The Balmiki critique of modernity


back to issue

IN this essay, I explore the conditions of an overtly political anxiety felt by many of the Dalit activists whom I met while conducting fieldwork among Delhi’s Balmiki community. I found that the grassroots activists who became my friends and acquaintances often disparaged their own movement for its ‘ego problem’, that is, the tendency of leaders to splinter off into their own groups and contribute to the endemic fragmentation that plagues the Dalit movement. The same activists were also quick to point out a problem of insincerity among organizations in the movement: an obsessive concern with registration, rank and office even when seemingly little ‘on the ground’ work is being done.

Such anxiety about the ability of their movement to successfully organize is a general feature of social movements but Balmikis too frequently interpret these problems as signs of a failure to be fully modern. In this essay, I argue that the problems of fragmentation and superficiality do indeed indicate a failure to be modern, but that this is not a failure of the Dalit movement itself, but is rather the result of a tension within modernity. Briefly characterized, this is a tension between, on the one hand, the needs of a modern state for disembedded identities and impersonal procedures and, on the other hand, the fact that the very oppression being fought by Dalits happens primarily in the interstices of such abstractions. The central argument of this essay, then, is that a subtle but radical critique of modernity is occurring within Dalit civil society in general and Balmiki civil society in particular.


In a series of influential essays on civil society, Partha Chatterjee argues that the traditional taxonomies of political theory are inadequate for describing the socio-political landscape of India and other former colonies in the contemporary era of world history (Chatterjee 2001, 178). In particular, it is the category of ‘civil society’ that does not accurately describe the relationship between the state and society for most of the globe’s population. He argues that ‘civil society as an ideal continues to energize an interventionist political project, but as an actually existing form it is demographically limited (39).’

The form in question is based on ‘those characteristic institutions of modern associational life originating in western societies which are based on equality, autonomy, freedom of entry and exit, contract, deliberative procedures of decision making, recognized rights and duties of members, and other such principles’ (172). The civil society form is limited to a ‘fairly small section of "citizens"’ (172) because ‘[m]ost of the inhabitants of India,’ he argues, ‘are only tenuously, and even then ambiguously and contextually citizens in the sense imagined by the constitution’ (38).

India’s majority of non-citizens, Chatterjee claims, do not relate to their government in the ordered and principled space of civil society, but rather in the much more tumultuous space of political society. Whereas civil society consists of individual citizens interacting with their state on the basis of a set of legally defined procedures, political society operates by a very different set of rules. For a start, the ‘non-citizens’ of political society, organized through political parties, unions and other organizations tend not to exercise simple civic or political rights but rather demand collective social rights as communities.

In stark contrast to the legalistic procedures of civil society, political society even allows mobilization on the basis of illegality, as for instance, slum-dwellers fighting to remain on their land. Because the state is more than willing to interact with its ‘non-citizens’ in these ways, we witness the extensive networks of patronage, corruption and influence-peddling that are notorious in contemporary India. This combination of illegality, social rights, collective subjectivity and governmental flexibility thus forms an alternative axis of state-population relations that Chatterjee believes to be absent from western modernity (see 177-78).


At the centre of Chatterjee’s argument then, is an empirical claim that civil society as an ‘actually existing form’ is present for some – namely the global elite – but not for the majority of the world’s population who instead relate to their states via the norms of political society. My fieldwork among Delhi’s Balmiki community provides evidence that contradicts Chatterjee’s claim.1

The Balmikis – the contemporary name adopted by most safai karamchari or ‘sweeping’ castes of northern-western India – ought to belong to the domain of ‘political society’ in so far as they tend to act as a community, demand social rights, depend on patronage in the form of municipal corporation sweeping jobs, and participate in various forms of patrimonial and legally suspect politics. The various sweeper castes of North India (Balmiki, Bhangi, Churha, Mehtar etc.) are categorized and counted as scheduled castes and as such are eligible for reservations in the public sector as well as for special financial assistance schemes (such as those granted by the National Safai Karamchari Financial and Development Corporation).


Nonetheless, during ethnographic fieldwork conducted among the Balmiki community between 2003 through 2004, it became clear that there was no shortage of Balmiki/safai karamchari organizations that fit the form of western civil society, but rather a vast surplus. Organizations dedicated to education (primary/secondary coaching and some technical training), Balmiki rights advocacy (anti-discrimination and anti-atro-city), and the struggle against the privatization of Delhi’s safai karamchari positions were active throughout the areas in which I worked. These groups were ‘grassroots’ in so far as they were composed almost entirely of Balmikis themselves. While it is true that some of the organizations were led by the so-called ‘creamy layer’ of educated Balmikis (mostly government servants), only a few of these could be truly called ‘elite’ as many leaders merely held Class III positions as clerks.

Of the organizations that I came into contact with (more exist), only two were led by persons who could be classed as ‘elite’ (i.e., they had some combination of the following: advanced degrees, Class I employment, housing in prosperous neighbourhoods, and/or better material possessions such as large flats, new cars, etc.). The remainder (many of which orbited around the Union leadership) were staffed by persons who were less educated (some had only passed Class 10), less gainfully employed (e.g., they owned small shops, worked in them, or were otherwise under-employed or even unemployed) and tended to live either in slums, regularized slums, resettlement colonies, or in government housing projects.


Although forming an underclass of NGOs, these associations fit Chatterjee’s description of civil society because they ‘are based on equality, autonomy, freedom of entry and exit, contract, deliberative procedures of decision making, recognized rights and duties of members, and other such principles’ (Chatterjee 2001, p. 172, emphasis added). Most of the organizations had a complete set of office bearers whose names could be rattled off by any of the other officers. They also usually had some sort of constitution or manifesto, a set of bye-laws and plenty of greeting cards, letterheads and receipt books. Decisions were made democratically or, at least, voting was nominally a part of the life of the organization. My observations of associational life among Delhi’s Balmiki community show that the ideal and normative discourses of civil society as defined by Chatterjee are alive and well.

It is relatively easy to show that Balmikis are mimicking the ideals and discourses of civil society, but Chatterjee is asking whether or not they actually put them into practice. This question, it appears, is incorrectly formulated. To Chatterjee, there is a difference between the ‘ideal’, which energizes political activity for most Indians (in so far as at least the trappings of civil society/democratic process are in place) and the ‘actually existing form’, which exists only for a small elite. Rephrasing these terms more precisely, a difference is being drawn between mere form and actual practice.

Let us for a moment reflect on the seemingly compulsive interest in maintaining the proper form of the organization. Nearly every organization that I encountered, no matter how small and fledgling, had a letterhead, business cards, some sort of manifesto or other official paper detailing the names of the president, the general secretary, the treasurer, etc., even when there weren’t necessarily any members other than the office bearers or any funds in the bank.

The more active groups did occasionally hold rallies and write letters to politicians or government officials, and these activities required member participation and fund management. However, these activities hardly seemed to warrant the elaborate institutional structure of the groups. ‘Members’ were culled from the vast network of family and friends in which the office holders were embedded; the small amounts of funds involved were raised and spent informally on things like banners, stationary and pens. Record keeping was at best spotty. Despite the apparent meaninglessness of the actual posts, titles seemed to be highly valued. Most Balmiki organizers, it seems, were reducing the form of democracy to a mere formality.


At the meetings and functions of these Balmiki groups, the mass exchange of greeting cards was standard practice. Employment, when it was prestigious, was mentioned, otherwise people gave their social service and political affiliations top billing. Cards often listed more than one affiliation. I was given a greeting card by a man who listed himself as the spokesman of one organization, the general secretary of one its branches, the national general secretary of another organization, the chairman of another group, and the general secretary of yet one more association.


To put this in perspective, let us also consider one organization that did not follow this pattern. Unlike the others which usually claimed to have general Balmiki welfare as their agenda, this group was dedicated exclusively to education. Though its founders were Balmiki residents of a Balmiki slum, they organized in the name of Dalits on the basis of Ambedkarite principles. The two lead activists of this group frequently dissuaded me from talking to other organizers because, as they put it, they were bogus; they had all the trappings of the organization but didn’t do any actual work.

My experience verified this claim and that my friend’s organization was quite different: they conducted regular classes for several hours per day with minimal resources and virtual disregard for formal titles and obligations (I still don’t know who was officially president, even though it was always quite obvious who was in charge). The leader of this group frequently warned me not to waste time talking to most Balmiki and other Dalit leaders saying that ‘everybody is the president of a sangh, a kendra, or a dal, but if they’re all leaders, who are the followers?’ So the problem with most Balmiki organizing, as far as this group was concerned, was that it was all form and no content, formality without any practical consequence.

I gradually came to realize that there was considerable merit in their remarks but also, more importantly, that this reduction to formality is not without its own cultural significance. Obviously, more is happening at these meetings than just social or political activism; they are socializing opportunities as well. The meetings, which came complete with alcohol, food and gift exchange, often felt more like parties. Such events, not unlike their western counterparts, make ordinary socializing seem more legitimate while simultaneously providing opportunities for networking to caste compatriots across the vast city (useful for making marriage alliances and business deals).


In one group, the president, also a well-known union leader, was always the centre of attention, the host, and the man to which all the others deferred, even in their joviality. Each gathering felt more like a fête for him than an actual meeting. This was perhaps a relatively extreme case of sycophancy, but it appeared to be the general rule that hierarchy of position (e.g., president, general secretary, treasurer, etc.) followed predictable patterns. Those with greater seniority tended to be honoured with higher positions and when younger people were made president there usually seemed to be some external way to account for their high prestige level, such as family connections, power within the union, success in business, or a good government job.

The organizations, even though ostensibly democratic, appeared to be using the titular structure of the organization as a way of recognizing and reiterating hierarchy. This hierarchy had a clear impact on the deliberative-democratic procedures, but it did not obviate the form. Discussions were held, opinions were heard, and even voting was held (non-secretively). Yet at the same time there were clear acts of deference during the discussion and not once did I witness the will of the president being openly contradicted (if so, the organization would fissure). Formal equality and deliberative procedure were in some sense present, yet democracy was failing in crucial ways. Something, I am arguing, is wrong with the way that Chatterjee has conceived of the problem because the difference between ideal (mere) form and actual form is in fact quite difficult to distinguish.


By way of clarification, I am not arguing that this is just homo hierarchicus dressed in new clothes. What I hope to demonstrate, rather, is that some configuration of everyday practice is interfering with the execution of the process of civil society without disturbing the form (it doesn’t particularly matter what that configuration of practice is).2 In other words, all the formalities can exist and ‘actual’ democracy can still fail because moment-to-moment pragmatic social interactions (in this case hierarchically configured ones) can interfere with the process. Moreover, the distinction I am drawing – in contrast to Chatterjee – between success at the level of form and failure at the level of everyday practice is intrinsic to civil society and all similar legal-bureaucratic, procedural institutions.

It is well beyond the scope of this short essay to prove the universality of the claim but some simple commonsense examples can illustrate the point. Nepotism is extremely common among the elite level organizing that Chatterjee would classify as civil society. (Think of Indira Gandhi’s many stints with esteemed social service organizations in her early career. Also think about the Bush family!). Western activists and scholars have also argued that race and gender oppression occurs in the micro-practices of everyday life which is precisely why simple legal and political change is not considered adequate for their rectification.3

Chatterjee’s mistake is in not accepting democracy/civil society as anything other than form – a set of procedures for ensuring equal access, voluntary participation, and majority rule. None of this, however, can happen without actual human beings executing the procedures. People, however, can only interact with one another through pragmatic encounters (even though these may be mediated) and in these encounters any number of anti-democratic social practices can interrupt democracy without disturbing its form.

This is as true of the UK and the US as it is of India and Indonesia. What differs, I would argue, contrary to Chatterjee, is not the number of people who do or do not engage in ‘the actually existing form’ of civil society, but rather the types of pragmatic social practices that interrupt the abstract procedures of civil society, be they nepotistic, hierarchical, or gendered. What differs, in other words, is the manner and modality of civil society’s failure rather than its extent.


Modernity, which includes civil society and democracy, is unique in that it is a social system that effaces its own pragmatic nature through social processes of abstraction,4 but it is impossible for modernity to completely efface the context-bound moments of everyday life from which it is built;5 erasure always leaves a smudge. Having said that, there is something radical about the way Balmiki – and most Indian – civil society practice exposes the truth about democracy. In its very obsession with titles, it takes the procedures of democracy and turns them into formality, that is to say, mere form. This desiccation of civil society serves, I argue, as a reminder of the concrete social relations that civil society and democracy must efface in order to realize themselves. For this reason, liberal-modernists perhaps find these practices disturbing (see for example Clarke 1998b; Ndegwa 1996).


It is in this context that Balmiki politics can be read as at once a part of civil society and a critique of it. To elaborate this point, I turn to Aditya Nigam’s argument that Dalit political action is unknowingly a critique of modernity precisely because it challenges the abstract and universal notions of humanity, citizenship, etc. upon which the discourses of civil society and democracy depend. Nigam argues that, while espousing a thoroughly humanist ideology, Dalit politics:

‘...represents in its very existence, the problematic "third term" that continuously challenges the common sense of the secular-modern. This resistance to these categories of modern politics is, at its core, a resistance to the very universalisms that characterise the emancipatory discourses of modernity which placed at their very centre the abstract, unmarked citizen – Universal Man – or the equally abstract "working class", as the subject of history. Dalit politics in my reading is deeply resistant to both the ideas.’ (Nigam 2000).

This is a familiar theme in North Atlantic race and gender politics (see Hooks 1981), but what is not clear in Nigam’s formulation is why the category ‘Dalit’ is not simply a new abstraction. While ‘Dalit’, like ‘African-American’ or ‘woman’, is non-universal in so far as it demarcates a subset of all possible members, it is nevertheless another abstract social category that homogenizes its constituents. The Combahee River Collective famously dealt with this political problem when it boldly proclaimed to a white, middle class, heterosexual feminism that not all women were equal in their oppression (Combahee River Collective 1982). Similar ‘new social movements’ and difference-based politics have developed in India as well (Omvedt 1994). The new politics addresses serious issues, but its reasoning steers dangerously close to a paralyzing problem: that of infinite differentiation. Once what Nigam calls the ‘insurrection of little selves’ (2000) starts, how do you stop the potential schizophrenia long enough to have a political movement?


This is, I believe, where the specificity of caste becomes relevant. By its very nature, caste references postulate potentially endless division. Banias, for example, are divided into 30 to 40 ‘subcastes’ with names such as Disawal, Kapol, and Modh that are in turn further divided (Shah and Desai 1988). Government categories such as ‘SC’, ‘ST’, and ‘OBC’ are, of course, (politically contentious) lists of historically underprivileged groups and are thus divided by nature. Ambedkar argued that caste was particularly pernicious in its capacity to endlessly divide and subordinate political subjects and hence simple class based ideologies like socialism were inappropriate for India.6 This fissiparous propensity has always plagued the formation of a Dalit identity;7 Balmiki politics exemplifies the problem.

In Delhi, Haryana, Punjab and other areas where Balmikis are strong, the Balmiki activists will argue that when Mayawati and others invoke the word ‘Dalit’, they are in reality referring to Chamar-Jatavs, who enjoy a virtual monopoly on reserved seats. The Balmikis argue that, due to the dominance of Chamar-Jatavs among Dalits, a separate set of reservations is now necessary. In Haryana, reservation policy divides SCs into a ‘Block A’ and a ‘Block B’ as a way of insuring that not all the reservations go to the relatively prosperous among the underprivileged of that state.8


In challenging the legitimacy of one sub-universal (Dalit) and replacing it with another (Balmiki), the Balmikis force our attention to the absurdity of infinite differentiation. They are politicizing one division and it is not hard to imagine the SCs and STs who get left behind by the ‘B’ list asking for yet further divisions. More pointedly, it is not hard to imagine some future world in which the former Mehtars of Delhi politicize their identity against the relatively prosperous ‘sheheri’ Balmikis.

Balmiki politics, then, is not marking an unmarked category but rather an already marked one. To a conservative, this mincing of political identity is reason enough to ridicule the very idea of positive discrimination as a hopelessly flawed project (see Shah 1996). To a progressive, such fissioning goes too far and destroys the unity of social movements. To both sides, there is something absurd about dividing a division. But the absurdity is not the fault of the Dalits. Rather, it lies in the contradictions of modernity itself. In order for the liberal notion of citizenship to have meaning, a level playing field must be created. But the creation of the level playing field through positive discrimination leads to the problem of infinite differentiation and hence contradicts the very nature of citizenship as an institution of equality and universality.


This contradiction is no coincidence. Abstract universalisms by their very nature erase the concrete, context-bound particularities of social life. It is precisely in the latter that the most pernicious and intractable mechanisms of oppressions exist, be they of caste, gender, community, race, or what have you. The only way to rectify this situation within a modern welfare/developmentalist state is through positive discrimination which, in turn, requires new forms of abstraction (like the identity of Dalit itself). It’s a vicious circle: the only solution to the ‘violence of abstraction’9 is more abstraction.

That said, we must not abandon ourselves to a romantic and ultimately impossible politics of anti-modernism. As Nigam argues, Dalit politics and new social movements do not obviate citizenship as an abstract universalism but actually make it more meaningful by forcing it to deal with social reality. I agree, but not only because positive discrimination helps produce a more level playing field. The ‘insurrection of little selves’ embodied in the Dalit critique of nationalism – and in the Balmiki critique of the Dalit movement, and in the gender critique of both, and in all the would-be identities that are struggling for political recognition – makes the concept of citizenship more meaningful because the very discomfort produced by its threat of infinite differentiation is a nagging reminder of the social reality that modernity tries to erase.


I have described two different ways in which Balmiki politics troubles the liberal-modernist ideas of what a political movement should be. The first is its reduction of civil society’s form to mere formality and the second is its potential to infinitely divide political identities. Both tendencies may appear as ineffective, inefficient and self-defeating and to some extent they are. But neither Balmikis nor Dalits, nor any other oppressed people need to change their tactics. Their politics should not be seen as flawed but rather as a complex response to the contradictions of modernity that does not fall prey to the sirens of an appealing but futile anti-modernism.



Ambedkar, Dr. B.R. 2002. The Essential Writings of B.R. Ambedkar edited by Valerian Rodrigues. Oxford University Press, New Delhi.

Butler, Judith. 1993. Bodies That Matter: On the Discursive Limits of Sex. Routledge, New York.

de Certeau, Michel. 1983. The Practice of Everyday Life, translated by Steven Rendall. University of California Press, Berkeley.

Clarke, G. 1998. ‘Non-governmental Organizations (NGOs) and Politics in the Developing World.’ Political Studies XLVI, 36-52.

Chakrabarty, Dipesh. 2000. Provincializing Europe: Postcolonial Thought and Historical Difference. Princeton University Press, Princeton.

Chatterjee, Partha. 2001. ‘On Civil and Political Society in Post-colonial Democracies’ in Civil Society: History and Possibilities edited by Sudipta Kaviraj and Sunil Khilnani. Cambridge University Press, Cambridge.

The Combahee River Collective. 1982. ‘A Black Feminist Statement’, in All the Women Are White, All the Blacks Are Men, But Some of Us Are Brave: Black Women’s Studies, edited by G.T. Hull, P. Bell Sciott and B. Smith. The Feminist Press CUNY, New York.

Hanisch, Carol. [1969] 1978. ‘The Personal is Political’ in Feminist Revolution: Redstockings of the Women’s Liberation Movement edited by Kathie Sarachild. Random House, New York.

Harper, Ian and Christopher Tarnowski. 2003. ‘A Heterotopia of Resistance: Health, Community Forestry, and Challenges to State Centralization in Nepal’, in Resistance and the State: Nepalese Experiences edited by David N. Gellner. Social Science Press, New Delhi.

Hooks, Bell. 1981. Ain’t A Woman: Black Women and Feminism. South End Press, Boston, MA.

Omvedt, Gail. 1994. ‘Peasants, Dalits and Women: Democracy and India’s New Social Movements.’ Journal of Contemporary Asia. 24,1:35-48.

Ndegwa, S. 1996: The Two Faces of Civil Society: NGOs and Politics in Africa. Kumarian Press, West Hartford.

Nigam, Aditya. 2000. ‘Secularism, Modernity, Nation: Epistemology of the Dalit Critique.’ Economic and Political Weekly, 25 November (Web version).

Sayer, Derek. 1987. Violence of Abstraction: the Analytic Foundations of Historical Materialism. Blackwell, Oxford.

Sayer, Derek. 1991. Capitalism and Modernity: An Excursus on Marx and Weber. Routledge, London/New York.

Shah, A.M. [1991] 1996. ‘Job Reservation and Efficiency’, in Caste: Its Twentieth Century Avatar edited by M.N. Srinivas. Penguin, New Delhi.

Shah, A.M. and I.P. Desai. 1988. Division and Hierarchy: An Overview of Caste in Gujarat. Hindustan, Delhi.

Weisgrau, Maxine K. 1997. Interpreting Development: Local Histories, Local Strategies. University Press of America, Lanham.



1. My fieldwork was conducted primarily in a north-west Delhi slum. In order to protect the identity of my informants, neither their names nor the names of their organizations will be used in this essay.

2. Similar observations have been made by Wiesgrau (1997) and Ian Harper and Christine Tarnowski (2003).

3. This is a long standing theme of feminist politics embodied in the slogan, ‘the personal is political’ (see Hanisch [1969] 1978). A theoretical take on this point can be found in Butler (1993).

4. Derek Sayer argues a virtually identical point (1991).

5. Dipesh Chakrabarty (2000) makes the point in his analysis of ‘the two histories of capital’ that ‘History 2’ is not transparent to ‘History 1’ but nevertheless registers itself indirectly, through ‘traces’.

6. This is a recurring theme in Ambedkar’s writing; see especially ‘Annihilation of Caste’ and ‘Caste, Class and Democracy’ in Ambedkar (2002).

7. Manohar Yadav makes this point in an editorial titled, ‘Dalit "Sahibs" and Masses’ in the The Deccan Herald, 28 June 2002 and reprinted at

8. In E.V. Chinnaiah vs. A.P. (5 November 2004) the Supreme Court ruled that states cannot further sub-divide the SC/ST category: But (the state) cannot take away said benefits on the premise that one or another group among the Scheduled Castes has advanced and, thus, it is not entitled to the entire benefit of reservation. The impugned legislation, thus, must be held to be unconstitutional. (C.A. no. 6758/2000, p. 346).

9. The phrase comes from Derek Sayer (1987).