Some Comments on the Social Backgrounds of the April 1971 Insurgency – Gananath Obeyesekere

“This high level of political consciousness accounts for the fact that of the parliamentary democracies in Asia, Sri Lanka was the only country that consistently threw out governments through a popular vote so that no government has had more than two consecutive returns to power, and most governments have had only one term of office. However, while the peasants were politically articulate, they had no access to political power.’the classic paradox of the Marxist civil society was operative here in extremis. The only occasions where peasants could exercise their power was at an election; but then often, irrespective of the party in control, effective political power remained (as it still remains) in the same ruling elite. In other words, political consciousness is widely diffused but political participation and decision-making remains in a ruling elite. None of this would really matter if the opportunity structure were flexible; but jobs and other forms of privilege were dependent on access to the centers of political power from which these persons were debarred.”

This essay by Gananath Obeyesekere originally apeared in Vol. 33, No. 3 of The Journal of Asian Studies.

Some Comments on the Social Backgrounds of the April 1971 Insurgency in Sri Lanka – Gananath Obeyesekere

Part I

MANY scholarly articles and some of a polemical nature have been written on the 1971 insurgency in Sri Lanka. Recent studies of this movement (Fernando 1972, Egan 1972, Kearney 1973, Arasaratnam 1972, Warnapala 1972) have helped us to understand some of the underlying political, social and economic causes of the uprising. These articles have dispelled early speculations regarding the social backgrounds of the insurgents, particularly by government apologists and theoreticians from the traditional left parties of Sri Lanka.

suspected insurgents arrested.

One of the most popular interpretations is that the insurgency was caste based, springing from the frustrations of low caste youth. Other explanations were that it was a rural movement, while some speculated it consisted of unemployed youth predominantly from Sri Lanka’s universities. None of these explanations is correct and it is my intention to present some of the statistical information we have on suspected insurgents in order to get a more accurate picture of their social background. Initially I shall examine statistical data from 10,192 suspected insurgents, who have been incarcerated in rehabilitation camps. Insurgents who surrendered or were arrested or captured were interviewed by carefully selected senior government officials and this data was subsequently com­ puterized by the Government Department of Census and Statistics. These are all “suspected insurgents” and one has to be somewhat cautious about this data for the following reasons:

(a) these numbers represent both persons arrested by the police (44.1%) and those who voluntarily surrendered (55•9%). We do not know how accurately these numbers reflect the general composition of the insurgents. For example, a couple of weeks after the insurgency the Prime Minister issued a message to the insurgents urging them to surrender. The message was justifiably interpreted by the public as an amnesty or at least something approaching it. Many persons who were peripherally associated with the insurgency-such as those who attended some of the classes held by the cell leaders-surrendered. In many cases parents urged their children to surrender or brought them along to the police stations, since they felt that this would be a better strategy than being arrested eventually by the police. It is therefore very likely that many “suspected insurgents” were not fully fledged members of the movement.

(b) After the spirit of the movement was broken, the government urged members of the public to provide information on suspects to a special bureau established for this purpose, or to the police. In some instances minor monetary rewards were offered, and in all instances the informant’s anonymity was guaranteed. As would be expected, this led to a spate of information and many persons paid off private grudges, or hostile feelings against low caste persons or personal enemies by informing against them. There is no doubt whatsoever that several innocent people were implicated and subsequently arrested as a result of all this.

(c) We do not know the numbers of persons or the statistical breakdown of the persons killed during the emergency and of the many more executed by the security forces. Nor do we have any idea of the insurgents still operating in the forests, or quietly remaining in their villages. Nevertheless I think the data are fairly satisfactory, particularly in view of the large number of cases. In general I think that while many of these persons may not have been directly involved in the movement, they would in all probability have been in sympathy with it. Table I gives the total numbers arrested or “voluntarily” surrendered.

Simple statistical information is available for the total of 10,192 prisoners; identical information is separately available for the 5700 prisoners who surrendered, but not for those who were arrested. The statistics on suspected insurgents reveal some key in­ formation on the social composition of the insurgency movement. In the first place the movement is unequivocally a revolt of the youth of the country, and those who actively participated in it were predominantly males. Widespread speculation during the period of the insurgency that many females participated in the movement is sim­ ply not true. However, that women even in small numbers, should participate in revolution was a novel enough idea for many persons in Sri Lanka, and dramatic enough to be exaggerated totally out of proportion. Table II shows the essentially youth and male centered nature of the movement.

suspected insurgents in prison

Table II reveals the following information:

(a) 92.8% of the suspected insurgents belonged to the age group 16-32 years.

(b) 71% of the suspected insurgents belonged to even a narrower cluster of 17-26 years.

(c)The modal age for males and females is 20.

Since the statistics are unequivocal in this regard, we have here one of the most fascinating and unique social movements in human history, where the youth of a country organized themselves militarily to topple a government and came close to success.

It makes the youth movements in Western societies seem somewhat pale by contrast. It is interesting to contrast those who surrendered voluntarily with the total (arrested and surrendered). One would expect a large number of younger persons among those who surrendered voluntarily; while those who were arrested would be older, since parents would have greater control over younger people and would have persuaded them to surrender. This is indeed the case, though the differences are not very great: of the number surrendered 96% of the youth are between the ages 16-32 whereas 86.6% of those arrested belonged to the same age group.

The statistical information also reveals that the overwhelming majority of the suspected insurgents were Sinhala-Buddhists. Though there is no separate statistical breakdown for ethnic groups, it can be presumed for Sri Lanka that all those who are classed as Buddhists are ethnically Sinhala, and Hindus are ethnically Tamil. The breakdown in terms of religion is given in Table III.

Again the evidence is overwhelming: 94% of the suspected insurgents are Sinhala­ Buddhists. The next largest group is Roman Catholics who are probably ethnically Sinhala also. One simply has to view the insurgency as a Sinhala-Buddhist movement spearheaded by the youth of the country.

The caste factor

suspected insurgents being arrested by the armed forces

Are the Sinhala-Buddhist youth who participated in the movement representative of any particular class, caste, regional or any other sociological category? At one time the government suspected that it was a low caste movement, and this was explicitly voiced by left-wing theoreticians of the government as a condemnation of the movement as “reactionary.” However, the statistics show that while some caste factors were operative, it was certainly not a low caste movement at all, and in gen­eral cross-cuts the caste issue. Let us consider the ten Sinhala castes listed in Table IV below. We do not have an accurate knowledge of the percentage distribution of castes in Sri Lanka, but a rough estimate can be made from the information available in Ryan’s book (1953) and my own knowledge of Sinhalese ethnography.

(1) What is truly impressive is that the Goyigama, the major Sinhala caste in the nation (both numerically and status-wise), heads the list below, and if we make cer­tain allowances for class and religious affiliation, its percentage representation among the insurgents is higher than its national distribution, as we shall presently show.

According to Ryan the various subcastes that constitute the Goyigama comprise over 50% of the population; I think it is close to 60%.

(2) The Karava, the second major caste, is probably 10-12% of the total popula­tion, and is therefore under-represented in the insurgency sample. However, in evalu­ating the Karava data two factors have to be borne in mind:

(a) The urban elites in Ceylon are almost wholly recruited from the Goyigama, Karava and Salagama castes. Percentage-wise the Karava and Salagama are very strong in this lead. The insurgency movement (to anticipate the latter part of my argument) is not from this class of urban elites.

(b) Secondly, most Christians come from the Karava caste, very likely about 1/2 its total population. Probably the second largest number of Christians come from the Goyigama caste. If then we control (impressionistically) for social class (elite) and religious (Christianity) factors, the following conclusion seems valid. Viewed in terms of national population percentages of castes, although the Goyigama caste is perhaps slightly under-represented in the sample of suspected insurgents, and the Karava caste under-represented therein by about 40-50% !, a different perspective emerges when we control for class (elite) and religion (Christianity). Thus, in rela­tion to the Sinhala-Buddhist non-elite population of the country, the Goyigama and Karava are over-represented in the insurgency sample.

The JVP leader, Wijeweera’s ancestral home.

(3)Viewed in terms of their national population, I believe that the Salagama caste is under-represented in the insurgency sample by about 50%. The Salagama have a large elite representation but unlike the Karava they are almost 90% Buddhist by religion. What then accounts for their under-representation in the insurgency sample?

Several local factors are operative here. The Salagama are a tightly knit caste with a great deal of internal cohesion and well-defined links to caste members among the bureaucratic and political elite. Elite members of this caste have greater control over their fellow caste-men, who consequently could be kept away from movements that threaten the established order. An even more important factor is that Salagama com­munities are sandwiched between Karava communities on the West Coast and there has been a great deal of traditional hostility between the two castes. Since much of the insurgency leadership in this area was Karava, it is likely that the Salagama was not as easily motivated to join the movement which they perceived as Karava.

(4) Two of the low castes, the Vahumpura in the low country, and the Batgama in the Kandyan area seem over-represented in the insurgency sample (10.2% and 9.3% .respectively). I also believe that these castes suffered the greatest number of casualties since some of the fiercest and most prolonged fighting occurred in areas where these castes were localized-in Elpitya in the Western Province (Vahumpura) and in the Kegalle District (Batgama). Moreover, the army commander in one dis­trict was a native high caste person of the area who conducted the campaign with complete ruthlessness against low caste persons. Goyigama folk living in proximity to these low caste people used the post-insurgency period to inform against persons of this caste, so that in some Batgama villages in the Kegalle District the youth were practically decimated.

In general how should one view the caste data? The statistical data suggest that, as far as the nation as a whole was concerned, caste was not a major determinant of of the insurgency. Caste factors were locally and regionally operative, in the relative aloofness of the Salagama and the strong regional involvement of two low castes, Vahumpura and Batgama. However, some caution must be exercised in interpreting the caste data.

(a) We know from other sources that the leadership of the insurgency came pre­ dominantly from the Karava caste. But I would be reluctant to interpret this exclu­sively in caste terms, as for example a reaction against the domination of the Goyigama in national politics. The Karava have always been noted for their entre­preneurship; Karava leadership was perhaps a reflection of the organizational and entrepreneurial skills of that caste. Furthermore, since Wijeweera, the leader of the insurgency, was Karava, he may have used the caste and kinship network to organize the movement in its initial stages.

(b) The caste data have to be broken down in regional terms in order to draw more precise conclusions. For example, the Karava come from the Western and Southern coastal areas of Sri Lanka, and it may be that the Goyigama caste in these regions remained aloof from the insurgency owing to their traditional enmity with the Karava. I have interview data from the Southern Province which, however, makes me doubt this inference.

(c) The secretive, and exclusive youth based nature of the movement also perhaps militated against caste. The youthful insurgents were used to interacting with other castes in their schools, and were perhaps more tolerant of caste than their elders. Moreover, the movement tapped the idealism of the youth which may have helped them to transcend caste allegiances. It is also likely that the cells, secret meetings, indoctrination sessions brainwashed them of their caste identities for a higher and nobler goal, as they perceived the movement. Caste would have been much more important had the J.V.P. (Janatha Vimukti Peramuna, “Peoples’ Liberation Front”) been an “open” movement, drawing its recruits from a wider age range.

The occupational background of the insurgents

weapons used by insurgents

Thus far our conclusions are that the insurgents are mainly Sinhala-Buddhist youth belonging to the major castes in Sri Lanka. We suggested, incidentally, that they do not come from the ranks of Sri Lanka’s elite ruling class. I turn next to the socio-economic backgrounds of the insurgents, which I believe, will help us understand more thoroughly their economic discontent. The data I will use are the breakdown of the occupations of the insurgents, the occupations of their fathers and their educational backgrounds. Table V gives the occupational breakdown of the suspected insurgents. It should be noted that this table gives the true occupational background rather than the caste occupational background. For example, fishing is the means of livelihood of the 25 persons listed as fishermen, and not simply their caste status. Similarly barbers are not a caste but an occupation.

When these statistics were made available to the government there were cries of jubilation from its left-wing theoreticians: after all, a majority of these insurgents were employed, ergo they had no real reason for insurgency, and therefore it was a “reactionary” attempt to oust a truly socialist government. However, these statistics are certainly not that opaque, and a careful interpretation will bring out an extremely disconcerning picture of massive and widespread public discontent.

(I) Firstly, only the first 5 occupations on our list truly belong to the elite, occupying the top positions in the professions or bureaucracy. They constitute only 0.2% of the total. Thus the country’s elite is virtually not involved in the insurgency.

(2) Take the middle rung administrative positions, and other stable occupations. The occupants of these positions are in general extremely conservative, have styles of life modelled on the elites, and have aspirations for their children to eventually join elite ranks. They are 6-IO on the list, constituting 6.1% of the sample. This “lower middle class group” is again poorly represented among the insurgents. Iwill discuss their role later, only suffice here to state that about 66.3% of them are under thirty.

(3) There is a large number of persons who either occupy low positions in the government (“underdogs”) or are self-employed in not very remunerative positions; of those in occupations II-I8, constituting 21.4% of the suspected insurgents, 82% are under thirty years of age.

(4) The blatantly underemployed category (20-22)-intermittent and poorly paid employment-constitute 39.6% of the sample. Ihave included cultivators here since they are often virtually landless “tenant cultivators” or farm laborers with no regular employment who call themselves cultivators, instead of the derogatory term “laborer” or “unemployed”. Of this group 88% are under thirty.

The name board for Hambanthota town renamed by the insurgents as” liberated town”. Bala Tampoe, Prince Gunasekere and Lord Avebury who apeared for Human Rights of insurgents are here.

(5) There are the totally unemployed (17.5%); among whom about 97% are under thirty; and a related group of students (I2.5%), all of whom are under thirty. That is, as we go down the occupational series the gross numbers involved in the movement as well as the percentage of young persons (under thirty) tend to increase. Socio-economic position is correlated with age: the lower the socio­ economic position the greater the numbers of younger persons involved. This may, however, partly reflect the lower average age of these occupational groups themselves. The occupational data are summarized in Table VI.

On the basis of the data in Tables V and VI I draw further conclusions: the category of "unemployed" ( 17.5%) is deceptive. Only 6.3% (elite and middle positions) had secure employment with reasonable financial rewards. The rest were either poorly employed, underemployed or unemployed. Students were not only technically unemployed; their actual and self-perceived chances of employment were remote, so that after leaving school they will probably move into the category of unemployed or underemployed. Thus over 90% of the sample were in un­prestigious or economically unrewarding positions.

The inverse correlation between youth and occupational position reflect: (a) larger demographic pressures where the employment structure simply cannot cope with the increasing numbers seeking employment and (b) that these persons are being funnelled out into the hard world through the country’s educational system. The 1270 students are simply those waiting to be funnelled out into the categories above them, crowding the lower reaches of the employment (or unemployment) structure (see Kearney 1973 for more details regarding the educational problem in relation to the insurgency).

A comparison of the social positions of fathers and sons shows no real differences in the first three categories, but shows some interesting differences in the rest. The unemployed group in the father’s generation is only 2.5% (260) while in the son’s generation it is 17.’5% (and students 12.5%). This remarkable difference can be explained if we look at the category “unspecified” in the parents’ generation: 2214 or 2I.7%, a category which does not exist in the generation of the sons. I strongly suspect this unspecified category is an “underemployed or unemployed” one, prob­ably the former, made up generally of low caste people or of high caste ones with no stable jobs, so that the sons who were interviewed simply did not state the nature of their employment. If this interpretation of the data is correct, the discrepancy in the lower reaches of Table VII is bridged so that a collapsed category “unemployed­ underemployed” would show 8o% in the sons’ generation and 73% in the fathers’. In general both parents and sons belong to the low socio-economic groups, the parents somewhat lower than the children when it pertains to middle and lower middle positions but better off than their sons when it comes to the truly unemployed category. This is perfectly expectable since economically poor parents were educating their sons to occupy stable positions in the job and occupational structure. Some of them succeed, but the increase is only +1.6% in the lower middle positions, +0.6% in the middle positions and -0.2% in the elite positions. If the data are indicative of true trends, if parents hoped that their children would get stable jobs of amy sort, the true increase over the parental generation in regard to these jobs is only +2.0%! The generation of the sons is doomed to massive underemployment or unemploy­ment.

While both fathers and sons belong to similar economic and occupational levels, it is likely that they have radically differing views of their life situation. The parents are reconciled to their economic positions. They have a model of status mobility through their sons who will obtain an education and will move out of their present socio-economic level. The sons, who have cathected different goals for aspiration, have realized subjectively the near impossibility of achieving them, and are con­sequently reacting against their economic situation.

The educational background

Lord Avebury questions family members of Premawathie Manamperi, the killed female insurgent of Katharagama

Examination of the educational background of suspected insurgents reinforces the earlier view that they are village, non-elite youth. Table VIII gives the schools attended by them. Several interesting conclusions emerge from Tables VIII and IX:

(a) nearly 80% of all suspected insurgents come from the type of schools known as Maha Vidyalayas. An additional 64% come from Madya Maha Vidyalayas, which are schools of the same type. These are generally high schools serving village youth, and it is obvious that the education supplied in these schools is in some way an important factor in the insurgency. They were also probably important recruiting centers for the insurgency, for it is likely that the age category 15-19 consisting of standards 5-8 and GCE (o) level are currently enrolled in these schools, (or in the latter case have just left these schools). Their numbers are 2083, i.e., 20.4% of the total.

(b) Tables VIII and IX also show that only a few (4-0%) come from the technical schools and universities. Though the numbers here are small, I have other evidence to show that the Universities provided a major source of the leadership of the move­ment.

(c) Table VIII shows that only 2.4% had not attended school, and this is confirmed by Table IX which shows that 2.0% had no education. About 17% had a “grade school” education.

Consider some of the implications of this data. The suspected insurgents are a deprived group, but not a depressed group like a lumpen-proletariat. Potential revolutions are not created by demoralized, anomie groups: millenia! movements may come from such a source. True illiterates form only a very small percentage and the poorly educated are only 17% About 80% have a reasonable education, and had both the motivation and the capacity to participate in what they viewed as a rational solution to the political and economic ills that beset the nation. Con­sider the large numbers (37%) that were in school or just out of it. They were products of mass free education (and all of them were young persons born after independence). The schools they were attending were the recruiting grounds and leaders of the movement surreptitiously held “lectures” and discussions in these schools. The grievances that these students had were real, and could easily be high­ lighted by “cell leaders”. Indeed the traditional leftists in Ceylon did this themselves, for they too in public speeches pinpointed the inadequacy of this type of schooling. Schools had poor laboratory facilities and students were generally trained for arts courses, which provided them with no technical expertise. Many of them would simply not be able to enter University either due to lack of vacancies there, or financial difficulties. Even if they did complete a university degree there was no guarantee of employment. Thus if they were not already aware of the bleak future that awaited them, they could easily be made conscious of it. Their youth and idealism could then be tapped for a new movement that held some promise for the future.

Part II

The barbarians at the gates

The arrested JVP Leader

The data presented in the preceding section have a further implication which I wish to explore here. That is, the insurrection can be seen as an attack, not only on the government in power, but more importantly, on the elite as a class, whose ranks furnish the political and bureaucratic leaders of the country, irrespective of their political or ideological commitments. The insurgents were not peasants, and the Ceylon insurrection was not a peasant revolt. They were at best the sons and grandsons of peasants. The leadership of the J.V.P. in particular came from a village aristocracy of the type I have described elsewhere (Obeyesekere 1967). Twenty years ago these leaders would have been comfortably assimilated into the elite, but in 1971 such assimilation was well nigh impossible. Let us examine why.
It is well known that one of the features of colonialism everywhere is the development of an indigenous elite, who were initially recruited to fill the minor positions in the colonial administration. Later the pattern of recruitment widened and after independence the whole government bureaucratic apparatus and the political system were controlled by them. The elite generally consisted of politicians, an administrative and professional elite including the security services, and a busi­ness elite either having direct control of economic wealth or holding business appointments in large European owned enterprises, particularly the tea plantations. It was a ruling class holding access to political and economic power and patronage. In Sri Lanka it was not internally homogeneous, but split on major ethnic lines or religious lines (Sinhalese vs. Tamil, Buddhist vs. Christian) and even on lines such as caste. The most serious division was the ethnic one but the other lines of cleavage were also operative. But overriding these factional cleavages was a common style of life, which may be termed “aristocratic,” (partly based on the aristocratic styles of traditional feudal Sri Lanka, particularly notions of power and prestige). As far as Sri Lanka’s elite was concerned one crucial feature of this life style was the use of English a11 the major language in the home and also as the language which the elite used for social interaction. The school system too was a powerful ground for secondary enculturation of the elite. There were the privileged schools modelled on the English “public school” system and located in Colombo and some of the major provincial capitals. These major schools formed a kind of old boys club: and intercollegiate cricket matches, old boys associations were highly important social gatherings for these people (see Fernando 1972 for an illuminating discussion of Sri Lanka’s elite). Once caste and ethnic lines were drawn, marriages tend to be endogamous, though hypergamy and hypogamy were also used as mechanisms for “outsiders” to establish alliances with the elite.

The political leaders of all the major political parties-left and right-were members of this class with their early education in one of the elite schools and their later education in Oxford, Cambridge or the London School of Economics. The physical locus of the elite was in Colombo: the administrative and business center; and with a thin upper crust in the other major cities. Colombo like other Asian cities was an unreal city; its physical and social character marked it out dramatically from the rest of the country. The network of elite interaction was largely confined to that city. The ideal of public servants and professional people was to live in Colombo and even today sending one’s children to Colombo schools is a major concern of elitist parents. Top provincial administrators and professional people who have to work in the outstations often have their wives and children in Colombo or at least their children are boarded there. Through time the social isolation of the elite from the mass of population was paralleled by their physical isolation in the city of Colombo and its immediate suburbs.

The ruling elites in a meeting. Sirimavo the leader of left wing SLFP and Prime Minister, J.R, the leader of right wing UNP and the leader of opposition and DR. Colvin , the leader of Trotskyist LSSP.

Historically elite recruitment came from three major sources: (1) elite ranks themselves (2) the various caste groups (Salagama and Karava) that rose to eco­nomic prominence in the early period of British rule (3) the village aristocracy of minor colonial officials, such as headmen, who amassed a fair amount of wealth and could educate their children in elite boarding schools. The major source of secondary socialization for the children of the elite were the privileged schools mentioned earlier, the University of Ceylon, and English Universities for those who could afford them. The educational system was crucial for obtaining coveted bureaucratic positions and for the professions. Elite schooling combined with patronage was all that was essential for high school “drop-outs” to obtain highly paid jobs in business and the plantation industry. It is one of the greatest achievements of this patronage system that some of the least educated persons with the right school tie and the right connections could occupy some of the best paid jobs in the country, e.g., super­intendents of plantations.

I think one could conveniently call this group a “ruling elite” in that it had virtual control of the political machinery, government, business and the professions. The life style of the elite-and entry into its ranks through the educational (or marital) system provided a powerful source of motivation for those below them, and also a source of emulation. Who were those below them? Two significant groups should be mentioned: the middle ranks of the administrative hierarchy, teachers, technical personnel, and educated people in the village (headmen, school­ teachers) who could afford to send their children to elite schools. These middle ranks (category 2, in Table VI) were either “drop-outs” from above or those who had moved up from below. Even if they could not achieve elite goals, they emulated (in lowered tones) the elite life style, and they had high aspirations for the children. A major concern of this “middle class” was to get their children into “good schools” so that they would join the ranks of the elite. The ideal position was the coveted Ceylon Civil Service, the top echelon of the government bureaucracy.

Let me highlight one major consequence of this elitist system. One could view the major political parties sociologically as factions of a ruling elite. Take the major political parties moving from right to left: the United National Party, the Sri Lanka Freedom Party, the Lanka Sama Samaja Party (Trotskyites) and the Com­munist party. Party manifestos and ideologies might be radically different from one end of the spectrum to the other but the leadership of all these parties came from elite ranks, almost without exception. They came from the same schools, went to the same clubs, spoke English, and marriage alliances cut across political differences. The Communist and Trotskyite leadership was not (and is not) a “disprivileged” group: they spoke for the “disprivileged” but were themselves privileged, and lived an unabashedly high style of life. Close relations, siblings, and even parents, of this group might belong to right-wing parties, and a social network connected them with the sources of power and privilege. It is no accident that left-wing leaders could not mobilize a peasant organization: their class position and physical isola­tion militated against it. They were successful in building up working class trade unions; though here again the social gap between the elite leadership and the workers themselves was immense. Thus all Sri Lanka’s political parties could be viewed as factions of the elite striving to gain political control of government. Nevertheless, one must emphasize that these were clear-cut factions based on ideological differences.
Sri Lanka, with a dominant ruling elite, was till about 1956 an extremely stable country, a model democracy in Asia. However, this stability could not last very long for many reasons, some of which I shall examine here.

(1) Until around 1955 the educational system was crucial for entry into elite ranks, but with the development of free mass education this was no longer true. Universal free education started in 1945, and its impact began to be felt after 1955. By 1¢o practically every village or group of villages had its school, so that now we have near universal literacy in Sri Lanka. High schools known as Maha Vidyalayas and Madya Maha Vidyalayas were established everywhere in the villages, and sons of peasants were able to get a high school education. By 1960 sons of well-to-do peasants were entering the several newly created universities and technical and teacher training institutes. Thus by 1972 the earlier situation had changed radically: children of the elite now constituted only a very small proportion of university entrants. Neither the parents of the village students, nor the students themselves were satisfied with the station of life into which they were born. Their aspirations were to occupy upper positions in the governmental structure if they entered the Univer­sity, or lower but respectable jobs if they were high school products.

But though the school population expanded dramatically, the economic oppor­tunity structure remained stagnant. There were limited jobs for the hundreds of thousands who were aspiring for them. This had an immediate impact on the political patronage system: since jobs of any sort were scarce, competition for jobs became extremely acute, and jobs increasingly became dependent on political patron­age. By 1970 competitive examinations had become a virtual farce, and even extremely low positions such as office peon, driver, temporary clerk, or even casual laborer were appointed by high-ranking politicians such as Ministers of State. Access of village youth to the decision-making heads was through party henchmen and political mediators on the village level, through to the M.P. of the district, and from him to heads of government departments and Ministers. Educational qualifications which were at one time paramount ceased to mean anything: there were too many qualified persons for practically no jobs. Economic pressures then forced persons to drop their aspiration levels so that there are now University graduates in menial occupations. This does not mean, however, a decrease in frustration. On the contrary, a conscious dropping of a highly desired aspiration level probably leads to increased frustration. It is this phenomenon that accounts (at least partly) for the clustering of insurgents at the lower levels of the occupational structure: they are there because they have simply nowhere else to go.

(2) Simultaneously with the problems noted above there was an increasing political consciousness of the electorate. It should be remembered that Sri Lanka was the first country in Asia to obtain universal franchise. This in combination with increasing literacy and education made Sri Lankans an extremely politically conscious community. The Trotskyite and Communist parties were largely respon­sible for the early politicization of the urban proletariat, and the rural peasantry came into political prominence by the massive and dramatic victory of Mr. Bandara­naike in 1956 over the conservative UNP. The appeal of Mr. Bandaranaike was that he could raise the self-esteem of peasants and appear as a champion of their interests. Peasants had the vote but lacked an organization to articulate their in­terests and draw them into the political arena. Bandaranaike’s success was due to the fact that he used a village aristocracy-Ayurvedic physicians, monks, various influential officials in the village-to mobilize the peasant vote. In 1956 the peasant vote was systematically mobilized, and this resulted in his sweeping victory. This was the first real breakthrough in the politicization of the peasantry. Soon Bandara­naike’s S.L.F.P. had built a kind of grass roots organization-led by the aristocracy I mentioned earlier-in the villages. Since then Sri Lanka has had a highly politically conscious peasantry: one clear manifestation of this is that almost all local govern­ment elections in the villages are now run on party lines, and kinship and other factions in the villages became rationalized and articulated in terms of party ideology and affiliation.

JVP rally in Colombo in1970

(3) This high level of political consciousness accounts for the fact that of the parliamentary democracies in Asia, Sri Lanka was the only country that consistently threw out governments through a popular vote so that no government has had more than two consecutive returns to power, and most governments have had only one term of office. However, while the peasants were politically articulate, they had no access to political power.’the classic paradox of the Marxist civil society was operative here in extremis. The only occasions where peasants could exercise their power was at an election; but then often, irrespective of the party in control, effective political power remained (as it still remains) in the same ruling elite. In other words, political consciousness is widely diffused but political participation and decision-making remains in a ruling elite. None of this would really matter if the opportunity structure were flexible; but jobs and other forms of privilege were dependent on access to the centers of political power from which these persons were debarred.

A good illustration of this problem came out in casual interviews I had with young men (not insurgents) in an area where insurgent activity was very high. Among them I interviewed a group of five persons who were all high school graduates (GCE ‘O’ level), unemployed for the last three years. All had occasional sporadic employment, and all of them assisted their parents in their agricultural work. But their parents owned only small units of paddy land, not even sufficient for their own subsistence and hence these young men were mostly unemployed. One person passed out of high school three years ago. He applied for many jobs but did not get any. Eventually he worked as an unpaid apprentice in a garage in a nearby town. He worked there for a year and, having acquired some mechanical skills, applied for a technician’s examination held by the state owned bus service (the C.T.B.). He passed this exam and was called for the interview. Suddenly without any provocation, the results of this examination were “cancelled”. But the appoint­ments were nevertheless made, not by examination, but on the recommendations of the M.P. of the government in power. Since my informant had, like many youths in the area, worked for the present government, he approached the M.P. to try to get the job. But it transpired that the number of positions allocated to the M.P. had already been filled-by young men who were in fact political opponents of the M.P. Rightly or wrongly, the youths I interviewed felt that the M.P. had been bribed by persons of wealth in the area.

The story is utterly typical. Since jobs are scarce, all competitive or open methods of recruitment have been abandoned and the government M.P. of the area is given tremendous power in these appointments. In the appointment of teachers, for example, each M.P. is given a fixed quota to fill. But very often the persons who get the job are kinsmen of the M.P., or those who have access to the patronage network through their elite connections. In some instances the wealthy of the area who are politically opposed to the M.P. in fact get the job, owing to bribery or the capacity to manipulate the patronage system. Naturally the young men I interviewed har­bored a felt sense of injustice, a feeling that is extraordinarily widespread among the unemployed youth. This I think was an extremely important motive for young persons to join the insurgent movement.

Hardening of class boundaries

the LSSP leaders Dr. N.M. , Bernard Soysa , Prof. Doric De Souza and Reginald Mendis.

I do not see the peasantry as a class in any meaningful sense, but I do see the elite as one. I noted previously that political parties in Sri Lanka could be seen as factions of an elite. A common style of life, marriage alliances, and kin networks cut across the political factions. These overarching cultural and social factors make for easy changes in party allegiances, so that the movement of individuals from one party to another is relatively easy. This is true even of the Trotskyite and Communist parties: crossing of party lines has been an easy and frequent feature from extreme left to extreme right. Only a small, hard core of party leaders have stuck it out, and even these people could easily link up with a party like the S.L.F.P. (with a strong right wing). I am, of course, not suggesting this is the only factor involved, but a common life style and interaction patterns outside of party ties facilitate and make possible both the continual crossing of party boundaries, and the formation of po­litical coalitions between ideologically incompatible groups. Indeed one of the fascinating problems deserving of further study is the inexorable process whereby young dedicated left-wing radicals of the forties were gradually assimilated into the elite life style, and the process whereby their political ideologies (Marxism) were transformed and made consonant with their class interests.

With the increasing problems faced by the society-particularly the scarcity of jobs-the overarching cultural factors and sociological networks become even more important. Irrespective of who was in power, the prestigeful positions went to those who came from elite ranks; and even minor jobs went to those who can establish connections with them.

Rarely could a village lad, even with a B.A., get an administrative job in a firm or large business because of his poor knowledge of English. In spite of lip service paid to the National Language, the same restriction applies to jobs in government-it operates in many devious ways. Many students in the Universities are quite aware of it. The few jobs available then, in general, go to the elite because of their social ties with persons holding power and because they may have special cultural advantages over the others (e.g., knowledge of English). The upshot of all this is the strengthen­ing of the overarching social ties among the elites and a weakening of political (factional) boundaries. For example, a Minister, a former Trotskyite who later crossed over to the S.L.F.P., and is viewed as one of the left leaders in the S.L.F.P., married his daughter to a wealthy businessman from the right-wing U.N.P. who owned a factory for assembling a Japanese product. Not long afterward, another Minister from the Trotskyite L.S.S.P. ordered government corporations to buy this product at three times its true value. This is not corruption in the simple sense: no one got a bribe I am sure. It simply reflects the operation of the extra-political ties I mentioned earlier. This kind of behavior is not unique to the present regime in Sri Lanka; on the contrary it was characteristic of all the regimes Sri Lanka has had since her independence.

The examples given here illustrate one critical feature of the extra-political ties that link the elite together: kinship and marriage. Numerically the elite is very small, but marriage and kinship shrink them further into smaller factions owing to ethnic and caste endogamy. This means that the “marriage market” is a small closed one; spouse selection and marriage alliances unify individuals and groups who are politically divided into larger, cohesive, extended kin groupings. This is especially true of the numerically small castes, such as Karava and Salagama, where persons who are politically divided often manifest extraordinary kinship loyalty and solidarity. These kinds of kin-links, and the intimate nature of elite interaction partly account for the extreme personalization of politics in Sri Lanka.


Within a year of its eruption the Government had the insurgency well under control, but the process of alienation of the elite from the rest of the country has accelerated further, resulting in the hardening of the elite class boundary. The most vociferous condemnation of the insurgency came from the traditional left (C.P. and L.S.S.P.). The effect of the external threat was to make political divisions within elite ranks fuzzier and their social links stronger. Appointment to jobs nowadays is based on social rather than on political connections. Along with the increased solidarity within the elite is their increasing physical isolation in the city of Colombo. Colombo-centrism has increased. Immediately after the insurrection members of the elite located in the outstations pulled into Colombo, and this trend has continued, increasing the unreality of Colombo as a city. In the city social and political life goes on in isolation from the rest of the country. Confronted with massive economic and social problems, elite bureaucrats, planners and politicians seek bizarre solutions. Planners, for example, are preoccupied with paper plans impossible of implementa­tion. Posed with what seem unsolvable economic and political issues, politicians and bureaucrats are increasingly using the psychological mechanism of denial. The social effect of “denial” may be called picayunism, an overwhelming preoccupation with trivia, irrelevancies, elaborate paper plans, and social legislation that cannot be implemented.


ARASARA TNAM, Sinappah, 1972, “The Ceylon Insurrection of April 1971: some causes and conse­ quences,” Pacific Affairs, Vol. 45, No. 3, 356-371.

EGAN, Michael, 1972, “Some reflections on the 1971 Ceylonese Insurrection.” Paper read at the Canadian Sociology and Anthropology Association Meetings, Montreal, May 28-June 1, 1972.

FERNANDO, Tissa, 1972, “Elite politics and revolutionary movements in Ceylon: a sociological perspec­ tive.” Paper read at the Canadian Sociology and Anthropology Meetings, Montreal, May 28-}une 1.

KEARNEY, Robert, 1973, “The 1971 Ceylonese uprising: an ecological interpretation of its origins and fate.” Paper prepared for the Research Conference on Communist Revolutions, Estate Carlton Beach Hotel, St. Croix, V.I., January 24-28, 1973•

OBEYESEKERE, Gananath, 1967, Land Tenure in Village Ceylon. Cambridge University Press.
RYAN, Bryce, 1953, Caste in Modern Ceylon. Rutgers, The University Press.
WARNAPALA, W. A. Wiswa (Politicus, pseudonym), 1972, “The April revolt in Ceylon,” Asian Survey,
Vol. 12, No.3, March, 259-274.