Tamil migration abroad was the largest regional component of Indian emigration during the colonial era. More than 1.5 million ethnic Tamils from south India were enumerated in 1931 in other (mainly British) colonies where they had poured in during the previous one hundred years. A typical feature of Tamil emigration was the ‘kangani’ system in which labour recruitment from India and supervision on the plantations were in the hands of Tamil headmen. Tamil workers were sent mainly to the newly developed plantations, but they were also active in the urban economy. Ceylon, Malaya and Burma were tire main recipient countries of Tamil labour. Other colonies (including French ones) received only several thousands of workers. After independence former colonies with strong local pressure groups tried and got rid of what they saw as disturbing legacy of the British period.In this paper an attempt is made to interpret migration processes in terms of migratory cycle. The cycle of migration streams is divided into three phases: perfect regulation, growing independence, government-controlled termination. These stages of the cycle correspond to the progressive constitution of a permanent migrant community in receiving countries. Such a pattern can help analyse other examples of international labour migration in the contemporary situation.
Introduction
International labour transfers are often considered a recent phenomenon, typical of modern capitalist economies, distinct from earlier waves of population movement converging particularly on the Americas. Yet there has been a long history of labour migrations in Asia and numerous examples occurred involving Chinese or Indian populations from the 19th century onwards. The aim of this paper is in fact to retrace the history of the migrations which affected an area of south India whose inhabitants scattered all around the Indian Ocean. This region, Tamil Nadu,1 did not wait for the population exchanges of recent development to enlarge its migration field: under the aegis of the British colonial rule, overseas emigration became very widespread from the middle of the last century onwards; and the international exodus from this area was probably both more intensive and more long-lasting than from any other part of the Indian subcontinent.
Although the political linking of a disparate collection of countries under the colonial system did not completely abolish the historical and geographical distances which separated the various colonies, it did reduce them considerably and thus created favourable conditions for large-scale redistributions of population. Among the available resources. Tamil labour was the first to be exported, and it is worth emphasising that the internationalisation of this aspect of the demographic regime started even shortly before complete integration of the regional economy into the world-system, which dates rather from the introduction of export-oriented agriculture and the end of the last century.2
Migrations and Colonial Era
In statistical terms, little is known about migration in pre-colonial India; but the geographical distribution of various communities such as linguistic groups or subcastes gives a good indication of the scale of ‘permanent migration within the Indian sub-continent. As far as the Thmil area is concerned, the sociological composition of the population from the last century onward gives a picture of a strong influx of immigrants, with a significant implantation of Telugu populations from Andhra Pradesh, as well as smaller communities originating in Kerala, Maharashtra and even distant Gujarat. Inversely, dispersion of the Tamil population outside its historic region was very insignificant, except in border areas such as Kerala. The sole exception to this rule was the Tamil colonisation of the north and east of the island of Sri Lanka, which took place long before the first European incursions into the Indian Ocean (see table). Tamil communities elsewhere, such as the merchants from the Coromandel coast whom the Portuguese encountered in Malacca in the 16th century, or slaves exported to southeast Asia, hardly constitute significant examples of Tamil emigration.3
The establishment of British control over the Indian sub-continent by the beginning of the 19th century gave an extraordinary stimulation and redirection to the exchange networks, and cheap labour was one of the first raw materials to be exported from India by the British. The British colonial area provided the privileged framework for this movement. linking India to other colonies in the Indian Ocean, but also to other more distant lands (Melanesia, the Caribbean) and to other European colonics. During the early decades of the 19th century, while Great Britain was establishing its supremacy in international exchanges, slavery was being progressively limited in the empire, until finally in 1843 all slaves were freed. The lack of the slave labour which the British had installed on the tropical plantations (producing sugar, coffee, to bananas, tea, . . .) quickly made itself felt, and international migrations of free labourers replaced the recruitment of slaves. Countries whose economy had depended on the continuous importation of slaves found other sources of labour supply within a few decades.4 South India, because of its favourable geographic position and the importance of its colonial ports, was to take on a very special significance within this new system.
In Sri Lanka, the first immigrants arrived towards the end of the 1820s, and their numbers increased in the course of the following decade. Considered as ‘indentured labour’ (labourers bound by a contract which it was almost impossible for them to withdraw from), these immigrants were subject to a quasi-military regimentation, which was later replaced by the ‘kangani’ system, a more flexible arrangement.
Recruitment for Malaya began at almost the same time, dating from the 1830s. Migrations to Mauritius started equally early and very quickly drained off several thousands (as well as immigrants from Bombay). During the two subsequent decades the streams of migration spread to the French Mascareignes (after the abolition of slavery in the French colonies in 1848). and to the colony of Natal in South Africa, where however most Indians originated from the Bombay presidency. There was less Tamil participation in emigration to the Antilles and Fiji; but the French, who controlled trading ports in the Tamil area (Pondicherry and Karaikal) consequently imported a number of Tamils to Martinique, Guadeloupe and Reunion.5
Almost all of these migrations were controlled by numerous limiting regulations, but the Indian government gradually yielded to the pressure of demand from planters in other colonies and liberalised the process of migration. From the time of the first census in 1871 most of these migratory flows became of marginal importance to the Tamil population, apart from those to Sri Lanka and to the Malacca Straits colonies (the Malaysian peninsula and Singapore). Later movements to Burma (Myanmar) followed the progress of colonial conquest into the Burmese hinterland and the consequent penetration of capitalism. The latest target of emigration was doubtless Fiji in the period before the first world war.6 The table summarises estimates of the Tamil population at the different census dates from 1871 to 1981 in these three former crown colonies, which garnered the immense majority of Tamil immigrants.
Also included are more recent figures, from 1981, in order to give an idea of how the communities, which developed as a result of immigration went on evolving long after the mass migrations had come to an end.
The Table brings together various census statistics pertaining to Tamil expatriates; indigenous communities, such as the Tamils of northern Ceylon, are naturally not included in them. The last line of the table summarises the development of the emigrant population during the second half of the colonial period. Its demographic weight was already significant in 1871, the date of the first censuses taken in the colonies; at that time the emigrant population represented 1.5 per cent of that of the Tamil territory in India-about two years of regional demographic growth in the average conditions of the period.
In the course of the last century, the size of the overseas Tamil population fluctuated as a direct function of migratory movements; the immigrant populations were not settled, with many individuals returning regularly to their own country to be replaced by new arrivals, and the sex distribution was very imbalanced. But the Tamil population did gradually settle overseas and a true diaspora developed. Families formed, and there was an ever increasing proportion of women among the migrants.
The internal growth of the emigrant population then became significant, and the percentage of individuals born in the Madras Presidency decreased rapidly in favour of a 'second generation' made of locally-born Tamils; after the second world war, immigrants as such constituted less than half of the ethnically Tamil population, the remainder having been born locally. From this time on, migratory exchanges diminished greatly, with the exception of movements of populations expelled from Sri Lanka and Burma. In 1981 the population of the Tamil diaspora could be estimated at 4.3 per cent of the population of Tamil Nadu, a proportion which has become slightly lower since independence because of expulsions to India. In the absence of regular and reliable statistical series, we have not mentioned the figures for people of Tamil origin recorded elsewhere, even though they may number more than 100,000 individuals, as in Reunion [1987 estimate].
Table: Tamil Population in Sri Lanka, Burma and Malaya 1871-1981 - (figures in thousands) |
| | 1871 | 1881 | 1891 | 1901 | 1911 | 1921 | 1931 | 1946 |
| Sri Lanka | 203.3 | 320.2 | 313.3 | 497.9 | 563.8 | 635.7 | 854.8 | 816.2 |
| Malaya | 27.5 | 36.3 | 62.7 | 98.0 | 220.4 | 387.5 | 514.8 | 461.0 |
| Burma | | 35.1 | 71.4 | 99.6 | 125.7 | 152.3 | 184.l | 90.0 |
| Total | 230.8 | 391.6 | 447.4 | 695.5 | 909.9 | 1175.5 | 1553.7 | 1367.2 |
As proportion of Tamil Nadu
Population (per cent) | 1.5 | 2.5 | 2.5 | 3.6 | 4.4 | 5.4 | 6.6 | 4.9 |
Notes: Sri Lanka: Population of Tamils and Indian Moors according to censuses from 1911 onwards; figures for 1981 [Guilmoto, 1987]; indirect estimates before 1911 based on the total Tamil population.
Malaya and Singapore: Tamil-speaking population, estimates before 1931 based on the population of Indian origin.
Burma: Tamil-speaking population according to censuses; free estimates for 1946 and 1981 due to lack of statistical information. Source: Censuses of countries concerned and my own estimates.
|
The size and direction of migration flows at different periods are the complex outcome of the action of three factors: the availability and the demand for labour on one hand, and the other, the institutional conditions (political or social) which permit such migrations. Analysis in terms of attracting and repelling factors (pull and push) makes it possible to distinguish different periods in the history of migratory exchanges. The classic illustrations of the effects of these factors are the departures precipitated by demographic crises in South India (1847, 1919 and especially 1876-77), and the returns or repatriations of 1930-32, as a consequence of changing circumstances (see Figure).
Yet it is difficult to separate these factors: we cannot cynically isolate the poor conditions in Tamil Nadu, or the narrow interests of the planters, and make them the sole determinants of migration. Rather, changes in the economic system from the 19th century onwards form a framework within which the migratory mechanisms operate. Apart from the coastal regions of south India, the economic system had previously been segmented and enclosed at a regional level, and the labour force was relatively immobile, often statutorily assigned a particular position at the local level (by professional specialisation according to caste).
As a result of colonial unification, the system of exchanges intensified and became more diverse; in addition the internationalisation of the colonial economy had the effect of globalising the labour market, allowing new transfers of labour. At the same time the capitalist system in the colonies experienced a rapid leap forward in certain peripheral regions, particularly in those zones suitable for plantations; this development /was obviously linked to the existence of reverses of cheap and mobile labour.
Tamil Nadu was already densely populated in the 19th century, in some irrigated coastal regions reaching almost 200 inhabitants per sq k h (1871 figure). While the population underwent a noticeable increase, in spite of recurrent spurts of crisis mortality (epidemics and famines), possibilities for emigration within India were limited: urban industrial development had hardly begun in Tamil Nadu, and even Madras, the capital of the presidency, was growing but slowly. Only a few small mountainous areas in the southern ghats, such as the Nilgiris, were able to attract the migratory currents. Economic development was on the other hand, more rapid and concentrated in other parts of the British empire.
This structural imbalance between concentration of population and concentration of capital could not fail to lead rapidly to significant demographic transfers. At the individual level, overseas employment often represented an insurance against the risks involved in the irregularity of farming seasons, as well as a substantial increase in earnings. Since work opportunities for labourers were very limited in south India, and the job market unstable and stagnant, a new possibility of employment at a regular cash salary on overseas plantations represented for many Tamils an unheard of hope The wages offered by the planters were far higher than those current in India [Kumar, 1965: 140-41] . Migration was thus shaped by inequalities in the colonial economy. We should note in addition that once the international migratory streams had been established, they tended to reinforce themselves by a cumulative effect, growing more independent of the economic conditions which had originally given rise to them [cf Massey, 1988: 396-991. The existence of migratory networks between Zuni1 Nadu and other colonies accelerated migration by reducing obstacles which had hindered free movement between the two countries (transport, uprooting, job-hunting. . .): this above all was the role of the 'kangani' system of recruitment which supplanted contract-recruitment (indentured labour), and which we shall examine below.
Destinations of Tamil Migration
Economic development in Malaya (including Singapore), Burma and Sri Lanka was sufficiently different to cause noticeable variations in the orientation of the migration streams at different times and in the different target countries, even I when the overall world situation also sometimes exercised a parallel effect on the demand for labour. We shall present a summary of the development of immigration in these three countries, emphasising particularly Malaya and Sri Lanka, which received the greatest number of Tamils. These descriptions are complemented by the Figure, which shows the annual statistics for net migrations of Indians in Malaya and Sri Lanka; the series for Sri Lanka is not entirely homogeneous because before 1911 it includes only figures for plantation workers (and their families) (hence the irregularity in the series). These data are not without defects, but here they are used only to illustrate more general hypothesis about the development of the migratory currents.8
They broke off in the early 195Os, when significant flows finally dried up (the area was hardly affected by partition). The most recent international migrations affecting Tamil Nadu have remained on a relatively small scale. These have been the temporary recruitment of semi-skilled workers for the countries of the Persian Gulf, and the 'brain drain' towards western countries, particularly the United states.9 Today demographic exchanges between different regions within India are on a much larger scale than departures for foreign countries.
Because of its geographic and cultural closeness, the island of Ceylon enjoyed ideal conditions for the massive transfer of labour from south India.10 It takes only a few hours from the Tamil port of Rameshwaram to Talaimannar on the western coast of the island- at least as long as the currents are not too strong, as they sometimes are during the north-east -monsoon. From there, until the opening of the railway in 1917, the migrants had to travel on foot to the hills in the centre of the country, or follow the coast to the port of Colombo. At the end of the last century new sea connections were opened up between Colombo and various ports on the Tamil coast such as Tuticorin [1891]. The journey up to the plantations in the hill-country was no picnic, for it passed through some particularly unhealthy areas where malaria was a tremendous scourge right up to Independence. Mortality along the way, from cholera. malaria, etc, was high.
The migrants were organised by an overseer, the kangani, who was responsible for them at every stage of their journey to the plantations. Originally he was also the recruiting-officer in Tamil Nadu, who visited the villages of his native region to persuade agricultural labourers and indebted peasants to accompany him to work on the plantations for varying lengths of time.
The kanganis travelled with the groups of Tamils all the way from the place of recruitment to Sri Lanka, advancing them the money required on their journey to the plantations. Once there, they continued to act as supervisors and were responsible to the plantation owners or managers for the workers they had brought, who were usually also indebted to them for the advances given. The system was soon complicated by the introduction of a very Indian stratification of power distribution, with different types of kanganis ranged one above the other.
The role of the kanganis was very important because of the social and financial powers they exercised within the limited geography of the plantations. The advantages of the system were naturally distributed: the migrants were taken care of from beginning to end without having to risk anything in the process of transplantation except their freedom and health; and the kanganis and planters shared the profits from a labour force that could be recruited in the most flexible manner.
The hill-country of the island experienced cycles of prosperity, depending on productivity and the world market for the produce grown there. Coffee, which was the original crop, collapsed during the 1880s under the pressure of competition coupled with a parasitic disease affecting coffee. It was replaced principally by tea, which remains to this day one of the major exports of the island, but also by rubber, which underwent a rapid development at the turn of the century because of an exponentially increasing demand on the world market.
Tea cultivation brought about a qualitative change in the workforce of the plantations, because it requires constant care which can be divided into several precisely-defined tasks. So the migrations became less seasonal and more stable, and started to involve families: unpredictable changes in the workforce would endanger the schedule of cultivation. and the settlement of entire families on the plantations, besides stabilising the labour force, made it possible to employ women and children for the regular plucking of the leaves. However. the nature of the exchanges (gross totals of migration far higher than net totals) shows that the Tamils returned frequently to India, and the proportion of newcomers ('puthal') was practically always lower than that of workers who had been on the island before ('palaiyal').
Fluctuations in the overall volume of new migrants depended principally on economic imbalances. The first Tamils left for Ceylon at the beginning of the 19th century, but emigration became sizeable only from 1830 onwards, and especially after 1850. After a period of prosperity. the disappearance of coffee was reflected for a short time by a lessening of migratory exchanges between the presidency and Sri Lanka, with even some depopulation of the plantations during the early 1880s.
The success of new crops such as tea stimulated fresh recruitments for the estates. and the following decade registered a record number of net migrants (2,71,000 in 1891-1901), with the proportion of women nearing 45 per cent. At the beginning of the 20th century the demographic growth of the ‘Indian Tamils’ (the official label for the Tamil immigrants) was maintained, although immigration slackened slightly, while new methods used on the plantations tended to increase productivity. During the first world war, the moves reversed direction as a consequence of the disturbed economic context and of new legislation aimed at protecting the Indian workers.
The vitality of the resurgence which followed is exceptional: between 1921 and 1931 plantations and other sectors of the economy of the island absorbed more than 300,000 Tamils, not only plantation workers but also both labourers going to work in the, towns (in the ports, on construction sites.. .), and merchants (and moneylenders). In 1927 alone, the official statistics recorded the arrival of 285,000 Indians in Sri Lanka. And yet at the end of the 1920s there was again a reversal in the economic situation, and when salaries drop a great many Indian migrants departed.
The worldwide depression, which hit the island with full force mainly because of the drop in international trade, led to a reduction in the employment of foreign labour, and for the first time in 60 years, the inter-censal migration balance on the island was negative. During the same period, immigration regulations were changed to enable the governments to check mass migration into a region impoverished by recession, and these culminated in an almost total halt to new migrations on the eve of war, after the prohibition of entry to unskilled migrants. This trend continued until the end of the second world war, and the brief acceleration which followed was quickly limited by the policies of the newly-independent governments, especially that in Colombo, which wanted to get rid of a community of foreign origin which in 1946 represented more than 11.6 per cent of the resident population of the island. Immigration was broken off completely in the 19SOs, and official hostility towards the Indian Tamils, who remained stateless after Independence. went on increasing. The crisis finally led to an inter-governmental agreement between India and Sri Lanka, signed in 1964, which provided for the repatriation to India of almost two-thirds of the population of Indian ancestry.11
Malaya
English settlement of Malaya took place at a later date, but the lack of sufficient labour was felt almost immediately.12 The peninsula was relatively sparsely populated, especially in the interior, covered with malaria-infested jungles. Like the Sinhalese, the Malays showed little inclination to work for the planters, and the Chinese, who were already settled there in numbers, were felt to be less manageable than south Indians. The latter, already present in Penang by 1786, arrived in their thousands from the second half of the last century onwards, to work on the coffee and sugar estates; this was the time of ‘indentured labour’, under which the labourers were bound to their employer by three-year contracts whose implications they rarely grasped.
In addition, the distance which separated them from their native presidency discouraged them from leaving at will. Immigration to Malaya increased towards the end of the last century. In the first place, the cost of a passage (from Madras, Nagapattinam or Karaikal) to Penang or Singapore became much cheaper. Secondly, the planters introduced rubber in 1897 and this very soon took the place of coffee. From this time on, more than 10,000 net entries of Indians to Malaya are counted annually (according to Malayan official statistics), but male migrants are still three times more numerous than women. There is a spectacular development of rubber growing and of processing industries connected with it from the beginning of the century, accompanying an unprecedented world demand as a result of the growth of the automobile industry in the rich countries.
At this same period, the ‘indentured labour’ system, which was connected mainly with sugar production, was finally dismantled and the kanganis become predominant. Travelling money was advanced to the aspiring migrants, who then worked on the plantations under a contract from which they could withdraw.
Arriving in Malaya later than in Ceylon, the kangani system never took on the same importance there, and the workers were less in the grip of the recruiters. The number of migrants arriving independently- among which there were less Tamils (but more north Indians and Keralites, etc) and less going to work on plantations (but often merchants and labourers hired by the government)-went on increasing. and created a more fluid and noticeably less rigid employment situation, within a society of a markedly polyethnic character which, to quote Stenson [1980], functioned more like a commercial undertaking than a state.
Another difference from Sri Lanka at this time is the predominantly masculine character of the migrations throughout the period, mainly because of less opportunities for female employment on the Malayan estates, and of independent immigration. It may however be noted that Muslims amongst the Tamil migrants were sometimes able to marry Malay women. Wherever Tamil Muslims immigrated (Sri Lanka, Burma. . .), their community was able to forge very solid links with their local co-religionists, often through matrimonial alliances.
In spite of the distance between south India and Malaya, permanent settlement by the migrants remained insignificant; yet it was to prove more long-lasting than elsewhere, for the Indian community found a favourable niche in the developing Malayan society. The prosperity of Malaya during the years after the first world war led to major influxes of migrants, with more than 3,50,000 arrivals recorded in 1926-27; in 1931 Indians, of whom 83 per cent were Tamils, represented almost I5 per cent of the total population, and an even greater proportion of the labour force, especially in rural areas (Penang, Selangor, Perak. . .)Together with the strong Chinese presence which was concentrated more in the towns, this led to’ the ethnically Malay population becoming a minority in Malaya after 1911.
In 1930, there was a brutal reversal. The planters, who could not dispose of their produce because of the world economic crisis, rapidly reduced production, cut wages and demanded that the coolies from their estates be repatriated. Economic stagnation spread to many other activities in a colony that was based entirely on economic links with the industrialised world. Between 1930 and 1932 more than 150,000 Indians were repatriated, and assisted migrations (kangani recruitments) were totally stopped. A brief-renewal of migration is recorded after 1934. but when the price of rubber dropped again in 1938 the Indian government prohibited the departure of unskilled labourers. In December 1941 the Japanese invasion put a full stop to all migration from India for the duration of the war. The early years of relative independence in Malaya were troubled by numerous industrial disputes, and even more, by the communist insurrection in which other ‘migrants’ from the subcontinent (the Gurkhas of the British army) became involved.
Migrations decreased throughout these years, as the new governments attempted to reduce interdependences which had been initiated by the colonial presence. Nevertheless, the Indian communities which had grown up from immigration, amongst whom the Tamils. were largely predominant (some of them originating in fact from Sri Lanka), were not much threatened by regional political developments. Both in Malay dominated Malaya and in Singapore, the Tamils were integrated without difficulty into a society where ethnic tensions were more pronounced between Chinese and Muslim Malays. Many Tamils maintained close economic or family relations with their homeland, especially the Muslim merchants from the Tamil coastal areas, whose presence in Malaya predates colonisation.
Burma
Migration to Burma is far from being an exclusively Tamil phenomenon; there had been links between Burma and India, especially Bengal, for hundreds of years before colonisation, and there was a well established community of Indian Muslims (Rohingyas) in Arakan on the borders of present-day Bangladesh.13 The migrations began almost immediately after occupation of Rangoon (or Yangon), then only a village, by Anglo-Indian forces in 1824; the spectacular growth of the town after this is moreover very closely connected with the arrival of Indian immigrants, who after 1881 even largely outnumbered Burmese in the municipal population. In 1852 and then in 1886 the British gained control over the rest of Burma and particularly the very fertile Irrawady valley as far as the towns of Mandalay and Pengu, and the territory settled by Indians increased, while exploitation of the valley soils led to a great increase in rice cultivation.
Many of the Indian migrants came from north India, reaching Burma via Calcutta. Another section originated along the whole coast of the Bay of Bengal, from Orissa (Ganjam) to southern Tamil Nadu (Ramanathapuram). During the period covered by the censuses the proportion of Indians in the total population increased more moderately than in the other countries discussed above, reaching a platform of 5.8 per cent in 1931 (not counting the Indians in the district of Akyab which constitutes the Arakan). As in the other cases shown in the Figure, the migratory movements sometimes reversed direction as a result of changing economic circumstances (1910-11, 1930-31.. .), and these reversals were sometimes activated by the anti-Indian agitation which started in. the 1930s.
Amongst the Indians, migrants from Tamil Nadu, identified for all intents and purposes with the Tamil speaking population, never represented more than 20 per cent. More than half the Indians born in the presidency of Madras recorded in the Burmese census were not in fact Tamil, but Oriya or Telugu.
One of the characteristics of emigration to Burma was its temporary and seasonal nature, with a great many Indian workers making the return trip within one year. It is possible too that this kind of circular migration may have caused a relative overestimation of the immigrant population in the censuses.
The female population hardly increased, representing about one third of the Tamils recorded in the census of 1931; yet this proportion of women among the Tamil immigrants is far higher than that recorded among the other Indian immigrants. One cannot help linking the relatively large scale of female immigration among the Tamils, whether in Burma or elsewhere, with the significant involvement of Tamil rural economy and their better status in society.
Since Burma was administered as a part of India until 1936, the movement of Indian labour was not very strictly controlled (hence the shortcomings of the port statistics). A system similar to the kangani system was set up, under the leadership 'of overseers called 'maistries' who controlled the job market, recruiting either directly from Indian villages, or in Burma itself. The maistry exercised a great sway over his recruits which was usually based on indebtedness, and supported by the law. Tamil labour supplied particularly agricultural labourers for the rice fields which made Burma one of the largest riceexporters of colonial Asia.
Another set of migrants, who had arrived earlier, contributed to urban prosperity, particularly in Rangoon, in very diverse occupations: in the port, in factories (rice mills), small businesses, communications (railways, cycle rickshaws) or services (administration, the professions). A small number of Tamils belonging to the chettiyar castes of Ramanathapuram (in the south-east of Tamil Nadu) established the first foundations of a banking system in the country, in which they played a role quite out of proportion to their numbers. During the slump of the early 1930s, which as in other colonies led to the return of many migrants, they gained possession of much of the cultivated land, and were to become the first victims of the nationalist policies of the Burmese after independence. More than elsewhere, the Tamil population in Burma was very heterogeneous during the colonial era, and included coolies with no assets apart from the strength of their arms as well as some of the wealthiest men in the country.
After the trauma of the Japanese occupation, immigration began again, but Burmese independence brought about new legislation that was unfavourable to the Indian presence, and the immigrant population, most of which had refused Burmese citizenship and remained stateless, again began to decline rapidly. During the 1960s, after many businesses had been nationalised, several tens of thousands of inhabitants of Indian origin had to be repatriated: 1,50,000 in 1964-68 according to the government of Madras [Chakravarti, 1971:184]; but unlike the repatriations from Sri Lanka, the rehabilitation of the Burmese immigrants in Tamil Nadu was more successful. In the absence of precise data, it is estimated today that the Indian community in Burma numbers about 350,000, of which a minority are Tamils [Bahadur Singh, 1984]
Tamils Involved in Migration
The development of immigrant Tamil society in the British colonies was characterised for a long time by irregularity of the in- and out-flow of migrants. The intensity of the migratory exchanges and their short-term instability constituted permanent destabilising mechanisms right up to the second world war. Thus from 1925 to 1935 about 400,000 annual displacements (gross migration) are recorded between India and Sri Lanka, which recorded only 600,000 Indian Tamils in 1931; during the same period, the average for more distant Malaya, where 620,000 Indian immigrants were domiciled, is close to 160,000 movements per year.
For several reasons the intensity of the exchanges leads us to think that the numbers of Tamils counted outside India represent a virtual population, constantly depleted and renewed by migratory flows, rather than a settled, self-renewing population. We shall now see that this demographic instability, which looks like a case of severe socioiogical precariousness for overseas Indian society, is on the contrary an outstanding advantage for the colonial economy, where the size of the available labour force reacts instantaneously to the needs of the productive sector.
The immigrant population, also because of its characteristically temporary nature, had for long an unbalanced age and sex composition, with a preponderance of young men-even though among the Tamils, as we have noted, women were often relatively more numerous. The birth rate remained very low until the 1930s. Unable to reproduce itself in a normal rhythm, Tamil immigrant society had in addition to face living conditions that were much more difficult than the attractive wages offered on the plantations would suggest. Not only was the journey dangerous (epidemics on board ship, marches through the jungle...), but once the Indians reached their destination the sanitary conditions there were drastic.
Ecological transplantation, the extremely unhealthy areas where the plantations were situated, lack of hygiene and health protection, all contributed to an extremely high death rate among the immigrants; the oft repeated argument that the Tamils had everything to gain by leaving a land of poverty and famine has difficulty in standing up in the face of the deplorable situation which long prevailed outside India, and which the colonial governments took note of only very tardily, from the 1920s onwards.14
The war in south-east Asia brought about a very severe deterioration in conditions. Although this period was not long-lasting, it was marked by extremely high mortality amongst the immigrant populations except, in Sri Lanka. In Malaya, the Japanese forcibly conscripted tens of millions of plantation workers for the construction of the ‘death railway’ linking Thailand and Burma.l5 In Burma thousands of Indian immigrants died on a forced march, fleeing towards India via Assam to avoid the Japanese advance [Tinker, 1976; Chakrwarti, 1971]
In fact, expatriate Tamil communities retained their fragile character for almost a century because of their dependence on migratory movements for renewal. The real Tamil diaspora formed only long after the start of departures to the British colonies, when two phenomena coincided to give this population demographic stability: on one side, a lessening in the importance of the role of annual labour displacements in the population, and on the other, a rising proportion of women. The integration of the Tamils into local society varied greatly, ranging from the brutal rejection which occurred after independence to permanent settlement.
The geographic isolation of some immigrant groups, who were often sequestered on the plantations, doubtless hindered integration, but other factors had more determining influence. Communal, national or religious affiliations are very vital in this region of Asia, and Tamils were never able to get assimilated into native groups; even in Sri Lanka, where there are many native Tamils. those of Indian origin remained cut off from the rest of society. In this latter country, thirty years after the end of the colonial period, most descendants of immigrants still lived in the same region and followed the same occupations as their forebears.
The lack of diversification, geographically (in ghettos and pockets of concentration) and in work (because of specialisation or lack of qualifications), combined with the maintenance of their distinct ethnic identity, gave the overseas Tamils a specific social profile which marked them out as scapegoats during periods of tension.
In Burma, where resistance to the colonial regime was most powerful, riots against them prefigured the vigorous measures of Burmisation (and expulsions) applied after independence and reactivated under the rule of Ne Win. Even the best-established Indians in government or business had to make way for new native elites.
Similarly in Sri Lanka the wealthiest Indians gradually withdrew: the Tamils on the tea plantations were a kind of forgotten relic of the colonial period, and perhaps owe to the hostilities between Sinhalese and native Tamils in the Jaffna area the relative peace they were able to enjoy after independence, until the repatriations of the 1960s.
It would be reasonable to suppose that their role in the tea industry made them indispensable, even though socially undesirable. It was perhaps been only when the internal growth of the Tamil population, with an attendant risk of unemployment on the plantations, started to endanger an equilibrium which was based on their geographical and political inconspicuousness, that expulsion to India began to seem an appropriate solution. Malaya provides a very different example: the presence of Tamils-or Indians in general-did not provoke there the same outbreaks of violence or administrative hostility as in the other colonies; but it is not certain that the economic position of the immigrants was the only factor favouring permanent settlement there, in spite of their gradual liberation from the plantation economy.
In the history and composition of its population, Malaya resembles the ‘creole’ countries: regions that were sparsely populated before the arrival of the colonisers, who, by engineering large-scale immigration (whether of slave or free labour), totally changed the ethnic make-up of the population, to the extent that the supposedly ‘indigenous’ groups became minorities. This was the case on many islands in the Indian Ocean, but also in the Caribbean and the Pacific Several of these countries house a population of Tamil origin that is well entrenched, even when a minority among other Indian groups: Mauritius. Reunion, Fiji, the state of Natal in South Africa. and the French Antilles, for example.
In more homogeneous countries, independence heralded the arrival on the political scene of an indigenous group with a strong nationalist agenda. The descendants of immigrants. like the Tamils in Burma or the Gujaratis in east Africa, then had to bow before the storm or run the risk of expulsion. The polyethnic character of Malayan society prevented such a polarisation between the sons of the soil (the Malays) and Tamils. Since we have chosen to follow the migration from its point of origin, we shall say a few words about the effects of the phenomenon in Tamil Nadu. The impact. of emigration on the home population was considerable, frequently responsible for a lowering of more than 10 per cent in the natural growth rate between censuses. Some regions, particularly along the coast, undoubtedly felt the demographic effects more than inland districts which were subject to other migratory pressures.
In the cases of Sri Lanka and Malaya, the location of recruitment offices fixed certain focal points, while the kanganis too operated for preference in their own native areas and thus helped establish specific. migratory routes between particular villages and plantations. In some villages, emigration affected all families of a particular caste, and the migratory drainage even led to appreciable local drops in population.16 Most of the Tamils who emigrated belonged to the lower castes (harijan, kallar, vanniyar. . .) and from the poorest sections of the population. Farmers or pastoralists who owned a little property-land, cattle, houses were more reluctant to leave their villages.
For those who left, the opportunity to work abroad offered a way out of an often closed situation in their home villages, where they were almost serfs to the big landowners, bound not only by economic and financial dependency (often deep in debt), but also by their position in the rigid structure of caste relations.
The introduction of the market economy into rural Tamil Nadu, together with the emergence of commercial farming directed towards international markets, weakened the communal socioeconomic fabric, which had been based on traditional exchanges of tribute between different sections of the village society, simultaneously creating a reservoir of potential labour which was constantly fed by a sustained demographic increase. Individual strategies (at the level of the family or social group) gained importance over traditional behaviour patterns based on the largely autarchic economic equilibrium of the rural community, and heralded the appearance of more profit seeking attitudes while stimulating new forms of social mobility.
The profitability of migration for those who returned has been described in very different terms and contexts; but it seems that few migrants were able to accumulate sufficient capital to make any lasting improvement in their situation once they returned home.17 On the other hand, the migrants did regularly send their savings home to India creating a constant transfer of wealth which was fostered by the political integration of the British colonies (migrations to other countries, French colonies or South Africa, lacked just this advantage). This mechanism made possible a significant redistribution of revenues, and productive reinvestments whose positive effect on Tamil Nadu is difficult to assess. One additional indirect effect of international migration was certainly to lighten the labour market in Tamil Nadu and lead to an increase in wages intended to retain potential migrants.
The social agitation which began to develop during the 1930s amongst agricultural labourers in Tanjavur, the district most affected by out migration, is an unmistakable indication of the impact of migration on local economic relationships; but it is beyond the scope of this article to explore all the implications of emigration for social relations. The economic effects within Tamil Nadu of the changes brought about by the departure of a section of the agricultural labour force also remain even more difficult to assess than the decisive contribution made by Tamils to the prosperity of the colonies where they settled-in many cases only to be driven out again in the y a k following the break-up of the British empire.
Perspectives of Migration Cycle
Standing at a certain distance from the context of Tamil Nadu in the 19th century, we may ask what lessons are to be drawn from the migratory experience we have described here. Indian emigration does share certain features with the labour migrations which characterise the wealthy nations after the last war, in that it involved large-scale transfers of low-skilled workers, transplanted into areas with a different culture, who initially had no intention of permanently settling in the target country.
Without this demographic supply, the inability of the local population to respond to the increasing availability of unskilled work, especially on the plantations, would have hindered the economic development of the host countries.
Internationalisation of the area of recruitment. mainly within the British colonies, enabled a nascent capitalism to profit from economic imbalances between different regions and from many advantages connected with the importation of labour: lower wages, more docility, more readiness to accept difficult conditions, etc. Returns home, high mortality and an absence of young married couples combined to reduce the immigrant groups to a floating population unable to reproduce or establish itself, thus delaying the formation of the Tamil diaspora and of established social classes. These phenomena, which we shall link to the migratory cycle as a whole, are not exclusively Tamil experiences.
Before returning to this wider perspective, we would like to attempt to identify different periods, so as to establish the various stages of the migratory cycle. By this term we mean the entire history of the migratory exchanges taking place between two periods of demographic equilibrium (before and after migration). It is certainly possible to conceive of emigration as an integral, and even permanent, part of a given demographic regime; however, a regular migratory deficit over a very long period, as in the case of Ireland, is a very rare phenomenon, since it leads logically to a demographic decline which will tend to stop the human drainage. The migratory cycle, on the contrary, corresponds to a rupture in demographic equilibrium within an environment that is often characterised by an enlargement of recruitment areas and the commercialisation of labour.
The Tamil migratory cycle passed through three successive phases: Migration began with an early period marked by a strict control of the flows both in volume and composition, which were male subject to the production requirements of the host country. These migrations, which are male and temporary, were apparently under the full control of the governments concerned, and served the economy of the colonies.
The regulatory role lied with the governments, for whom laissez faire was often the sole social philosophy, and the power of negotiation available to the fragmented immigrants was kept to a minimum. This period represented a sort of golden age for colonial entrepreneurs: a labour force that was plentiful, non-local, and undemanding responded precisely to fluctuations in production, without any risk of becoming an autonomous pressure group. In fact, the employers saved the costs of labour reproduction in the wider sense, and the only additional expense involved in their reliance on immigration was connected with transportation of the workers.
We have seen that the 19th century was the time of the most rigorous forms of exploitation like indentured labour, which bore witness to the slave heritage of the production organisation. But in one way the system of individual contracts was less convenient than the collective framework of the kanganis. a system which soon came to predominate among the Tamils. The success to which the long life of this system attests is doubtless due to its affinities with agrarian relationships in the Tamil homeland.
The migrants were in fact mainly landless labourers whose working conditions in their original environment were very close to servitude. There too, temporary migrations for work were organised in teams, and departures were rarely individual affairs. The many levels in the authority structure represented by the kangani overseers (who belonged to higher castes) paradoxically ensured a preservation of the groups once they were in the colonies, acting as a buffer between traditional attitudes and the more 'modern' labour relationships on the plantations. The system certainly worked more effectively and completely in Sri Lanka than in Malaya, as if indicating the complete marginalisation of the immigrant community in the island,
The second phase of the migratory cycle is characterised by some relaxation of the economic constraints on labour migration. Demographic crises in south India lessened from the beginning of this century, and distrcss-induced migrations ceased. ln the target countries, expansion or contraction of the productive sector continued to influence demographic exchanges up to the beginning of the 1930s, but from then on a significant proportion of migrants are no longer dependent on these circumstantial phenomena. This is the time when an established overseas diaspora is in the process of formation, which, in spite of its alien ethnic identity, has ifs roots outside Tamil Nadu. Many factors contributed to this change. In the first place, there was an alteration in government attitudes towards the Indian expatriates.
Liberalism, which in practice used to mean indifference, gave way to an increasing concern for the lot of Indian expatriates: the Indian government showed more and more concern for the living conditions of their subjects, finally prohibiting emigration to several countries.
This new order of things was undoubtedly a result of the increasing participation of Indians in public affairs, and of severe criticism from nationalists about the exploitation their compatriots were undergoing abroad. In the host countries too legislation evolved and the situation of the immigrants improved: regulations were made about their stay, which facilitated the establishment of families. Recruitment of women in certain areas (on the Sri Lanka plantations, for example) was also doubtless an additional factor in making the expatriate populations more balanced.
But the distinguishing feature relates rather to independent male migrations: independent as opposed to assisted migrations (with financial advances), led by the kangani from India.. .) are the outcome of new attitudes. They often make use of the established framework, to the extent that the migrant may still rely on a kangani for his placement. But the 'spontaneous'. migratory system, on the basis of family village connections and with parallel circuits for embarkation and employment, lies outside official control. These migrations initiate strategies which become progressively more independent of the system of total care from the village to the workplace. This second period is marked by a progressive 'un-linking' of the migratory phenomenon from the need for labour expressed by entrepreneurs. Independent migrants such as merchants and entrepreneurs certainly played a pioneering role at a very early period, but these often belonged to particular castes and involved very small numbers. This second phase, difficult to date precisely because of the heterogeneity of the migrants (1920-1940?), came in when these changes applied to the majority of departures. Even those who were setting off for the plantations adopted individual strategies. In case of difficulties, they may return for a time to India, or change plantations.
The more dynamic among them looked for other jobs, leaving the sectors reserved exclusively for immigrant labour to take up urban occupations (in Singapore, Colombo, etc.. .). It is during this second stage that the native population begins to feel the threat posed by the foreign presence, because of the increasing competition it presents in many different sectors of the economy. For this reason social diversification on the part of the immigrants did not invariably lead to the third and last phase of settlement, a return to demographic equilibrium.
In several places, as in Burma, hostility towards the Indian immigrants led to their departure once colonial protection was removed. Thus the merchants, bankers and moneylenders from the Tamil chettiyar castes, whose influence had been considerable all over south-east Asia on the eve of the second world war, were driven out of Burma and then of Sri Lanka. The Tamils on the Sri Lankan plantations were, as noted above, protected from hostility by their importance to export-oriented production, and by their relative invisibility.
Many of them did start on the final phase of the migratory cycle after Independence, for example by breaking off ties with their ancestral homeland. In spite of their marginalisation on the plantations, their participation in trades unions indicate a new assumption of responsibility for their own economic future, and they were able io avoid [he worst (general expulsion) in the course of negotiations over their status. The great difficulty with which those who were repatriated have settled in India over the last twenty years testifies 'to the uprooting experienced by these descendants of immigrants. It remained to Malaya to enable the Tamils in general there to complete the migratory cycle by complete integration into the host society.
Very few expatriate Indian communities have been able to conclude their migratory history in this way, for there still exist countries where the descendants of the immigrants are treated as second class citizens (as in Fiji), if they have not been expelled in the end (as in east Africa). Paradoxically, this last stage of migratory historycoincides with the more or less total halting of migrations, to or from India: the end of independent migrations and of direct recruitment in India, and the slackening off of return migrations.
The closing of frontiers, which began before the second world war and became more complete after decolonisation. confronted the immigrant communities with a future in the host society; 18 this resulted in a lessening of social exchanges with India, and the diaspora therefore had to redefine its identity in relation to other local groups.19 Although cultural reference to Indian civilisation may have remained strong, geographic isolation implied i41 obligatory adaptation to living conditions in the country of immigration. The degree of integration probably varied according to social grouping and the possibilities for social mobility within the host society. The long time-span of the history of Tamil international migrations suggested to us this attempt at a division into three periods.
The dating of these different phases certainly varied in different social groups (skilled or not, wage-earners or self-employed...) and different countries. The sequence of these phases should be seen from the internal point of view, with reference to the immigrant populations and their composition, rather than from the standpoint of external determinants. Reconstituting this migratory cycle means above all following what happened to immigrants who, after being reduced for many years to an unstable and virtual population (the first phase), re-formed themselves in the course of the second phase through diversification, and finally became an integral part of the host society in spite of the possible persistence of distinctive social traits based on origin.20 On this point we differ from descriptions of immigration based on divisions into economic or political periods, which predominate in the literature on the topic; although it is quite suitable for taking account particularly of the development of the flows, we find the economic perspective which relies heavily on the development of geographical imbalances between labour and capital inadequate for studying the internal development of expatriate communities.
The perspective from which we have chosen to approach this historical example enables us to cast some light on contemporary labour migrations. in spite of profound differences between the socio historical contexts. The first phase corresponds to the earliest attempts at an international recruitment: recourse to foreign labour, through recruitment that is organised in the country of origin, under the control of the government of the host country, with strict control of the movements so as to give the active population a variable geometry enabling it to respond best to structural variations in employment.
Even the eastern European countries have made use of this policy (for Vietnamese labour). Although this type of system has disappeared in many countries, for example, West Germany where the principle of rotational migrations functioned up to 1970, 21 it still applies in others (Switzerland, the Persian Gulf countries). The migrations are temporary, because dependent on the ocerall situation and controlled by restrictive legislation.
Any possibility of permanent stay is out of the question. Most west European countries quickly passed on to the second phase, with an unregulated increase in independent migrations and the first movements of family reunification; Great Britain figured ;is a pioneer in this area.22 This phenomenon was accompanied by a redefinition of the relationship between the migrations and the economic structures of the host countries, moving in the direction of a progressive independence of migratory phenomenon from labour demand.
This change was made possible mainly by more liberal legislation and the establishment of migratory networks (ethnic- or family-based. . .) functioning independently of governmental institutions. For the western overseas countries policy oscillates between the first phase (national quotas, professional preferences) and the second (clandestine immigration, family groups. . .)
The draconian limitation of immigration in the European countries from 1973 onwards has not precipitated the final phase, because migratory routes-have often circumvented more lenient regimes (clandestine entrants, refugees); the mechanism of family reunification, the other aspect of the second migratory phase, is still functioning. But the serious reduction in the flow of entries observed during the 1980s characterises the last phase of the cycle, in which the question of integration arises because of the limited numerical threat posed by future immigration. The various forms of integration observed (assimilation, ethnic- recognition., .) involve phenomena which are essentially external to the migratory cycle.
The main distinguishing feature of the Tamil cycle is certainly the length of time over which it extended: more than 50 years passed between the earliest departures and the first signs of the formation of a permanent diaspora in comparison the European experience has taken place much more rapidly. Reluctance on the part of the Tamils to emigrate permanently for social reasons, and the strictness of the controls applied, provide the main explanations for this. Demography supplies an additional explanation: with the exception of Malaya, the Tamils rarely migrated to under-populated areas, whereas in Europe the demographic deficit was not inconsiderable.
The differences observed between the various target countries suggest that the ability to absorb a population of external origin, which is linked with the social and demographic flexibility of the host society, is relatively independent of the economic mechanisms which originally gave rise to the population transfers.
Notes
[This paper is drawn from a larger study of the demographic history of Tamil Nadu since the end of the 19th century. My research in India was made possible mainly by the hospitality of the Madras Institute of Development Studies, and the support of Romain Rolland grants from the French ministry of foreign affairs in 986-88. A first version of this paper in French appeared in Revue Europeenne des Migrations Internationales I, 1991]
1 This refers to the predominantly Tamil area at the south-eastern extremity of India, which formed part of the presidency of Madras during the colonial era. The overall demographic context is described in Guilmoto [1992]
2 The economic context is analysed by Baker [1984]; see also Kumar [1965].
3 Tamil dispersal in south-east Asia involved mainly Ceylon and Malaya; cf Nilkanta Shastri (19751 and Sandhu [1962 21-30].
4 This period is described in detail by Tinker [1974]. For migration in general, see also Kumar [1965: 128-31, Tinker [1976. 19771. Waiz [1934], as well as the chapters on migration in the various Census of Madras (from 1871 to 1931).
5 The recent work of Singaravelou [I9871 deals with Indian immigration to the Caribbean area. See too the Revue Carbet, [I989] which contains much information on Reunion in particular.
6 Here the Tamils who immigrated between 1903 and 1915 were only a few thousand amongst a north Indian majority, cf Cillion
7 Dutta [1972] shows moreover, on the basis of an econometric analysis of the migratory flows between India and Sri Lanka from 1920 to 1938. the specific role of economic differentials (measured in terms of salaries and living standards), in addition to the purely climatic factor (Indian agricultural seasons).
8 The data are drawn from the tables of Peebles [1982: 67-70] for Sri Lanka. and from the appendices of Sandhu [1969: 304-171. These are statistics from the target countries, which usually underestimate departures, thus causing owr-estimation of the total migrant population (this being the case for Ceylon in the early part of the period). No comparable figures are available for Burma, which in any case received a majority of non-Tamils.
9 Weiner [1982], and Burki and Subramaniam [1987].
10 On migration to Sri Lanka see especially Jayaraman (19671, Meyer (19781, and Peebles [1982].
11. Several hundred thousand repatriated Indian Tamils have been received in India since this, although for most of themit was not their country of birth. Their reintegration proved to be very difficult, mainly because of their lack of education and the absence of a receiving network. For more details on this ment period, see Guilmoto [I9871 and Fries and Bibin [1984].
12 The standard work on this question is Sandhu [1969]; see also Stenson [1980] and Jain [1970].
13 For migration to Burma, Chakravarti [1971], and Mahajani [1960] may be consulted.
14 In 1885, a semi-official document speaks of good living conditions for coolies in Ceylon, requiring no government intervention [McLean, 1885: 1,5031. For a much less optimistic assessment, and statistics which reflect the terrible mortality affecting the Indian immigrants, see especially Marjoribanks and Marakkayar [1917: 18-21], Sandhu [1969:85, 171]. and Chakravarti [1971: 49]
15 The Indian population of Malaya, estimated at 744,300 in 1941, sinks to 599,600 in the 1946 Census, without any significant migration in the interval. The description of a Malayan plantation during the Japanese occupation [Jain. 1970 297-3071 gives a glimpse of the mechanisms of this crisis.
16 See the examples of Tamil villages described by Slater [1918], and Thomas and Ramakrishnan [1940]; the reports of the decennial censuses in the presidency of Madras supply statistical details, in particular of the degree of demographic stagnation affecting districts with high emigration.
17 Cf. Dennery [1930]. Kumar [1965], Slater [1918], and Thomas and Ramakrishnan
18 Departures for certain destinations had stopped before this: Reunion (1882). the French Antilles (1888), Natal (1911). Fiji (1916).
19 It may also be noted that in the host country, Tamil identity was often disguised behind vague local labels such as 'Malbars'(Reunion), 'K(e)lings' (Malaya), 'Sammies' (south Africa), 'Z'indiens' (Guadeloupe), 'Kala' (Burma).
20 This is clearly the case in Asiatic societies where inherited distinctions (caste, ethnic origin.. .) are among the main criteria of social differentiation. The endogamy of social groupings guarantees their historic identity.
21 On immigration in Germany, cf, Rist [I9781 and Leitner [1987]. See also Hammar.[1985] for a comparative perspective.
22 As is testified by the facilities enjoyed by the Irish and, up to the I%, by Commonwealth immigrants [cf Holmes, 19821.. French migratory history of migration is too long established and complex [Noiriel, 19881 to allow of a brief interpretation here.
References
Bahadur Singh, 1 J (ed) (19841. Indians in South Asia, Sterling, New Delhi.
Baker, C J [1984], An Indian Rural Economy 1880-1955: The Tamilnad Countryside, Oxford University Press, Delhi.
Burki, S J and S Swamy [1987], 'South Asian Migration to United States: Demand and Supply Factors', Economic and political Weekly, March 21, pp 513-17.
Chakravarti, N R [1971]. The Indian Minority in Burma: The Rise and Fall of an Immigrant Community, Oxford University Press, Oxford.
Dennery, Etienne (19301, Foules dilsie: Surpopulation Japonaise, Exparirion Chinoise, Emigration Indienne, Armand Collin. Paris.
Dutta, A (19721, 'An Econometric Study of Indo-Ceylon Labour Migration, 1920-1938: A Critique', Economic Development and Cultural Chunge, 21(1), October. pp 142-57.
Fries, Y and T Bibin [1984], The Undairables Expatriation of Tamil People 's/ Recent Indian Origin'from the Plantations in Sri Lanka, K P Bagchi, Calcutta.
Guilmoto. Christophe [1987], 'Demographie el, Politique: Les Tamouls Entre Sri Lanka Et l'Inde', Population, 2, 283-304. [1992], Un Siecle de Demographie Tamoule, La Population du Tamil Nadu de 1871 a 1981, CEPED, Paris.
Hammar, Thomas (ed) (19851, European Im-' migration pS!icy: A Compamtive Study, Cambridge University Press, Cambridge.
Holmes, Colin (19821, 'The Promised Land? immigration into Britain 1870-1980' in D A Coleman (ed), Demogmphy of Immigrants and Minority Groups in the United Kingdom. Academic Press, London, pp 1-20. '
Jain, R K [1970], South Indians on the Plantations Frontier in Malaya. Yale University Press, New Haven and London.
Jayaraman, R [1967]. 'Indian Emigration IO Ceylon: %me Aspects of the Historical and Social Background of the Emigrants',
Indian Economic and Social History Review. 4(4), December, pp 319-59.
Kumar, Dharma [1965], Land and Caste in South India. Agricultural Labour in the Madms Presidency during the 19th Cenfury, Cambridge University Press, Cambridge.
Leitner, Helga (19871, ‘Regulating Migrant’s Laws: The Dialectic of Migrant Labour and the Contradictions of Regulatory and Integration Policies in the Federal Republic of Germany’ in Glebe, G and J O’Lughlin,
Fbreign Minorities in Continental European Ciries, F Steiner, Wiesbaden.
Mahajani, V [IW]. The Role of Indian Minorities in Burma and Malaya, Vora, Bombay.
Marjoribanks, N E and A K G Ahmad Tambi Marakkayar (19171. Report of Indian Emigrating to Ceylon and Malaya, Government Press, Madras.
Massey, Douglas S [1988], ’Economic Development and International Migration: Population and Development Reviek: 14(3), September, pp 383-414.
Meyer, Eric [1978], ‘Migrations Internes et Migrations Externes a Sri Lanka (Ceylan): Une Appmche Historique‘, CEIhnogmphie, Nilakanta Sastri, K A [1976], A History of Soufh Indi& From Prehisforical Times Io fhe Fallof V~ayanagar, Oxford University Press, Madras.
Noiriel. Gerard [ISSS]. Le Crewel Francais. Hktoire de L”immigmrion XIX-XX siedes, Seuil, Paris.
Peebles, P [1984], Sri Lanka. A Handbook of Historical Statistics, G K Hall, Boston.
Rist, Ray C [1978], Guestworkers in Germany: I3olrppCLs for Plumlism. Praeger, New York.
Sandhu, K S [1%9], Indians in Malaya. Some Aspects of their Immigration and Settlement 1786-1957, Cambridge University Press, Cambridge.
Slater, Gilbert (ed) [1918], Some Soufh Asian Villages, Oxford University Press, London.
Singaravelou [1987]. Les Indiens de la Camibe. 3 tomes, L‘Harmattan, Paris.
Stenson, M [1980], Clas, Race and Colonialism in West Malaysia: The Indian Case, University of Queensland Press, St Lucia.
Thomas, P J and K C Ramakrishnan (eds) [IW], Some South Indian Villages. A Resurvey, University of Madras, Madras.
Tinker, Hugh [1974], A New System of Slavery: The Export of Indian Labour Overseas 1830-1920, Oxford University Press, London.
-[1976], Separate and Unequal: India and the Indians in the British Commonwealth 1920,1950, C Hurst, London.
-[1977], The Banyan Tree: Overseas Emigrants from India. Pakistan and Bangladesh, Oxford University Press, Oxford.
Waiz, S A (ed) [1934], Indians Abroad Directory 1934, The Imperial India Citizen Association, Bombay.
Weiner, Myron [1982], ‘International Migration and Development: Indians in the Persian Gulf, Population and Development Review, 8(1) , March, pp 1-36. 77-78, pp 185-97.
