Caste in Jaffna

Prashanth Kuganathan** whose title runs thus: “Social Stratification in Jaffna: A Survey of Recent Research on Caste”

A SYNOPSIS: Since 1983, war has dominated the perception of Sri Lanka. This has affected scholarship on the country, such that the subjects of an overwhelming number of research proposals and publications have been on the war and the prospects and prescriptions for peace. This survey paper is an attempt to locate the system of caste in transition in the Jaffna Peninsula by reviewing recent literature written after the commencement of the war. While detailed ethnographies of caste in Jaffna may have temporarily come to a halt, caste practices have not and remain a salient part of everyday life among the Tamils in Sri Lanka. As the war ended in 2009, it is therefore important that social scientists on Sri Lanka revisit the topic of caste, that is an integral part of not just Tamil culture or society, but being Tamil itself. As the study of caste is dominated by research in India, a microanalysis of Jaffna and Sri Lanka, particularly the nuances of this system in transition due to war and militancy, could contribute to the macro-study of caste at a sub-continental perspective.


Two thousand and nine was a watershed year in Sri Lankan history. In May of that year, the Sri Lankan Armed Forces, under the command of President Mahinda Rajapaksa, successfully annihilated the Liberation Tigers of Tamil Eelam (LTTE) – simultaneously killing its leader, Velupillai Prabhakaran (1954–2009). Prabhakaran had spearheaded a rebel movement for an independent Tamil homeland in the northern and eastern parts of the island against the Sri Lankan state for over a quarter of a century. During this time period, Sri Lanka and war essentially became synonymous, with war dominating both the internal and external conceptions and perceptions of the country.

A core structure in transition has been that of caste boundaries and identity. Over 25years of war and militancy have altered systems of kinship and social formation, as well as transformed conceptions of nationalism and shifted allegiances in Sri Lanka, particularly in the conflict-affected north and east. The mechanisms of war and militancy have been able to alter, even for a short period of time, who one may marry, who they associate with, how they choose to identify, and how they may subdue the innately inherent element of caste.

This survey paper is an attempt to locate the system of caste in a post-war environment in the Jaffna Peninsula, known to be the most caste conscious and conservative region in the country (Pfaffenberger 1994, p. 162). As this journal is concerned primarily with recent literature, this article will only discuss caste research concerning Jaffna published after the commencement of the war. However, in order for non-specialized readers to grasp a complete historical understanding of caste in Jaffna, it is important for them to also look at Arasaratnam 1981, Banks 1971(1960), Casie Chitty 2004(1934), David 1972, Pfaffenberger 1981, Pfaffenberger 1982, Raghavan 1971, and Skjønsberg 1982.

The politics of naming the oppressed and a brief description of some castes in Jaffna

It is important to mention that in Sri Lanka, there is no unanimous consensus on the name or term by which to collectively refer to the various castes formerly known as untouchable, a term of condescendence, that cuts right to the core of one’s dignity. Various works have used a variety of terms over a period of over 50years. Michael Banks (1971[1960]) uses “Untouchable” (with a capital letter “U”). Kenneth David (1972) uses the Tamil term “korenja cati,” which literally translates to low caste (p. 66) or diminutive caste, a term just as condescending as untouchable. While Bryan Pfaffenberger (1981; 1982; 1994) also uses “Untouchable,” he uses the term “Minority Tamil” as well. The term “Minority Tamil” expresses the fact that Tamils from castes considered untouchable constituted the minority in Sri Lankan Tamil society. As this term predates Sri Lanka’s contemporary ethnic and political strife, the term “Minority Tamil” today may easily be conflated for Tamils in general, including those considered upper caste, who occupy a subaltern status vis-à-vis the Sinhalese majority of whom some perpetuated anti-Tamil violence, sometimes condoned and at other times coordinated by the Sri Lankan post-colonial state and its anti-Tamil policies.

Raghavan (1971) uses the at.imai/kut.imai1 dichotomy to divide Jaffna Tamil society (pp. 166–167). These terms, however, do not refer collectively to castes that have previously faced untouchability. At.imai and kut.imai refer to two categories of collective caste groups employed by the Vell..ā.lar2 caste, the largest and most dominant caste group in the peninsula, comprising over 50 percent of the total population (Banks 1960, p. 67). The Vell..ā.lars are landowning agriculturalists by traditional caste occupation. Raghavan (1971) describes the at.imai, who are also known as kut.imakkal, as castes “attached” to the Vell..ā.lar, as they were their former slaves (p. 167). The at.imai are comprised of the Kōviyar (domestic servants specifically to the Vell..ā.lar), the Nal.avar (toddy tappers and tree climbers), and the (agricultural laborers) castes. The kut.imai, on the other hand, were not private slaves to any family and served society in general. The kut.imai are comprised of the artisan Goldsmith, Blacksmith, Brazier, stone-mason, the (barbers), the Van.n.ār (dhobis), and the Par-aiyar (funeral drummers) castes. Only the, the Nal.avar, the, and the Par-aiyar castes were formerly considered untouchable in Sri Lankan Tamil society (Silva et al. 2009, p. 56). As they are split between both the at.imai and kut.imai groups, both of these terms do not specifically imply untouchability. The at.imai category may undergo some sense of subjugation, but this is definitely minimal in comparison to the kut.imai category, particularly with the artisan castes, which are considered intermediary in status and not low.

Pañcamar is a term used to collectively refer to the five castes of Van.n.ār,, Nal.avar,, and Par-aiyar. While it may not be widely used as a term of self-identification, Jeyarajah (2001) uses it in his writing and self-identifies as Pañcamar (personal communication, November 14, 2011). I use this term in this paper, as it has been used in Sri Lankan scholarship on caste (Jeyarajah, 2001; Silva et al. 2009); it has a sense of locality and is considered an inoffensive and somewhat neutral word to refer to the five castes formerly considered “low” in Jaffna society. “Otukkappatt..avarka.l” or “otukkappatt..a camukam,” both of which loosely translate to “those who have been set apart,”3 has been used in Jaffna by anti-caste activists in the past and the present to refer to these groups. These terms, however, were used by members of the Eelam People’s Revolutionary Liberation Front (EPRLF); as they were a paramilitary group, some feel otukkappatt..avarka or otukkappatt..a camukam to have political connotation or implication. In this paper, therefore in addition to the use of Pañcamar, I also adopt the term “depressed castes,” which Silva et al. 2009 use, and

“oppressed castes,” both of which are similar to otukkappatt..avarkal..

Today in India, Ambedkar’s term Dalit has become the approved and accepted appellation, most importantly, by the members of these traditionally marginalized and oppressed groups themselves, as well as the normative term of reference used by national and international scholars. The Puducherry based activist and politician Ravikumar (2002) uses the term “Dalit” when referring to these groups in Sri Lanka, which may be somewhat problematic as it is appropriative, since most of these groups do not use it as a label of self-identification. This new name is rarely used in Sri Lanka (Silva et al. 2009, p. v); however, in a recent field visit to Jaffna in the summer of 2013, I noticed its usage was on the rise, particularly by certain political elements that were mobilizing. The term Dalit then becomes an important term, perhaps not locally, but in forging certain sub-continental political alliances with other entities across the Palk Strait, who have undergone similar experiences of oppression and are better organized. It also facilitates for comparative academic study and discourse – so while I am careful to limit my use of the word Dalit, I do not necessarily shy away from it in the pages that follow.

The colonial period

While the system of caste is derived from ancient Hindu texts such as the Purus.asūkta4 and thus sanctioned by religion, caste as it is perceived today has undergone evolutionary changes over time, just as the progress of the term Dalit itself. It is important to comprehend the nuanced influence the colonial period in particular has had on the perceptions and lived realities of contemporary caste manifestation. In his article “Caste as a social category and identity in colonial Lanka,” John Rogers (2004) mentions that caste was viewed by colonial administrators as “prejudices of the people,” which led to the argument for its official nonrecognition and eventual policy of repudiation (p. 63). He goes on to explain that this policy countered that of the administrators in neighboring India, where caste was not only used for enumeration purposes but also exacerbated by the regime (p. 64). Historical anthropologists such as Bernard Cohn, Susan Bayly, and Nicholas Dirks have argued that processes undertaken by colonial officials in regard to the collection of caste-specific data and ranking have in fact idealized and solidified the conception of caste as it is known today in India (Bayly 1999, p. 4; Cohn 1990[1987], p. 248; Dirks 2001, pp. 8–9). Rogers (2004), however, highlights the fact that from the time of the inception of the modern census in Ceylon in 1871, no census has collected caste specific data (p. 72). In contrast, India has been able to harness this widely available data and apply it to constructive policies, such as reservations for Scheduled Castes and Scheduled Tribes. While the constitution of Sri Lanka outlaws caste-based discrimination (Silva et al. 2009, p. 15), no progressive caste based legislation has been implemented to directly uplift or empower depressed castes.

In a similar train of thought, B. E. S. J. Bastiampillai (1988) has argued in his article, “Caste in Northern Sri Lanka and British Colonial Administrative Practice in the Mid 19th Century: Compromise and Expediency,” that the British played a crucial role in eliminating caste as an official category of inhabitant classification from 1845 (p. 50). Paradoxically, it was during this colonial period that “the dominant caste was even strengthened and made use of by the British colonial power” (p. 48), which points to Vell..ā.lar domination in intra-Tamil community affairs, and Tamil domination in inter-ethnic affairs – largely due to colonial policies of divide and rule. Interesting are his observations in regard to caste and Christianity. While overt caste distinctions were eliminated through the colonial government, the Sri Lanka National Archives find that the Government Agent did not intervene in the late 1850s when two caste groups of Roman Catholics argued over the burial of their dead, stating “no wish on the part of the government to interfere with native customs…” (Bastiampillai 1988, p. 51). The church also seemed to exercise minimal protest over discrimination against students from oppressed backgrounds by those considered upper caste, possibly due to the repercussions missionaries would face had they upset or alienated the dominant Vell..ā.lar community. Bastiampillai (1988) writes of a pupil of the Van.n.ār caste who had to be withdrawn from a Wesleyan Methodist school due to opposition of his enrollment by those considered upper caste (p. 53). Similarly, he also writes of students considered upper caste who dropped out of a Wesleyan school when a student from the Nal.avar caste was admitted. Despite the introduction of western notions of egalitarianism through the high level of education prevalent in the peninsula, adherence to caste distinction was rigidly upheld (p. 51). While it was the belief of the state that universal, free education would extirpate caste inequality (Silva et al. 2009, p. 15), Bastiampillai (1988) argues that neither education nor conversion to Christianity could eradicate caste prejudice among Tamils (p. 58).

A subaltern perspective on how Pañcamars were treated by Vellālars and Christian missionaries has come from Benjamin Jeyarajah, a pastor with the Church of the American Ceylon Mission (CACM) in Jaffna. Jeyarajah (2001) argues that the Church, specifically the American Missionaries, not only adopted caste practices but propelled them into continuity by perpetuating favoritism of the Vell..ā.lar (p. 20). While he acknowledges that education did not reach oppressed caste communities until the entrance of the Christian missionaries (p. 12), he argues that the American Missionaries primarily educated people considered upper caste (p. 30). Furthermore, he claims that children from oppressed caste communities were still subject to humiliation and indignity, some having to sit shirtless at the back of classrooms (p. 12). His theological argument is that throughout history, the church has never stood in solidarity with “these marginalized people for their development including their spiritual growth” (p. 2) and that the Church needs to follow the gospel that itself preaches (p. 41) and not sanction caste distinction, particularly caste discrimination in the church. The element of caste was inherently woven into the fabric of Tamil culture that even Christian missionaries could not extract it from their converts.

One cannot, however, blame the missionaries entirely for an element that is inherent in Tamil culture. The Pañcamars may have overestimated the powers of missionaries. It is important to point out that the hierarchy of the Church of South India (CSI) to which the Jaffna Diocese belongs to, was, and continues to be Vell..ā.lar dominated. It was somewhat inevitable for the missionaries to favor the Vell..ā.lars dominated the political and social realms of Jaffna life, alienating them could indeed have threatened the stability of missionary activity. It is no surprise the Vell..ā.lars would definitely have numerically dominated the schools due to their proportion of population in Jaffna and their influence in society. However, despite Jeyarajah’s argument that students primarily considered upper caste went to these Christian schools, many Pañcamars, including Jeyarajah himself, patronized these institutions in Jaffna following the 19th century and are also the products of a western missionary education.

Temple Entry

The 1960s saw some change in regard to caste discriminatory policy, particularly in the Hindu religious arena, where caste originally stemmed from and was practiced most conservatively. In his article “The political construction of defensive nationalism: The 1968 temple entry crisis in Sri Lanka,” Pfaffenberger (1994) argues that the caste agitations and riots at Maviddapuram, one of Jaffna’s most sacred shrines, were not only a victory in regard to caste discrimination but also a “stepping-stone on the path to the creation of a decidedly modern Sri Lankan Tamil identity” (p. 162). He gives the reader a brief overview of some discriminatory restrictions imposed upon oppressed castes by the Vell..ā.lars:

In Jaffna in the 1940s and 1950s, for instance, Minority Tamils were forbidden to enter or live near temples; to draw water from the wells of high caste families; to enter laundries, barber shops, cafes, or taxis; to keep women in seclusion and protect them by enacting domestic rituals; to wear shoes; to sit on bus seats; to register their names properly so that social benefits could be obtained; to attend school; to cover the upper part of the body; to wear gold earrings; if a male, to cut one’s hair; to use umbrellas; to own a bicycle or car; to cremate the dead; or to convert to Christianity or Buddhism. To enforce these sumptuary restrictions extra legally, Vellalars have been ready to field gangs of thugs to punish upwardly mobile Pallars or Nalavars. Such gangs pollute Untouchable wells with dead dogs, fecal matter, or garbage; burn down Untouchable fences or houses; physically assault and beat Minority Tamils; and sometimes kill them. Preceding the Maviddapuram crisis were several altercations in which Minority Tamils had died (p. 148).

Pfaffenberger (1994) states that the 1968 Vell..ā.lar violence against oppressed caste groups that were engaged in Gandhian satyāgraha or peaceful protests proved and exposed Jaffna to be “the most caste-conscious and conservative region of the island country” (p. 143). This fact would be used by Sinhala politicians and parties (pp. 159–162) to undermine Vell..ā.lar dominated attempts at Tamil nationhood (Pfaffenberger 1981). The protest at Maviddapuram proved that something similar to Dalit mass mobilization in India was indeed possible in Jaffna and could have influence on the public and policy. After the protests ended, temple doors all over the peninsula were gradually opened to oppressed castes, with the Government Agent promising to intervene on behalf of them “to put more teeth into the Prevention of Social Disabilities Act” (Pfaffenberger 1994, p. 162). Pfaffenberger (1994) found, however, that temple entry to depressed castes still sparked scuffles in the 1970s, with a caste riot in 1973 being a major example (p. 162). He concludes by pointing out that it was militant groups such as the LTTE that have been committed to the abolition of caste, which had been given overwhelming support from all cross-sections of Jaffna society. “But the unity that has been achieved is founded in terror and wedded to an antidemocratic ethos…,” he comments (p. 164).


Likewise, Dagmar Hellmann-Rajanayagam (1993) also looks at the LTTE in her article “The Jaffna Social System: Continuity and Change under Conditions of War,” focusing on the evolving influential role of the Karaiyār (fishermen) caste during the period of war, and how they modified the notion that the Vell..ā.lar caste constituted the ideal Tamil (pp. 251–270). While she argues the military prowess and anti-caste social policy of the LTTE played the key role, she also points to pre-war facts of Karaiyār contestation of Vell..ā.lar hegemony, such as early conversions to Catholicism to evade caste obligations (p. 268) and a significant number of Karaiyārs taking advantage of education (p. 270), similar to the Vell..ā.lars. As the Karaiyārs often claim a ksatriya (ruler) lineage (David 1972, p. 261), it is pertinent that Hellmann-Rajanayagam (1993) recounts their history in taking the lead in the revolt against foreign occupation (p. 267) and points out their contemporary battle against the state as warriors, “protectors, but also to some extent, rulers of the people,” culminating in a possible “Kshatriyaization” of their caste (p. 278). She argues that this goes against the “slowly, slowly become Vellalar” proverb,6 since non-Vell..ā.lar castes nowadays “demand respect and equality in their own right” (p. 281). This aphorism, however, may never have applied to the Karaiyār to begin with, as David (1972) has shown that the Karaiyār have a long history in contesting Vell..ā.lar hegemony and asserting their distinctiveness. Hellmann-Rajanayagam (1993) also points out that allegiance to antiquated caste solidarity has shifted to a contemporary one based on ethnicity (p. 280); while caste hierarchies still exist, these hierarchies fade into the background as distinctions with the ethnic other are increasingly and boldly contrasted at the forefront. This is what Kalinga Tudor Silva (2000[1999]) argues in his article “Caste, Ethnicity and Problems of National Identity in Sri Lanka,” when he says the primary identity of individuals in Sri Lanka has shifted from caste to ethnicity (p. 201) largely due to militant movements like the LTTE and the Sinhala-Buddhist Janatha Vimukti Peramuna (JVP), which both fostered ideologies of ethno-nationalism while simultaneously trying to eliminate hierarchies of caste and class.

In contrast, Indian Dalit rights activist and politician Ravikumar (2002) in his provocative article, “Caste of the Tiger,” has argued that Dalit rights have taken a back seat at the expense of Tamil nationhood and that any assertive writing for the rights of Pañcamars was seen as traitorous to the Tamil cause, despite the number of LTTE cadres from depressed caste backgrounds. Ravikumar has given the reader a historical narrative of how Vell..ā.lar-dominated parliamentary politics failed to address the plight of the Pañcamars, if not going directly against their interests, and has also highlighted some discriminatory restrictions and atrocities committed against them. He adduces the particularly disturbing words of celebrated Jaffna Śaiva Hindu tradition reformer Arumuga Navalar (1822–1879),7 who stated that the Par-ai[yar], the woman, and the Pañcamar were “all born to be beaten” (para. 5). Ravikumar clearly conveys the normative casteist undertone inherent in the nature of Sri Lankan Tamil culture. As this was written during a time of ceasefire between the LTTE and the Sri Lankan government, Ravikumar argued that the LTTE must clearly articulate a policy that guarantees and safeguards the rights of Pañcamar citizens.

Sinhalese connections and recent Jaffna case studies

  1. M. M. Mahroof (2000), however, has made the argument in his paper, “A Conspectus of Tamil Caste Systems in Sri Lanka: Away from a Parataxis,” that militancy and “exteriorisation” of depressed castes, such as migration, may “soften caste deprivations,” causing the Vell..ā.lars to rethink their rank in relation to other castes (p. 56). Mahroof gives the reader a general summary of the features of the Tamil caste systems on the island and an explanation as to how the Vell..ā.lars were able to become the most dominant and powerful, by patronizing the educational institutions founded by American and British missionaries and mastering English, which created a dichotomy between elite and common (pp. 47–48). He points out that the 1956 Social Disabilities Act would have been more effective if it had been coupled with a system of reservations or quotas for depressed castes in educational institutions and government employment (p. 55), as they have in India. As Mahroof is a researcher of caste in Sinhala society, he makes useful comparisons between the Tamil and Sinhala systems (p. 43), namely
  • The Tamil systems are based on notions of pollution, whereas the Sinhala systems are not.
  • The Tamil systems are a part of or supplemental to Hinduism, whereas the Sinhala systems are not related to Buddhism, which is theologically free of caste.
  • Non-Govi8 castes in the Sinhala systems are often wealthy with power; non-.ā.lar castes in the Tamil systems too often are not.
  • The Govi negation of non-Govi is silent and subtle, whereas the .ā.lar negation of non-Vell..ā.lar is often overt and can be offensive.

Also looking at other Sri Lankan caste systems in addition to Jaffna, Editors Kalinga Tudor Silva, P. P. Sivapragasam, and Paramsothy Thanges (2009) in Casteless or Caste-blind? Dynamics of Concealed Caste Discrimination, Social Exclusion and Protest in Sri Lanka have attempted to situate caste in its present context in various Sinhalese and Tamil communities, including a study on caste discrimination in war-affected Jaffna society and a field study on camps for internally displaced persons (IDPs). Given the security restrictions and inaccessibility of camps to researchers, this paper has introduced some rare and usually unobtainable data, such as the caste composition of camp inhabitants. While this paper focused specifically on Mallakam in 2007, the conclusions and reasoning drawn by the researchers give us an indication of camp composition and similar social situations of IDPs held in more recent facilities established by the state at the end of the war in 2009. Thanges and Silva found that a disproportionate number of IDPs came from the Nal.avar, Pall..ā.lar, and Pa-raiyar castes (p. 70), due to the fact that many of them had no homes to return to (p. 65). Because many of these castes were bound in nature to the Vell..ā.lars (David 1972), they had no capacity or resources to purchase their own land. Many bounded subordinates depended on their superordinates for their livelihoods, which they may have lost with the vast exodus of Vell..ā.lars from the peninsula. M. Siddartan has found that over one million people have left the Northern Province since the 1980s, with a disproportionate number of them being Vell..ā.lars who have taken advantage of extended social networks in Colombo and abroad (Silva et al. 2009, p. 64). Silva et al. (2009) point out that the Vell..ā.lars who have remained in the peninsula are usually able to afford to stay outside the camps and avoid the stigma of camp habitation (p. 71). Camp inmates from oppressed castes have difficulty worshipping in local temples (pp. 72–73) and obtaining drinking water from wells (pp. 73–74). They also mentioned the issues oppressed castes face in regard to purchasing land, due to objections from potential neighbors (p. 71). Prior to the war, there was a strict code among kinsmen, especially the Vell..ā.lars, in regard to land control, where one would only sell one’s land to another of the same caste (p. 74). If a person from a caste considered inferior were somehow able to secure property in a supposedly upper caste hamlet, they would most likely have to make an offer above the market price (p. 75). While this structure was somewhat altered during wartime, due to both militant anti-caste policy and the desperation of emigrating owners to sell and liquidate, it was still largely prevalent and a hindrance to many Pañcamars who remained in Jaffna.

One can see that even as displaced Tamils during a time of ethnic conflict, depressed caste persons received little sympathy from other Tamils belonging to privileged castes. This was not always the case. Silva et al. (2009) argue that the LTTE placed a ban on caste discrimination and even caste discourse in Tamil society (pp. 55–56) in areas they controlled. This merely imposed a veil of silence on the topic. Silva et al. have pointed out, however, that if peace were ever to return, caste would most likely reappear on the political and social agendas of various political actors (p. 52). Ultimately, the studies of Silva et al. have unraveled what the larger parts of Sri Lankan society have intrinsically chosen not to see, acknowledge, or discuss: the incontrovertible practice of caste, ultimately leading to a counterculture of denial that it even exists. Indeed, on the ground and in the field, one will find a highly prevalent discourse of denial from most people, regardless of ethnic group or region, that there is caste-based oppression in Sri Lanka. The mere discussion of caste is taboo. Particularly disturbing are the institutional efforts that go into caste censorship; Silva et al. report that the University of Jaffna, for example, discourages any research on caste (p. 20).

Some foreign scholars, however, have pursued caste-based research. Delon Madavan (2011) spent the summer in 2005 during the ceasefire studying social relations in Jaffna, focusing specifically on religion and caste. His study was particularly focused within the city limits of Jaffna Municipal Council where he interviewed 45 residents about their opinions on caste, love, and marriage. Madavan argues that the LTTE’s policy on abolishing caste significantly transformed Jaffna society due to “the fear it inspired [in] people,” finding that many Vell..ā.lars viewed the LTTE and this policy as an attempt to weaken their dominant caste position in Jaffna society (p. 3). Even though the LTTE lost its control over Jaffna in 1995, Madavan shows that the after effects of their policy continued 10 years later when he did

Figure 1 Perception by inhabitants of castes localization in Jaffna Municipal Council.

his field research in 2005. He finds that overt acts of caste discrimination present in the 1980s were no longer prevalent (p. 3; p. 18). From the depressed castes, Madavan argues in particular that many members of the Nal.avar caste have successfully elevated their class status by converting to Christianity, giving up their traditional occupations, and taking advantage of educational opportunities (p. 11). He also finds that they received a significant amount of financial remittances from kin, as did the Vell..ā.lars (p. 19). The and Par-aiyar castes, on the other hand, did not have the same success as the Nal.avar in mobilizing (p. 11; p. 19). The success of a caste elevating its status included the purchase of land in traditional Vell..ā.lar dominated areas – a process he calls desegregation – much to the opposition of Vell..ā.lars (p. 20). The most valuable contribution Madavan makes is a map he constructs, locating the major composition of each caste group living within each of Jaffna Municipal Council’s 23 districts (p. 9), which he constructed himself with the input of 15 of his interviewees (Figure 1). The primary maintenance of caste practice is through endogamy, with inter-caste couples having to leave conservative Jaffna if they wish to marry and not be ostracized by their communities (p. 19). His prediction is that caste will slowly dissipate, especially due to the changing notions of the younger generation. Further research and time, more importantly, will be able to tell if his hypothesis will prove true.


This brief survey of the available research on caste in Jaffna since the commencement of the war has shown (that)

  • War, militancy, and migration have had a direct influence on how caste has been practiced in Jaffna. Almost all the literature reviewed expressed .ā.lar domination prior to militancy. As Hellmann-Rajanayagam (1993) has indicated, the LTTE and their Karaiyār leadership broke this domination, silencing and veiling caste difference in order to strengthen Tamil nationalism. This created what Madavan (2011) calls desegregation in regard to residential areas in Jaffna. However, Silva et al. (2009) found that acts of caste discrimination were perpetuated in spaces such as IDP camps during the war and possibly after.
  • While caste is practiced at some level or another in Sri Lanka, including within Sinhalese communities, it is still most strongly emphasized in Tamil communities. Mahroof (2000) gives us an inter-ethnic perspective, providing a brief comparison between Sinhalese and Tamil caste systems, while Hellmann-Rajanayagam (1993), Pfaffenberger (1994), and Ravikumar (2002) give us the often silenced intra-community perspective of the contestation of caste within Tamil nationalism. Arasaratnam (1981) and Jeyarajah (2001) also clearly show that caste affects Tamil Christians and not just Hindus.
  • The evolution of caste from the colonial period in Sri Lanka. Arasaratnam (1981) and Rogers (2004) articulate the divergence in colonial policy between India and Ceylon; India used caste as an official category of recognition, while Ceylon chose not to do so. In addition to Hellmann-Rajanayagam (1993) and Silva (2000[1999]) arguing that the ethnic conflict has solidified identity in Sri Lanka along ethnic as opposed to caste lines, Arasaratnam and Rogers give us another reason as to why there is less intervention by the Sri Lankan state in local caste issues, particularly those of discrimination.

The recent research that was reviewed does not provide an update to the prewar studies on caste in Jaffna (Banks 1971[1960]; David 1972; Pfaffenberger 1982). These prior ethnographies provide descriptions of castes, their traditional occupations, kinship systems, their particular rituals, and their relations with other castes in great detail. Moreover, they describe the social structure of Jaffna as being one overly based on purity and pollution. The people of the Jaffna Peninsula, primarily but not limited to the dominant Vell..ā.lar caste, live their daily lives through the lens of what they perceive to be clean and unclean; the rituals they perform – both on a daily and occasional basis – and their relations with other castes, exist to remove various pollutants (tut.akku). None of the recent research looks at these practices in detail due to the overshadowing of war and living under military or militant rule. However, pertinent research questions that were deferred due to the onset of the civil war in the 1980s still yearn to be answered today. More importantly, several new questions have arisen with the defeat of the LTTE and the end of the war in Sri Lanka that demand scholarly attention. For instance, more research is necessary for understanding

  • Power structure in Jaffna. With the migration of many .ā.lar elite and the social mobility various caste groups have undergone during war and militancy, it is important to determine if Vell..ā.lar domination is still in existence in Jaffna or if there are any new dominant caste groups. An updated listing and ranking of caste groups, as provided in David 1972, can provide an indication of caste mobility and current social structure. It is also essential to numerically or proportionately assess traditional caste occupation vis-à-vis caste group to analyze whether there was mobility within each particular classification.
  • The effects external militarized entities have on the traditional local system, as the Northern Province is heavily stationed and patrolled primarily by ethnic Sinhalese armed forces and police officers during this postwar period.
  • The current state of hierarchically traditional systems of kinship, ritual practice, and inter-caste interaction for each caste group, which were central to the ethnographies that were written prior to the war.
  • If caste practice and discrimination are actually on the rise. With the end of the war, Silva et al. (2009) predict caste will resurface on the political and social agendas of various actors, which Hellmann-Rajanayagam (1993) shows were veiled and silenced during the reign of the LTTE. However, war and militancy eradicated most of the horrendous and humiliating overt and open acts of caste discrimination, some of which Pfaffenberger (1994) described. There are several nuances that deserve further attention, which one can examine far more easily than before, now that the war is over and Jaffna is more easily accessible. It is thus important to analyze social structure, power, and domination in Jaffna with the absence of the LTTE, to see if old attitudes and sentiments are resurrected.


  1. Valentine Daniel, Rachel Fell McDermott, Amarnath Amarasingam, Francesca Bremner, and Joel Lee read various drafts of this paper and provided necessary modifications. Darshan Ambalavanar, Mark Balmforth, Victor D’Avella, Victoria Gross, Kaori Hatsumi, Dagmar Hellmann-Rajanayagam, Benjamin Jeyarajah, Delon Madavan, John Rogers, and Sharika Thiranagama pointed me in the right direction at one point in time or another and/or provided me with versions of their own work for me to read and use.

Short Biography

Prashanth Kuganathan is a doctoral student in Applied Anthropology at Teachers College, Columbia University. He received his MA in South Asian studies from the Graduate School of Arts and Sciences, Columbia University in 2012. This article has been modified for publication from the literature review chapter of his unpublished MA thesis entitled Borders/ Barriers/Boundaries: The Resurrection of Caste Practices in Postwar Jaffna, written under the supervision of Professors E. Valentine Daniel and Rachel Fell McDermott.


* Correspondence address: Prashanth Kuganathan, Doctoral Student in Applied Anthropology, Teachers College, Columbia University, Box 211, 525 W 120th St, New York NY 10027-6696, USA. E-mail:


Both At.imai and Kut.imai loosely translate to slave or servant. They are, however, viewed as two different categories of servitude varying in degree of ownership or control by the dominant caste. In South India, the kut.imai are known as the right side castes, and the at.imai are known as the left side castes (Pfaffenberger 1982, p. 38).

2 Vellā.lar is spelled in more than one way in academic sources, such as “Vellala” or “Vellalar.”



Otukkappatt..avar is the singular for otukkappatt..avarkal..

  • A hymn in the Rigveda that explains the creation of humanity. When a makes the ultimate self-sacrifice, the four hierarchical varn. as or “cosmogonic human social types” emerge from his body: brāhman. a (priest) from his mouth, ks.atriya (ruler) from his arms, vaiśya (commoner) from his thighs, and śūdra (servant) from his feet (Marriott 2002, pp.1–3). While caste is often conflated with varn. a, the lived reality of caste endogamy is actually practiced through jāti, which Marriott (2002) describes as “one of the many thousands of marriage networks among South Asian families” (p. 1).
  • Jeyarajah’s Master of Theology thesis was written in Tamil and completed in May 2001 at the Tamil Nadu Theological Seminary. I used a concise English version of his thesis entitled, “History of Sri Lankan ‘Panchamar’ in

Relation to C.S.I. Jaffna Diocese During the Period 1973–1998.”

  • There is a South Indian Tamil proverb that expresses the fusion of the .ā.lar caste. It has a connotation that the castes fusing to become Vell..ā.lar were inferior, as if to imply the Vell..ā.lar caste was the aspired ideal. Arasaratnam (1981) translates this as “Kallas, Maravas and Akampadiyas gradually come to be Vellalars” (p. 385). In the original Tamil transliterated into English, the proverb is “Kallar, Maravar, Kan-atākampat.iyar mella mella vantu Vell..ā.lar āvār.” The words “mella mella” literally translate to “slowly slowly.” 7

For a detailed study on Navalar and his reform movement, see D. N. Ambalavanar (2006). Arumuga Navalar and the construction of a Caiva public in colonial Jaffna. Ph.D. dissertation, Harvard Divinity School, Harvard University.


The Govigama, which may be spelled in a variety of ways including Goyigama or Goigama, and sometimes known simply as Govi, is the dominant Sinhalese caste in Sri Lanka. Also a caste of agrarian landlords, the Govigama caste is often viewed as the Sinhala parallel to Tamil Vell..ā.lar caste. Similar to the Vell..ā.lars in the north, the Govigamas in the south also dominated educational institutions and government jobs.

This article appeared in Sociology Compass vol 8/1 (2014): 78–88, 10.1111/soc4.12101 … while the author was attached to Teachers’ College, Columbia Univeristy


Ambalavanar, D. N. 2006. Arumuga Navalar and the Construction of a Caiva Public in Colonial Jaffna. Ph.D. dissertation, Harvard Divinity School, Harvard University.

Arasaratnam, S. 1981. Social History of a Dominant Caste Society: The Vellalar of North Ceylon (Sri Lanka) in the 18th Century. The Indian Economic and Social History Review 18(3): 377.

Banks, M. 1971[1960]. Caste in Jaffna. p. 61 in Aspects of Caste in South India, Ceylon, and North-West Pakistan, edited by E. R. Leach. Cambridge, New York: Cambridge University Press.

Bastiampillai, B. E. S. J. 1988. Caste in Northern Sri Lanka and British Colonial Administrative Practice in the mid 19th Century: Compromise and Expediency. Sri Lanka J.S.S. 11(1 & 2): 47.

Bayly, S. 1999. Caste, Society and Politics in India from the Eighteenth Century to the Modern age. New York: Cambridge University Press.

Casie Chitty, S. 2004[1934]. Castes, Customs, Manners and Literature of the Tamils. New Delhi: Asian Educational Services.

  1. in Casteless or Caste-Blind?: Dynamics of Concealed Caste Discrimination, Social Exclusion, and Protest in Sri Lanka, edited by K. T. Silva, P. P. Sivapragasam, P. Thanges. Copenhagen: International Dalit Solidarity Network.

Cohn, B. S. 1990[1987]. An Anthropologist Among the Historians and Other Essays (2nd impression with corrections. ed.). Delhi: Oxford University Press.

Daniel, E. Valentine. 1984. Fluid Signs: Being a Person the Tamil Way. Berkeley: University of California Press.

David, K. A. 1972. The Bound and the Nonbound: Variations in Social and Cultural Structure in Rural Jaffna, Ceylon. Ph.D. dissertation, The University of Chicago.

Dirks, N. B. 2001. Castes of Mind: Colonialism and the Making of Modern India. Princeton, N.J.: Princeton University Press. Hellmann-Rajanayagam, D. 1993. Jaffna Social System: Continuity and Change Under Conditions of War.

Internationales Asienforum 24(3–4): 251.

Jeyarajah, B. 2001. History of Sri Lankan “Panchamar” in Relation to C.S.I. Jaffna Diocese During the Period 1973–1998. Master of Theology Thesis, Tamilnadu Theological Seminary.

Madavan, D. 2011. Socio-Religious Desegregation in an Immediate Postwar Town Jaffna, Sri Lanka. Carnets De Recherches (2). (Last Accessed 31 July 2013).

Mahroof, M. M. M. 2000. A Conspectus of Tamil Caste Systems in Sri Lanka: Away from a Parataxis. Social Scientist 28(11/12): 40–59.

Marriott, M. 2002. Varna and Jati. Unpublished manuscript.

Pfaffenberger, B. 1981. The Cultural Dimension of Tamil Separatism in Sri Lanka. Asian Survey 21(11): 1145–1157.

Pfaffenberger, B. 1982. Caste in Tamil Culture: The Religious Foundations of Sudra Domination in Tamil Sri Lanka. Syracuse, N.Y.: Maxwell School of Citizenship and Public Affairs, Syracuse University.

Pfaffenberger, B. 1994. The Political Construction of Defensive Nationalism: The 1968 Temple Entry Crisis in Sri Lanka. p. 143 in The Sri Lankan Tamils: Ethnicity and Identity, edited by Chelvadurai Manogaran and Bryan Pfaffenberger Boulder: Westview Press.

Raghavan, M. D. 1971. Tamil Culture in Ceylon; a General Introduction. Colombo: Kalai Nilayam.

Ravikumar. 2002. Caste of the Tiger: Eelam and the Dalit Question. htm (Last Accessed 31 July 2013).

Rogers, J. 2004. Caste as a Social Category and Identity in Colonial Lanka. The Indian Economic and Social History Review 41(1): 51–77.

Silva, K. T. 2000[1999]. Caste, Ethnicity and Problems of National Identity in Sri Lanka. p. 201 in Nation and National Identity in South Asia. edited by S. L. Sharma and T. K. Oommen. New Delhi: Orient Longman. Skjønsberg, E. 1982. A Special Caste?: Tamil Women of Sri Lanka. London: Zed Press.

An Editorial Note

The map indicates that the focus is on the Jaffna Peninsula. Hence, the question arises: to what degree do thesame patterns prevail in the Mannar, Kilinochchi, Mullaitviut and Vavuniya Districts?


Filed under caste issues, centre-periphery relations, commoditification, communal relations, cultural transmission, demography, discrimination, economic processes, education, Eelam, ethnicity, governance, heritage, Hinduism, historical interpretation, Indian traditions, island economy, landscape wondrous, legal issues, life stories, LTTE, politIcal discourse, power politics, religiosity, Saivism, self-reflexivity, Sinhala-Tamil Relations, sri lankan society, Tamil civilians, Tamil migration, tamil refugees, truth as casualty of war, unusual people

4 responses to “Caste in Jaffna

  1. John Clifford Holt

    Excellent overview providing a helpful platform for research on an important topic.

  2. So many changes have happened today, you idiot. Are you still writing about caste? The name of Pallar today known to the world as Devendra Kulavelalar has been announced to the world by the Indian Parliament.

  3. Tharshika

    Sir, how do you make caste map without data.. Could you please share the data?

    • This article is NOT written by Thuppahi and/or Its Editor. The author is PRASANTH KUGANATHAN. You need to locate him whwrever he is in the world and address your inquiry (it seems to bea CHALLENGE) to him.

Leave a Reply