From jati to samaj


back to issue

NO view is a ‘view from nowhere’, and views on caste are no exception. This essay takes caste to be neither simply an intellectual puzzle, nor existential fact, but desires its demise as a fundamental social, political and ethical problem of our times. Such a view throws up a major paradox of caste today: rising caste groups/identities, declining caste system; rising casteism, and dwindling legitimacy. This paradox demands not only explanations about the persistence of caste, but also a rethinking of caste as problematic: Can caste (the social group) and casteism be disentangled, i.e., can there be caste without casteism? What kind of a problem is caste? How have caste and casteism adapted to changing political-economies? What would an anti-caste position need to consider for a liberatory politics of caste? These questions offer a framework to evaluate the arguments offered in this essay.1 I start with a discussion of the paradox, then introduce two explanations for the paradox – a dominant narrative that makes the problem of caste ‘disappear’, and an alternative one.

Scholars across disciplines have noted the recent (late colonial period onwards) dissociation of the social group called ‘caste’ from the ‘caste system’ and the rising significance of caste identities. Here some clarification of the terminology is useful. Caste groups have historically appeared as three socially identified collectives – varna or the pan-Indian category for the four classically defined groups or estates, jati or the approximately 4000 ‘large-scale descent groups’2 that are recognized by local populations and the state, and jati-clusters or federations of politically organized collectives of (not too socially distant) jatis.3 All three caste groups are invoked in socio-political life by social actors and social observers.

Further, the so-called ‘caste system’ has arguably been built upon a tripartite foundation of hereditary occupational specialization, ritualized/sacralized hierarchy, and mutual repulsion or separation,4 and kept in place by a variety of forces that are political (local kings and dominant caste groups, power relations including violence, patriarchy, and the modern state), economic (village economy with relations of economic interdependency, exploitative relations and segmented labour markets), social (kinship ties, marital practices and institutions of customary law), and cultural-ideological (ritual ranking, ideas of purity-pollution and religious sanction). The paradox then is that despite the relative decline of the caste system (due to the colonial erosion of the village economy as we knew it, increased ‘de-ritualization’ of caste, a staunchly modernist and interventionist postcolonial state, and resistance by so-called ‘lower’ castes to their own domination), it has been observed that caste groups (all three types) and caste identities are invoked in public, and persist and flourish in Indian society, not only in matrimonial markets and electoral politics, but also in everyday life shaping access to land, credit, capital, employment, housing and knowledge, prestige and power.5 


The other part of the paradox refers to the persistence of casteism despite the general illegitimacy accorded to that practice through state strictures and by public disavowal of casteism (not caste) in popular life (no one likes to be called jativaadi any more even when, or especially when, they practice casteism). Again, to clarify, casteism usually refers to a form of discrimination based on imputations of ‘caste’. While a focus on discrimination is crucial to proscribe casteism, it makes casteism – a deeply entrenched social practice – recognizable only through the narrow lens of law. To bring its social roots to the fore, casteism is viewed here as a set of monopolization strategies based upon domination, stigmatization (including humiliation), exclusion, exploitation, or annihilation on the ‘fact’ of caste.6 

Such a view of casteism encompasses discrimination while also bringing into focus a sociology of the gainful effects and social functions of casteism, i.e., who gains from casteism and how? Casteism is thus seen to be rampant despite a ‘declining’ caste system, the persistence of caste groups, a legal system that has made caste discrimination a punitive act, and illegitimacy of casteism.


The paradox is explained away by a discursive régime of truth that narrates the problem of caste as ‘resolved’ due to the forces of Indian modernity, democracy and capitalism. Five (scholarly and popular) tropes animate this discourse: a political trope of democratization or modernization which views castes positively as political interest groups, instead of as shackles for free association;7 a cultural trope of substantialization or ethnicization which views castes again positively as ethnic groups or independent communities of cultural identity contributing to India’s diversity, instead of as status hierarchy and inequality;8 an economic trope of capitalization or productivity which valorizes castes as innovative networks of social (trust) and economic capital instead of as exploitative relations.9 Two more tropes act as book-ends to this discursive régime: a benign-normal trope which views caste as privatized and domesticated practices (such as endogamy) that do not affect life-chances, instead of as public sites for reproducing caste networks and inequality, and a brutal-abnormal trope which views the frequent caste atrocities as aberrations of the above benign-normal existence of caste, instead of accounting for the quotidian existence of caste violence.


This 5-trope discourse advances the claim that castes can exist without a caste system because caste is fundamentally about ‘social difference’ not hierarchy and it is only the latter that really requires a caste system. Further, the existence of castes need not be considered a social problem since caste groups are simply ‘cultural communities’ or ethnicities, and caste identities are cultural identities that enrich India’s diversity. The second half of the paradox is simply dismissed by asserting that casteism is on the decline or only persists in restricted arenas of social life and is not a determinant of ‘life-chances’. Hence, the ‘explanation’ for the paradox is: castes without caste system or casteism.

Taken together, these five tropes facilitate the ideological and triumphalist narrations of ‘shining’ India, heralding a ‘new society’ (purportedly free of casteism although not caste-free) to go with a ‘new economy’ (globalized, liberalized, privatized) and a ‘new polity’ (matured democracy). This discourse is emblematic of a liberal, bourgeois and neo-Hindutva view of Indian society that appears to be anti-casteist since it asserts that the problem of caste has been overcome in India today. In place of caste is a mythology of the ‘ethno-futures’ (of caste) that embraces and celebrates caste identities as cultural identities, caste pride as cultural assertion, and caste endogamy as ‘natural’ affinity to those who are ‘culturally alike’. Viewed as a positive contribution to India’s cultural diversity, caste is set to enter public policy, not as a problem of inequality, but as a problem of ‘management’ of ethnic and cultural diversity. In an age of multiculturalism, we can then only aspire to tolerate caste by ‘respecting’ the caste Other as cultural Other, rather than challenge the existence of caste as a fundamental problem of Indian society.


An alternative explanation of the paradox acknowledges the emergence and social embeddedness of two phenomena based upon claims to ‘culture’ and ‘difference’. The first is a modality of casteism that I call cultural or differentialist casteism. Taking ‘difference’ to be at least as important as ‘hierarchy’ in the workings of caste today, it views casteism as operating within twin modes, as heterophobia and as heterophilia.10 As heterophobia (a psycho-social fear of ‘difference’ as embodied in peoples, behaviours, relations, artifacts, beliefs) casteism is a quasi-permanent social separation or ‘apartheid’ operating along a graded hierarchy for economic exploitation. This is the historically visible modality of casteism.

In recent times, casteism operates simultaneously as heterophilia (a preference or Bourdieuian ‘taste’ for difference) which does not require a dread or fear of difference, but nurtures, requires and maintains a desire for cultural difference along lines of caste. It does not require annihilation, assimilation or apartheid of different bodies, but only that the caste Other appear as the cultural Other, and that the Other remain within the cognitive-socioeconomic ‘place’ marked by casteism.11 Both forms of casteism support monopolistic practices and ‘mixophobia’ (aversion to caste mixing), ensuring that markets in production (capital, labour, credit) and reproduction (marriage, housing) are segmented along caste lines. Since most scholarship only recognizes casteism as heterophobia, it has been easy to mistake heterophilia as anti-casteism rather than as a new form of casteism.12 


In an increasingly conscious multi-cultural India, cultural casteism based upon heterophilia acquires legitimacy by representing caste discrimination as cultural differentiation, and status distinctions as cultural difference. Instantiation of cultural casteism gives rise to the second phenomenon – a modality of the caste group, samaj, based upon ideological representations of caste identity as cultural identity, and caste distinctions as cultural difference. The paradox is then explained by how cultural casteism (as monopolization) produces a caste system (a system of monopolies based on the criterion of ‘cultural difference’ which governs caste as economic, political, social and cultural capital) which in turn produces caste groups as samaj (embodiments of cultural difference). In short, caste groups/identities persist as samaj, caste system operates as monopolies, ‘cultural’ casteism emerges as the legitimate form of casteism.

This explanation requires a serious look into the ‘interiors’ of caste groups where cultural-ideological ‘work’ is continually performed to make caste identities and interests appear far more ‘stable’ and caste groups as far more integrated ‘wholes’ than in reality. Caste elites perform this work by taking the lead in forming samaj through the ‘culturalization of caste’, a process that socially constructs caste as cultural community by producing, organizing, instituting and disciplining the ‘culture’ of a caste group and representing caste officially as cultural difference or ethnicity.

The ‘culture’ of samaj is thus sought to be made explicit, even written down, codified and legislated as community customs and traditions in an attempt for political ‘recognition’ by its believers, including the state, which confers legitimacy to samaj under new forms of capitalism and multicultural democracy. Samaj exists alongside jati in ideological conflict because these are qualitatively different ways of enjoining group ‘belonging’. As samaj, caste difference and distinctions become reformulated and naturalized as cultural difference; group membership shifts from what Foucault has termed ‘degrees of affiliation’13 to ‘degrees of normalcy’ wherein the former, exemplified by jati, works through the logic of ‘blood’ (birth and descent), while the latter, exemplified by samaj, works ostensibly through the logic of performance of cultural belonging (‘norms’ of behaviour).


Samaj is then neither simply a gloss on jati, nor does it qualify as an ethnic group so long as its ‘culture’ is derivative of the historical caste system which operated by demanding difference in order to rule.14 Moreover, the production of the ‘culture’ of a samaj proceeds through material contradictions and ideological challenges. Despite attempts by elites to homogenize its interiors, samaj does not always imply a shared cultural identity or interests among members. The claims of samaj to transforming caste into cultural identity is, therefore, at best tenuous and unstable, due to internal contestations of ‘culture’ and also since jati ultimately comes back to haunt samaj. For, the conditions of possibility of a cultural identity for samaj itself depend upon a caste identity which underlies it and is prior to it.15 


Using Jean-Luc Nancy,16 we may think of samaj as suspended between jati (a community based upon ‘a commonality of being’ or fetishized shared substance such as ‘blood’) and a yet-to-emerge community based upon a ‘being in-common’ (an existence in common but without recourse to an essential sharedness). This negatively affects the potential of samaj for anti-caste collective action although it allows castes to collectively act from time to time within the logic of electoral politics.


The arguments above acquire significance given the widespread formation of samaj by large numbers of ‘upper castes’ over the last century. Dalit groups, who are best located in class and ideological terms to lead the building of an anti-caste politics, need to view this trend (culturalization) with caution due to the presence of a ‘culture trap’ that inheres in the formation of samaj.17 Tellingly, samaj continues to perform historical functions of caste, such as controlling access to means of production, allocation of labour to different processes, and rights over distribution of products of labour. Moreover, it also performs other equally historical functions such as aiding the reproduction of patriarchy (since the ‘culture’ of a caste requires imposition of gendered views of social relations and the ‘patriarchal compact’), the active disorganization of any class formation and class consciousness within samaj, and enabling state governmentality (since the state confers legitimacy to claims of samaj and cultural distinctions between caste groups).18 Not sanskritization (which even to the extent it did exist lies dead now due to the loss of prestige of upper caste symbols), nor ethnicization, but it is through culturalization that caste groups ensure the durability of casteism and caste.


In conclusion, the problem of caste is that caste groups presuppose casteism; the latter gives rise to the former. Imagining that castes can exist without casteism is what Nicholas Dirks has called a ‘bourgeois dream world.’ Castes cannot be assumed to be ‘privatized’ (as endogamous practices) since the ‘private’ sphere is where casteism regenerates itself through creating monopolies. If we are to pay heed to Ambedkar’s intellectual legacy which convinced him that the ‘monster of caste’ must be killed, then scholarship needs to engage in enabling caste-conscious anti-caste political mobilization of multi-caste groups. It cannot simply be a disengaged observer of cultural casteism and samaj. It should critically demand an anti-caste agenda from the caste-conscious mobilization of jati-clusters for electoral purposes, explicitly reject the liberal disavowal of casteism as a fundamental organizing principle of current Indian society, and recognize the need for anti-caste politics to work through identity towards solidarity.19 



1. This essay draws upon my book (The Culturalization of Caste in India: Identity and Inequality in Multicultural Age. Routledge, London, 2012), which is an ethnographic study of a group of potter-artisans, Kumhar in Chhattisgarh (classified as OBCs or so-called ‘lower castes’) and which tackles many of the above questions.

2. P. Kolenda, Caste in Contemporary India: Beyond Organic Solidarity. Waveland Press, Illinois, 1978.

3. L. Carroll, ‘Colonial Perception of Indian Society and the Emergence of Caste(s) Associations’, Journal of Asian Studies 37(2), 1978, pp. 233-250.

4. C. Bouglé, Essays on the Caste System. Cambridge University Press, Cambridge, 1908.

5. A. Deshpande, The Grammar of Caste: Economic Discrimination in Contemporary India. Oxford University Press, Delhi, 2011. S.K. Thorat and K. Newman, ‘Caste and Economic Discrimination: Causes, Consequences and Remedies’, Economic and Political Weekly, 13 October 2007, pp. 4121-4124. Takahiro Ito, ‘Caste Discrimination and Transaction Costs in the Labour Market: Evidence From Rural North India’, Journal of Development Economics 88(2), 2009, pp. 292-300.

6. I draw upon Randall Collins who has shown how monopolization strategies underlie the existence of status groups in the most modern sectors of the economy and society. See R. Collins, Weberian Sociological Theory. Cambridge University Press, Cambridge, 1986.

7. Rudolph’s is a classic statement. See L. Rudolph, ‘The Modernity of Tradition: The Democratic Incarnation of Caste in India’, American Political Science Review 59(4), 1965, pp. 975-989.

8. C.J. Fuller (ed.), ‘Introduction: Caste Today,’ in Caste Today. Oxford University Press, Delhi, 1996, pp. 1-31; D. Gupta (ed.), Caste in Question: Identity or Hierarchy? Sage, New Delhi.

9. A recent example is H. Damodaran, India’s New Capitalists: Caste, Business and Industry in a Modern Nation. Permanent Black, Delhi, 2008.

10. See the recent work on racial prejudice in the French context by P. Taguieff, The Force of Prejudice: On Racism and its Doubles. University of Minnesota Press, 2001.

11. I have elaborated elsewhere on the need to view ‘place’ or its closest gloss in Hindi – aukad, as a far better marker of how casteism and untouchability operate, than the criterion of ‘touch’. B. Natrajan, ‘Place and Pathology in Caste’, Economic and Political Weekly XLIV, 19 December 2009, pp. 79-82.

12. Here, viewing casteism solely as caste atrocities makes the error of thinking of casteism only as heterophobia.

13. M. Foucault, Discipline and Punish: Birth of the Prison. Vintage Books, New York, 1995, p. 184.

14. U. Chakravarti, Gendering Caste: Through a Feminist Lens. Stree, Calcutta, 2003. The only caste groups who are capable of ethnicizing in this sense are Dalits, and even then only those Dalit groups who have consciously tried to construct what Ambedkar called an ethnic identity that is not derived from caste. Jaffrelot comes closest to acknowledging this. C. Jaffrelot, ‘Sanskritization vs Ethnicization in India: Changing Identities and Caste Politics Before Mandal’, Asian Survey 40(5), 2000, pp. 756-766.

15. One way to think about the imbrications of caste and culture or jâti and samâj in such projects is to ask the question posed by Walter Benn Michaels in the context of race. Thus we could ask: ‘If to be a Kumhâr you have to do Kumhâr things (as prescribed by the codes of samâj), then can you really count as doing Kumhâr things unless you already are a Kumhâr (as recognized by your jâti)?’

16. Jean-Luc Nancy, The Inoperative Community. University of Minnesota Press, Minneapolis, 1991.

17. Samaj is very common across caste groups ostensibly for matrimonial purposes, and oftentimes among ‘upper castes’ for creating residential enclaves. But even a cursory look at the ubiquitous Brahman samaj reveals that a larger project is to build a ‘cultural community’ which is a monopolization strategy that reproduces and entrenches caste inequalities.

18. Interestingly, caste cultures are not ‘protected’ in the way that tribal, linguistic-ethnic, and religious ‘cultures’ are protected under the Indian Constitution. Yet, the ‘cultural defence’ is precisely how the notorious khap panchayats seek to continue their hold over samaj members and stave off an intervening state and judiciary.

19. In this regard, the most recent ‘beef wars’ on Indian university campuses could be a rich site for an anti-caste politics that reveals the process of culturalization of caste and cultural casteism (in this case strategies for monopolization of power and status on campus). Here casteism is camouflaged as a clash over cultural identities. The well-organized student body that is calling for the cafeteria to serve beef is rightly focused on the stigmatization of beef-eating by battling the casteist and culturalist claims of the ABVP that ‘beef-eating as a practice is not part of ‘Indian’ or ‘Hindu’ community practices or heritage’ (as if there is such an entity). An anti-caste politics around beef, however, need not fall into a ‘culture-trap’ set up by the terms of the discourse which uncritically defines caste communities by purported ‘essential’ cultural traits. The beef war is a caste war on cultural grounds; not a cultural war on caste grounds. It is casteism that defines caste communities, and not cultural practices – almost all of which are put in place by casteism – which define caste communities. For a good exposition of the ‘beef war logs’ see