri Lanka Tamils lived in the NP and a quarter in the EP. The densely populated Jaffna district alone contained 42 percent of the community and was 95 percent Tamil - a factor that had an integral bearing on the weight that the views and interests of the Jaffna-based elite had on Tamil affairs as a whole. The other important minority groups are the estate Tamils (now 6 percent of the population) and the Sri Lanka Moors (or Muslims), forming 7 percent of the population.

(2) Overall, the TULF got 72 percent of the vote on the Jaffna Peninsula, 58 percent in other NP electorates and 32 percent in the ER It received a total of 50 percent of the votes cast in the two provinces but only 42 percent of the electorate supported it. Absentee voters were highest in proportion on the peninsula (19 percent of the electorate) and lowest in EP (13 percent).

(3) The areas that came to form the Amparai district were 8 percent Sinhalese in 1921. The population increased from some 68,000 persons to 388,786 by 1981, along with an equally drastic change in ethnic ratios-the Sinhalese grew to 38 percent of the population (Peiris 1985:20- 21).

(4) Parallel academic arguments using historical and census data have been made to refute the Tamil claim that large areas of the EP are part of their “traditional homelands” (notably Peiris 1985; de Silva 1986:213- 14). While informative, the methodology employed is ques­tionable. For instance, Peiris first isolates the EP for close scrutiny and then argues that the interior areas had been earlier inhabited by Sinhalese and are now depopulated or continue to be (sparsely) inhabited by them. He equates sparse interior settlements (often of fewer than ten persons) to the Tamil and Moor agglomerations of several thousand persons along the coast. No examination of a similar Tamil claim that might be made of parts of Sinhalese-dominated districts is made to balance the analysis. Further, the majority principle for dominance is upheld at the national level but explicitly derogated at the provincial and administrative district level to undermine Tamil claims. Their analyses thus support the homogenizing policies of the central state in regard to language, employment, resettlement, and education. In a fundamental sense, these writers miss the contemporary and dynamic politico economic nature of the conflict between a majority dominated expansionary state and a resistant regional minority.

(5) Colombo’s importance can be gauged by the fact that its population is equal to almost 12 percent of the national population and is five times larger than the next largest urban area- Jaffna

(6) Interview with author, in the lobby of the old Parliament in August 1981. As he clarified and I confirmed from other sources, the outline of this argument had emerged as early as 1976 from the Vaddukodai Convention. Leading academics affiliated with Jaffna University were involved in developing the idea after 1973. To the best of my knowledge, it has not been spelled out in this cohesive manner in public or printed sources. The closest to it is the discussion in Manogaran 1987:179-80. However, conversations with politically active, articulate Jaffrna Tamil expatriates and students revealed they were well aware of it.

References

ABEYSEKERA, DAYALAL. 1981. Regional Patterns of Intercensal and Lifetime Migration in Sri Lanka. Honolulu: Papers of the East-West Population Institute.

CENTRAL BANK OF CEYLON. Review of the Economy. Annual. Annual Report. (Abbreviated as AR)

DE SILVA, CHANDRA R. 1979. “The Impact of Nationalism on Education: The Schools Take-over (1961) and the University Admissions Crisis, 1970-1975 in Collective Identities. Nationalisms, and Protest in Modern Sri Lanka, Michael roberts ed. Colombo: Marga Institute.

DE SILVA, K. M. 1986. Managing Ethnic Tensions in Multi-Ethnic Societies: Sri Lanka 1980-1985. Lanham, Md.: University Press of America.

HECHTER, MICHAEL. 1977. Internal Colonialism: The Celtic Fringe in British National Development, 1536-1966. Berkeley and Los Angeles: University of California Press

HOROWITZ, DONALD. 1986. Ethnic Groups in Conflict. Berkeley and Angeles University of California Press.

JAYAWICKRAMA, NIHAL. 1976. Human Rights in Sri Lanka. Colombo: Ministry Justice.

KEARNEY, ROBERT N. 1967. Communalism and Language in the Politics of Durham, NC: Duke University Press.

--- 1985. “Ethnic Conflict and the Tamil Separatist Movement in Sri Lanka” Asian Survey 25, no. 9:898-917.

LTTE. 1983. Liberation Tigers and the Tamil Eelam Freedom Struggle.

--- 1984. Towards Liberation: Selected Political Documents of the Liberation of Tamil Eelam.

MANOGARAN, C. 1987. Ethnic Conflict and Reconciliation in Sri Lanka. Honolulu University of Hawaii Press.

MOORE, MICK. 1985. The State and Peasant Politics in Sri Lanka. London: Cambridge University Press.

PHADNIS, U., and L. M. JACOB. 1972. “The New Constitution in Sri Lanka” in India Quarterly (New Delhi) 28, no. 3:291-303.

PEIRIS, G. H. 1985. “An Appraisal of the Concept of a Traditional Tamil Horn in Sri Lanka.” Unpublished paper. Kandy: International Center for Ethnic Studies

PONNAMBALAM, SATCHI. 1983. Sri Lanka: The National Question and the Tamil Liberation Struggle. London: Zed Press.

SHASTRI, AMITA. 1985. “Politics of Constitutional Development in South Asia in the Seventies: Case-study of Sri Lanka.” Ph.D. diss., Jawaharlal Nehru University, New Delhi.

--- 1987. “Electoral Competition and Minority Alienation in a Plurality System:. Sri Lanka, 1947- 1977.” Paper presented at the Association for Asian Studies meeting in Boston.

SRI LANKA DEPARTMENT OF CENSUS AND STATISTICS. 1962 and 1973. Census of Agriculture. 1962 and 1973.

--- Census of Population. 1981, no. 2.

--- National Accounts of Sri Lanka. Occasional. (Abbreviated as NASL)

--- Statistical Abstract. Annual.

TAMBIAH, S. J. 1986. Sri Lanka: Ethnic Fratricide and the Dismantling of Democracy Chicago: University of Chicago Press.

TAMIL UNITED LIBERATION FRONT. 1977. “Manifesto 1977.” Logos 16, no. 3:10-25. (Abbreviated as TULF)

THIRUCHELVAM, NEELAN. 1977. “The Making and Unmaking of Constitutions: Some Reflections on the Process.” Ceylon Journal of Historical and Social Studies 7, no. 2: 18-24.

WILSON, A. J. 1975. Electoral Politics in an Emergent State.’ The Ceylon General Election of May 1970. London: Cambridge University Press.

---1988. The Break-up of Sri Lanka. London: C. Hurst.

WRIGGINS, W. H. 1984. “The Present Situation and Outlook for Sri Lanka.” in U.S. House of Representatives, Proceedings Before the Subcommittee on Asian and Pacific Affairs and the Subcommittee on Human Rights and International Organizations.

---