Browsing Through 51 Years of EPW | Caste and Castelessness: Towards a Biography of the ‘General Category'

As a modern republic, India felt duty-bound to "abolish" caste, and this led the State to pursue the confl icting policies of social justice and caste-blindness. As a consequence, the privileged upper castes are enabled to think of themselves as "casteless", while the disprivileged lower castes are forced to intensify their caste identities. This asymmetrical division has truncated the effective meaning of caste to lower caste, thus leaving the upper castes free to monopolise the "general category" by posing as casteless citizens.


The quickest way to map the terrain I wish to cover is to recount a joke that has been circulating on the internet. Popularised five or six years ago when the 93rd amendment to the Constitution introduced reservations for the Other Backward Classes (OBCs) in elite educational institutions, the joke goes as follows: India decides to send a space exploration team to the moon. Feverish negotiations begin immediately on the composition of the team, and after much haggling it is decided to include nine OBCs, six members of the scheduled castes (SCs), three from the scheduled tribes (STs), and, if there is any place left, two astronauts. This joke unintentionally offers us a deep insight into the central predicament of caste today. The insight is contained in the fact that the “astronauts” are not identified by their caste but only by their qualifications (as astronauts), whereas the quota-walas are identified only by their caste and not by their qualifications. In short, the joke correctly assumes that “we” will know the caste of the astronauts without being told, but will agree that it is irrelevant in the face of their qualifications, while simultaneously agreeing that though the quota-walas too would presumably have qualifications, these are irrelevant in the face of their caste. To put it differently, upper caste identity is such that it can be completely overwritten by modern professional identities of choice, whereas lower caste identity is so indelibly engraved that it overwrites all other identities and renders them illegible, along with the choices that they may represent.

This, to my mind, is the central predicament of caste today – its hypervisibility for the so-called lower castes and its invisibility for the so-called upper castes. Having started out at Independence with the common goal of transcending caste – an objective that hardly anyone dared to question publicly and almost everyone seemed to share – we appear to have reached a dead-end six decades later where society is split into two unequal and implacably opposed sections. For one section, caste appears to be the only available resource with which to try and improve life-chances in a game where the playing field is far from level. This section, which constitutes the large majority of the population, includes many disparate groups that nevertheless share an interest in caste-based politics. For the other section, which is far less numerous and (relatively speaking) much more homogenous, caste-qua-caste has already yielded all that it can and represents a ladder that can now be safely kicked away. Having encashed its traditional caste-capital and converted it into modern forms of capital like property, higher educational credentials and strongholds in lucrative professions, this section believes itself to be “caste-less” today. Not only is there no dialogue possible between the two sides, they are trapped in a perverse relationship where each is compelled to unravel the arguments knitted by the other.

What I would like to emphasise here is the mismatch in the public perception of the two groups. The story of the political encashment of caste is often told – indeed it has dominated public discourse over the past two decades. This is a noisy and raucous account, full of the rough and tumble of political contestation, and it has also attracted ample attention from social scientists, as attested by concepts such as “dominant caste” or “the Congress system”. The other story – that of the “extra-electoral” coup effected by the upper castes through the transformation of their caste capital into modern capital – is not so well known. Because it runs with the grain of the dominant common sense – which is for obvious reasons monopolised by the vocal upper caste minority – this story is almost unseen and unheard. That is, it is seen and heard in other garbs – it appears to be a story about something other than caste, like the story of nation-building for example, or the story of a great and ancient tradition modernising itself.

I want to suggest that one reason why caste has proved elusive is because we have not recognised the consequences of this asymmetry. While it is of course necessary to address the question of the lower castes and their demands for social justice, we will not get a grip on the contemporary complexities of this institution unless we pay close attention to its taken-for-granted side, namely the “naturalisation” of the upper castes as the legitimate inheritors of modernity. In brief, my contention is that caste can be understood only if we pay as much attention to it when it is invisible or infra-visible as we do when it is hypervisible or ultra-visible. Whether it is represented as a chosen goal or claimed as an actual achievement, castelessness holds the key to caste.

Therefore, my objective is to attempt an initial account – a brief biography – of the emergence and rise of the notion of castelessness and its main form-of-appearance in everyday life, namely the “general category”. Needless to say this is a preliminary and incomplete effort, a rehearsal rather than a performance. Such an effort must begin by asking how a journey (apparently) originating in a common starting point – the desire to “abolish” caste – could lead to such sharply divergent paths. There are two obvious places where answers may be sought. First, we must examine the starting point to check whether it was in fact common or shared, and the extent to which this was so. Second, we must examine the particulars of the initial part of the journey to check whether something happened along the way that magnified existing differences among fellow travellers or else manufactured new divisions. The next two sections of this essay attempt to take up each of these options in turn. Section 1 examines the apparently universal goal of “abolishing” or transcending caste and its many distinct strands in the decades leading up to Independence. Section 2 deals with the ways in which the constitutional ideals, legal norms and policy practices of the new republic tried to give expression to the variously understood objective of “abolishing” caste. The concluding section (Part III) speculates on the current and possible future trajectories of the “general category”.

1 The Provocation of Caste

Caste offers a paradoxical union of the overfamiliar and the poorly understood. As the unique institution that indelibly marked Indian society as fundamentally inegalitarian and therefore unfit for modernity, caste was the universal provocation. No Indian, and certainly no Indian wishing to claim modernity in any way, could remain indifferent to it. This response was pre-given by the encounter with modernity, that is to say, something had to be done about caste – it could not be allowed to continue “as is”. And this generalised urge to change caste, or to act upon it, was typically expressed by the term “reform”, which “proclaimed the existence of a community…of the enlightened, working in harmony towards improvement and “uplift” in the life of the nation” (Bailey 2008: 155).

However, this apparent commonality was very deceptive because of the divergence between implicit intentions and explicit rhetoric. Public statements about caste were more constrained by the normative pressures of modernity than communitarian intentions, which could always manage to create some space for manoeuvre. What this meant in practice was that the language in which political and social programmes were expressed was far more convergent than the divergent projects that these programmes actually contained. Even when these disparate positions eventually seemed to congregate around firmer terms like “abolition” they continued to subsume wide variations in perspective and intent.

Thus, when tracked through sites such as the Indian National Congress and its official resolutions, for example, it is clear that the public language in which caste was addressed acquired the motif of “abolition” very late and only through a slow and reluctant process. As Ambedkar has documented in his famous tracts What Congress and Gandhi have done to the Untouchables and Mr Gandhi and the Emancipation of the Untouchables.1 even after talk of “abolition” became common, it remained facile and was rarely accompanied by a concrete understanding of caste and the practical course to be followed to achieve its abolition. Moreover, caste appeared to be unique in the sense that it was the only all-encompassing institution that was slated for abolition rather than reform. The obvious comparison is with religion which, even when it admittedly harboured numerous “social evils”, could still be presented as possessing an indispensable positive residue well worth preserving. Finally, while “everyone” had religion including the colonisers and others who were undeniably modern, caste was uniquely ours and it seemed unquestionably “un-modern”, or, indeed, anti-modern.

In this sense, therefore, when speaking of the “abolition” of caste, reformist public rhetoric was leaning far ahead of its constituency which was still located well to the rear of the rhetoric. This ideological overhang is most clearly visible in the early stages of the campaign against caste, namely the last quarter of the 19th century and the beginning of the 20th century. The most prominent voices here are still those that are seeking the reform of caste groups qua castes. The practical measures advocated here require nothing more than the simplification of an over-intricate system and the dissolution of proliferating sub-castes in favour of a larger, more effective collective caste identity. Examples of such campaigns were many, calling for rationalising and modernising, say, the kayasths or the brahmins as a caste by promoting inter-dining and intermarriage among sub-castes and urging sub-castes to rise above petty rivalries and claims to precedence. Associated measures might include those that were part of the larger social reform agenda such as education of girls, raising the age of marriage, doing away with arcane taboos and superstitions, and so on. So, by and large, campaigns whose stated intention was to abolish caste were actually designed to prepare castes to meet the modern world as castes.

A second set of agendas was less parochial and attempted to address the severe disabilities that the caste system imposed on the lower and especially the lowest castes. These efforts matured at the national level into the “constructive programme” of the INC launched in 1922 soon after Gandhi’s virtual takeover of the Congress. One of the major themes of this programme was the campaign against untouchability, easily the most visible and damaging practice associated with caste. However, it is important to emphasise the self-imposed limits that this programme functioned under. One way of mapping the gradual and reluctant widening of the ambit of the anti-caste campaign within the Congress is to trace the evolution of Gandhi’s positions on caste. Gandhi, too, began with what was essentially a rationalisation and reform programme whose overall objective was to simplify the needlessly intricate system of castes into the four broad varnas. The legitimacy of varnashramadharma remained an article of faith with Gandhi that he gave up only towards the end of his life, after sustained interaction with powerful opponents like Ambedkar and Periyar and radical anti-caste groups like the Jat Pat Todak Mandal.

1.1 Abolition of Caste

At the start of this political and moral journey, we have Gandhi declaring in 1921, just before the launch of the “constructive programme” that: “The caste system is the natural order of society. [...] I am opposed to all those who are out to destroy the caste system”.2 From there, by a gradual process helped along by the constructive programme and escalating in 1932 after the Poona Pact and the launch of the harijan uplift campaign, Gandhi had arrived, by the mid-1930s at the view that “Caste must go”. Gandhi’s faith in the basics of the caste system, which he understood in terms of the doctrine of varnashramadharma, endured for nearly a decade and a half of his career as an activist against casteism. During this period, all that Gandhi desired was the purification and simplification of the system that would help it to rediscover its sanatana or eternal virtues. Despite some discomfort with the unjustifiable reliance on birth to determine varna or station in life, it is instructive from a contemporary vantage point to see just how far even such a committed political reformer was willing to go in his support for caste.

Similarly, striking instances of reformist political beliefs coexisting with extreme anxiety about “losing caste” were seen among early Tamil brahmins supporting the Congress and other pro-change groups in the south (Pandian 2007). The clearest evidence for the gradualism that attended Gandhi’s slowly evolving views are visible in the intercaste marriages that he began to advocate in the 1920s. As Mark Lindley (2002) has shown, intercaste initially only meant inter-subcaste and strictly intra-varna marriages. This slowly expanded in the late 1920s to intra-savarna marriages that could be across the three twice-born varnas. It took a significant and clearly difficult interregnum before Gandhi could bring himself to advocate the marriage of twice borns with the shudra castes. Ultimately – after the 1936 publication of Ambedkar’s famous text The Annihilation of Caste – Gandhi graduated to his most radical position of advocating intermarriage between harijans and caste Hindus. By 1946, two years before his assassination, he publicly declared that the only marriages that would be celebrated in Sewagram Ashram would be those involving a harijan bride or groom. Thus, even for the most famous campaigner against caste “abolition” in pre-Independence India could have – and did have – many meanings that evolved over time.

Finally, there was a third version of “abolition” that stood at the far end of the spectrum, a position represented by Ambedkar with his stated goal of “annihilation”. For him, mere intermarriage was necessary but far from sufficient to uproot caste. Unlike his more moderate allies in the Jat Pat Todak Mandal, who were enthusiastic advocates of intermarriage, Ambedkar wanted to attack the ultimate foundations of caste which he believed were located in Hindu religion and especially its revered scriptures. It was precisely his call to disown the scriptures that alarmed his hosts in the Mandal and led to the withdrawal of their invitation to address their national convention in 1936.

In short, the moral pressure faced by the institution made abolition the preferred motif for programmatic public utterances on caste. However, this apparent unanimity of purpose concealed a broad spectrum of attitudes ranging from revitalisation and rationalisation to annihilation. Ambiguity of language was desirable and even sought after because of the pervasive nature of caste. It was (and is) no easy matter to “abolish” an institution that affects almost all areas of social practice – an institution so broad and inclusive that it constitutes a “way of life”. In the final approach towards Independence, therefore, these varied positions tended to find expression in similar sounding phrases and slogans that were intentionally vague and imprecise. It is no surprise, then, that the Constitution makers should have carried these ambiguities into the founding document of the new republic.

But civil society was not the only active force working on caste during this period – the colonial state too was an important actor. Indeed, an influential strand of scholarship has argued that caste as we know it today is “a modern phenomenon, that it is, specifically, the product of an historical encounter between India and western colonial rule” (Dirks 2001: 3). Possibly the most consequential intervention of the state was its effort, via the Census of India, to enumerate caste. As has been argued by Bernard Cohn, Arjun Appadurai, Nicholas Dirks and others, the very effort to enumerate caste led to important changes, with the institution becoming progressively more and more “substantialised” and fixed than it had been previously. In the 1930s the enumeration question was also tied to the question of electoral politics as provincial legislatures were formed and a gradually expanding electorate was demarcated. Two events are particularly relevant from this decade, the Census of 1931 and the negotiations around separate electorates that culminated in the so-called Poona Pact of 1932.

The census is particularly relevant from the point of view of the emergence of “castelessness” as a possibility and as a conscious political and social desire. Writing in the chapter on “Caste, Race and Tribe” in the Census Report of 1931, J H Hutton, the census commissioner, observes:

As on the occasion of each successive census since 1901, a certain amount of criticism has been directed at the census for taking any note at all of the fact of caste. It has been alleged that the mere act of labeling persons as belonging to a caste tends to perpetuate the system...It is, however, difficult to see why the record of a fact that actually exists should tend to stabilise that existence. It is just as easy to argue and with at least as much truth, that it is impossible to get rid of any institution by ignoring its existence like the proverbial ostrich...(Census of India, 1931, Ch XII, p 430).
He goes on to wonder whether, by aggregating castes across regions wherever feasible, the census “may claim to make a definite, if minute, contribution to Indian unity”. But his most interesting revelations concern the “no caste” category which was specifically provided for in the Census of 1931. Nearly 19 lakh people seem to have made use of this category in 1931, with 98% of them being from Bengal. Although this amounts to a little less than 0.8% of the total population of Hindus in India, it is still true nevertheless that the number of “no caste” returns in 1931 is greater than that in any previous census.

There is evidence to suggest, therefore, that the possibility of refusing a caste identity – at least in response to the colonial state – was already well-established by the 1930s. However, we need other sources of socio-historical evidence to evaluate the precise nature of this response and the reasoning that lay behind it. On the other hand, the census results can also be said to demonstrate the pervasiveness of caste as a ubiquitous form of identity. As Kingsley Davis has shown, the vast majority of Indians were willing and able to state their caste, including most of those belonging to sects like the Arya Samaj or Brahmo Samaj that were actively opposed to caste. Caste was reported extensively by Muslims – indeed, well over 80% of them reported castes, with 133 castes being exclusively Muslim. Caste was also reported by Sikhs and to a lesser extent by Christians.

Equally important, if not even more so, is the question of the electoral significance of caste identities. With the Muslim demand for a separate electorate having already crossed the point of no return by the 1930s, intense attention was focused on the Depressed Classes. Gandhi and the Indian National Congress had particularly high stakes in this issue because of the way the numbers stacked up. According to the 1931 Census, Hindus accounted for 68.2% of the population of India, while Muslims made up 22.2%. Given that the “Exterior Castes” (mostly corresponding to the Depressed Classes) accounted for as much as 21.1% of the Hindu population, the grant of a separate electorate to them would greatly reduce the Hindu majority (roughly to under 54%). More importantly, this would be a major blow to the moral authority and hence the eventual political power of the Congress as the representative of “India” rather than only a caste Hindu minority. While there were strong inequities marking the relationship of even the so-called “interior castes” (or shudras) with the twice-born minority within caste Hindus, these divisions could be papered over and prevented from emerging into the open. However, the disabilities imposed on the Depressed Classes were so severe and shocking that no amount of propaganda could hide them. Thus, the distinctness of the untouchable castes was already an established empirical and political fact.

It is this fact that Gandhi was addressing in his negotiations over the question of separate electorates for the Depressed Classes being demanded by Ambedkar. By embarking on a pre-emptive fast unto death – the very first time that he had taken such a radical step – Gandhi ensured that Ambedkar would have no option but to succumb. The Poona Pact of 1932 thus cemented the claims of the Congress and specifically of Gandhi to represent all of India, thus helping to conceal the fact that the leadership was exclusively upper caste and the even more closely guarded “public secret” that these castes represented a very small minority of the Hindu population. The muting of caste identities was thus a necessary precondition for the construction of a Congress “majority” – a development of immense significance in the emerging era of electoral democracy.

However, a peculiar and paradoxical twist was imparted to this by Ambedkar’s vigorous championing of the untouchable cause. The Poona Pact agreed to significantly increase the guaranteed political representation for the Depressed Classes, but a very heavy price was paid for this “concession”, as Ambedkar realised only too clearly. Separate electorates could be seen as articulating a consociational demand for a full share in the nation, a demand that underlined the equal claims of the untouchables. Although this was not immediately obvious, the grant of reservations reduced the Depressed Classes to the status of supplicants for whom a special concession was being made by the majority that “owned” the nation. This effectively positioned the upper caste minority (which was in control of the majority) as the de facto owner of the nation, with the power to grant favours to this or that subgroup. It is this mindset that has shaped upper caste common-sense on issues of caste and especially reservations. This is also the origin of the hypervisibility of the lower castes, with the untouchable castes being at the extreme end of hypervisibility. Until the eruption of the “interior castes” in their avatar as the “Other Backward Classes” in the Mandal conflagaration of 1990, it was the dalit-upper caste axis that was central to questions of visibility and invisibility.

2 Caste, Constitution and Citizenship in the New Republic

It is to be expected that the birth certificate of the nation reborn as a republic – the Constitution of India – will bear the marks of the contradictions that have shaped it. Among the most significant contradictions are those that concern the vexed institution of caste (Galanter 1963, 1968, 1969, 1984). The central tension is between, on the one hand, the need to “abolish” – or at least to delegitimise – caste as an institution that affects all citizens; and, on the other, the commitment to redress the disabilities of caste imposed on one section of citizens, namely the lower castes. These conflicting demands – requiring in the first case that caste be derecognised, and in the second case that it be recognised – have to be accommodated within the overall framework of a “passive revolution”, that is, a revolution from above whose fundamental tendency is to minimise the impact of change on already entrenched enclaves of power and privilege. As the manifesto of the passive revolution, the new Constitution must be faithful to both terms. Being in some sense a revolutionary Constitution, it is full of radical good intentions. But it is also a passive or an orphan Constitution in the sense that “there is no class backing the Constitution with its iron will”, as Madhava Prasad (2011: 45) has written, so that it lacks “the will to change” and offers only “the letter of the law… without the spirit”. The legal career of caste in the passive revolution is thus shaped through the disparate effects of constitutional intention, judicial interpretation and the policy initiatives of the
new republic.

In colonial and precolonial India caste identities were compulsory for all – only those who renounced the world could be caste-less (Burghart 1983). Nationalist efforts to exorcise the embarrassment of caste succeeded to some extent in valorising a worldly ideal of castelessness, but they were unable, and also largely unwilling, to mount an all-out assault on caste. This ambivalence is translated into the Constitution through the inclusion of, on the one hand, the rights to equality and non-discrimination, and, on the other hand, the charge on the state to show special consideration to the STs and SCs, to “socially and educationally backward classes” (SEBCs), and more generally, to the “weaker sections” of society. The two kinds of entitlement are neither equal nor symmetrical. Being a fundamental right, the right to equality and non-discrimination takes precedence and is pre-emptive – the state’s duties towards the lower castes and weaker sections may be discharged only as “permissible abridgements” of this always-already established right.

Thus, the Constitution promises to redress the injustices suffered by the SCs and STs, and also to ameliorate the disabilities and disadvantages suffered by the SEBCs, but these promises are contained in the Directive Principles of State Policy that are not justiciable. The relative weight to be attached to these principles in comparison with the Fundamental Rights may be a matter for judicial interpretation, but the pre-eminence of the latter is never in any doubt. Moreover, to keep its promises to the SCs and STs the state must first recognise them as castes, and this in itself is sufficient to confine such initiatives within the bounds of a benevolent exception to the prior and stronger commitment of the state to not discriminate among its citizens on the basis of caste.

By contrast, the biggest boon that the state grants to the upper castes is a guarantee of anonymity in caste terms. This effectively means that regardless of the extent of their past or present privileges, their caste identity can never be used directly to prohibit or limit access to any public resource. In other words, the upper castes cannot be prevented from cornering a disproportionate share – or even all – of a public resource because they belong to caste A or B; their share can be limited only by setting aside portions exclusively marked for castes X and Y. But as we have seen, such an exclusive setting aside – or reservations – is already designated as an exception to the norm of non-discrimination and equality. From the perspective of the upper castes, therefore, the constitutional guarantees of equality and non-discrimination amount to a licence to capture unequal shares of public resources. This licence is limited only by two things, first the rules of the market or open competition, and second, the exceptional device of reservations. The most significant aspect of this licence is that it can be worked without having to name one’s own caste.

Although the commitment to redress caste injustice was integral to the social contract upon which the nation was founded, the new Constitution constrained the victims of caste to demand justice as a caste-marked exception, while its beneficiaries were empowered to demand the perpetuation of their advantages as a casteless norm. Neither route leads towards the annihilation or even the diminishing of caste; but in dominant common sense, one route is presented as having already passed this destination, while the other is accused of leading us away from it. In our time, it is the unbridgeable divergence of these two routes – and their unshakeable internal certainties – that make the “annihilation of caste” seem more like a disabling dream than an empowering utopia.

2.1 Victory in Defeat

Barely six months after the Constitution of the Republic of India was formally adopted, the Madras High Court upheld in July 1950 the plea of two brahmin petitioners, Champakam Dorairajan and R Srinivasan, who claimed that their fundamental right to equality and non-discrimination guaranteed by the Constitution were being violated by caste and community quotas then in force.3 Although the specific order being challenged – known as the Communal GO – predated constitutional reservations, these petitions also had an impact on the new legislation. The unanimous verdict of the full bench of three judges striking down the Communal GO sent shock waves through Parliament when the Supreme Court concurred with the high court in April 1951. The law ministry (then headed by B R Ambedkar) and the government (headed by Jawaharlal Nehru) responded swiftly with the first amendment to the Constitution protecting reservations in higher education with the same special proviso already included for job reservations. The first amendment was passed in June 1951, less than two months after the Supreme Court verdict, but the state was put on the defensive.

In reality the courts had been victorious in defeat. They had managed to firmly establish the primacy of the meritocratic norm over the aberrational status of social justice initiatives. At the same time the courts made explicit and endorsed a new kind of agency that the Constitution implicitly offered to the upper castes, an agency based on the universal-normative position of “castelessness”. This was, however, a presumptive castelessness – that is, it did not require the upper castes to “give up” their caste in reality; it simply assured them that they would be presumed to be casteless as long as they did not invoke their caste explicitly. In effect, the new Constitution forbade the state to name or act against caste-based privileges or advantages as long as they wore the garb of secular modernity. Constitutionally and legally, caste was henceforth to be recognised only as a source of disadvantage or vulnerability, not as a source of privilege or advantage. And when it was invoked as a liability (as in social justice legislation), it was promptly imprisoned in the straitjacket of a regrettable and hopefully short-lived exception to the meritocratic norm.

The arguments in the Dorairajan case in the Madras High Court bring this out very clearly. The court is told that if the Communal GO had not existed and selection to the roughly 400 seats in government engineering colleges were made solely on “merit”, i e, in terms of a ranking based on the marks obtained in the qualifying examination, then brahmins would have obtained 249 seats instead of the 77 they were allotted under their communal quota. The court sees this as clear evidence of injustice against brahmins, with no attempt to reflect on how a republic committed to ending caste inequalities ought to deal with a situation where a historically privileged community numbering 3% of the population would corner 62% of the seats in a state-subsidised engineering college. Nor does the court evince any interest in the facts that once admitted, all candidates would receive the same instruction, and would be held to the same academic standards in the qualifying examinations. Justice Viswanatha Sastri is both eloquent and unequivocal in his defence of caste-based advantages:

It may be that through the fortuitous operation of a rule, which in itself is not discriminatory, a special advantage is enjoyed by some citizens belonging to a particular caste or community. This advantage is not taken away by Article 15(1). If, for instance, students belonging to a certain community or caste by reason of their caste discipline, habits and modes of life, satisfy the prescribed requirements in larger number than others, it is not permissible to shut them out on that score (para 44).


It would be strange if, in this land of equality and liberty, a class of citizens should be constrained to wear the badge of inferiority because, forsooth, they have a greater aptitude for certain types of education than other classes (para 54).
(Madras High Court 1950)
2.2 Unmarking

Clearly, what is taking shape here is not the “general category” as such, but rather its immediate ancestor, namely an explicitly caste-marked identity that has suddenly been freed of its particularistic burden by the “fortuitous” advent of a historically unprecedented category – the unmarked universal citizen. Pictured in this liminal moment at the cusp of tradition and modernity, and posed (so to speak) with its progenitors, the new republic and its freshly minted Constitution, this figure is clearly recognisable as brahmin even as it proudly wears the new clothes of the citizen. These are still clothes; they are not – or not yet – a disguise or costume, and the upper caste subject is at this early moment in its career remarkably uninhibited about exhibiting signs of caste belonging. But already, even at this inaugural stage, there is an awareness that “in this land of equality and liberty” the public declaration of upper caste identity has been made voluntary, and that this could be a decisive tactical advantage. Unlike the compulsory marking of lower caste identity which the new republic perpetuates and intensifies, upper caste identity may now be declared or not at will. Most important, the privileges and benefits that accrue to the upper caste identity may now be accessed anonymously, while its political-moral debts and liabilities are written off by the new Constitution.

The broader consequence of these changes is that the welfare of the upper castes need no longer be pursued in visible fashion through the mediation of public politics. It can now be made congruent with impersonal collective goals like nation-building, development, or later in the story, by equally anonymous forces like the market or globalisation. While it is not perfectly symmetrical, the situation of the lower castes is a study in contrasts – the pursuit of their interests requires the mandatory mediation of public politics, and their needs must usually be articulated as particularistic demands. In brief, upper caste interests go with the grain of development and the market and appear to involve the exchange of equivalents, whereas lower caste interests appear as transfer payments that can only be justified as exceptions. But what is taking shape here is not casteless egalitarianism, but rather an unequal “democracy” of castes qua castes, as Suditpa Kaviraj has suggested (Kaviraj 2011: 291).

It is important not to exaggerate the legal or policy impact of the 1950-51 judgments. Their immediate effects were largely overturned by the swiftly enacted first amendment. Also, the Dorairajan-Srinivasan case had little to do with the constitutional reservations for the SCs and STs; it was part of an ongoing local contest between the brahmins and the “non-brahmins”. The main impact of these early judgments was ideological, and in this respect it was considerable. By creating the conditions for the infra-visibility – ultra-visibility division the category of the unmarked citizen helped to mystify the category of caste and its social relations. In fact, it authored and disseminated a new kind of common sense where the very definition of caste was truncated and equated with the lower castes. This was in keeping with similar effects produced by other universal-modern categories invested with power – for example, patriarchal common sense reduces gender to women, while white supremacy limits race to non-whites or “people of colour”. The unmarked universal becomes the abode of normal, naturalised power, its transparent invisibility being a sign of its privilege in contrast to the compulsory markings that subaltern identities were forced to display. It took four decades and the emergence of the key category of the “Other Backward Classes” (OBCs) before the infra-visibility of the upper castes qua castes – or at least its normal-natural status – could be named and challenged.

3 After Mandal

The 1990s were probably the most momentous decade in the life of our republic, and one of its defining events is the social revolution now known simply as “Mandal”. The Mandal moment marked the long delayed arrival on the national stage of the critical category of the OBCs. Given its electoral weight and sociopolitical significance, this was like an avant-garde play where the Godot-like protagonist arrives very late and disrupts the narrative, forcing the audience into hurried retrospective revisions of the storyline. I will focus here on only those aspects of Mandal that are immediately relevant for the “general category”.

Although it might seem rather obvious in retrospect, the first consequence of OBC assertion for the general category was not immediately recognised. It took sometime for the realisation to sink in that, with the OBCs too being added to the “reserved category”, the general category had now become a euphemism for the upper castes. In fact this had been true since the advent of the Republic, because the general category had been comprehensively colonised by the upper castes and OBC participation in it was negligible. Of course, it was this very fact that had triggered the Mandal upsurge in the first place. Until this moment, the fiction of the “general” being the all-inclusive universal had been easy to maintain, given that reservations were locked into the exceptional mode from the beginning. The constitutional attempt to be “caste blind” had worked against the public naming of caste (outside the reserved categories), thus offering an “equal” anonymity to both the upper castes and the OBCs. But the unequal distribution of material and cultural resources had turned equal anonymity into severely unequal access to privilege and power.

One of the curious facts about Mandal – perhaps also the reason why it gets elevated into a rare instance of national catharsis – is the way in which it seemed to drive home lessons that should have been learnt already. Thus, the basic trajectories of OBC politics had already been traced in many regional contexts, notably in the southern states, for whom Mandal was just so much deja vu. Despite this, the national media and even academia seemed to realise for the first time that the upper castes who had been accustomed to regarding the general category as their ascriptive birthright were actually a minority while the reservation categories constituted the vast majority of the population. Even though the absence of reliable statistics on the OBCs fuelled initial scepticism and controversy, this conclusion was hard to resist. With the publication of separate statistics for the OBCs by the National Sample Survey Office at the end of the decade, many doubts could be resolved, and the logic of numbers was strengthened. What this brought to light was the long-forgotten “power sharing” or consociational argument for reservations. If power flowed from aggregated majorities in a democratic polity, then it was hard to explain why the distribution of opportunities could be allowed to be so far skewed in favour of a minority that it supported stark forms of durable inequality. And though the traditional meritocratic arguments continued to be advanced, they seemed to lose their lustre. Mandal provoked a re-evaluation of the symbolic and practical scope of the general category.

One instance of such a revaluation is in the so-called Mandal case of 1992. Here the Supreme Court consciously invokes the history of the checks and balances played out between non-discrimination, equal protection of the law, and the special charge on the state represented by the SCs and STs on the one hand, and the other unspecified “weaker sections” and the SEBCs. Coming full circle from the ratio of the Madras High Court in its Dorairajan and Venkatramana decisions of 1951 that quashed the Communal GO, the court reiterates that the unreserved or general category cannot be treated as a de facto quota for upper castes. It states unequivocally that those reserved category candidates who qualify to be included in the general category must be included in it – they must not be forced into the quota seats, nor can the size of the quota be reduced because of such inclusion in the “merit category”. Although, once again, this is not new (various court judgments had appeared to reach similar conclusions since 1958 (Galanter 1984, especially Ch 12)), there is something about the context that adds weight to this revaluation.

However, the most recent national level assertion of castelessness is that provoked by the proposal to enumerate caste in the Census of 2011. It is here that we see the media and civil society organisations mounting campaigns claiming precisely a casteless identity. Similar proposals to count caste in the 2001 Census had been summarily rejected by the then home minister L K Advani. The discussion at that time had remained largely confined to the pros and cons of gathering such data; castelessness did not emerge as a visible and vocal identity. Despite efforts to delay, block or otherwise scuttle the proposal, efforts which were ultimately successful, it did seem for a while that caste was actually going to be counted in the 2011 Census. It was this prospect that energised the upper caste elite and crystallised its claims to castelessness. For the first time, the anti-caste enumeration campaigns placed more weight on the claim of castelessness rather than on the other consequentialist arguments they had emphasised earlier (Deshpande and John 2010). One of the best known instances of protest came from the veteran actor Amitabh Bachhan who declared on his blog that census enumerators who came to his home would be told that the caste of its inhabitants was “Indian” and nothing else. Soon after, a new civil society initiative led by similarly inclined upper caste elites calling themselves “Meri Jati Hindustani” was launched. As its name suggests, the campaign urged citizens to join in the effort to scuttle the counting of caste by insisting on identifying themselves as “Hindustani”.

One sees the emergence here of a voice and a sensibility that is beginning to believe in its own castelessness. The fact that such claims invariably emanate from the upper castes – indeed from the elite among them – continues to elude proponents. Once they are successfully interpellated by the ideology of castelessness, upper caste subjects see their caste identities as incidental or irrelevant to the claim. They can thus assert with some sincerity that it is mere coincidence that everyone who makes such a claim happens to be from the upper castes. This is the generation that is (generally speaking) distanced from the process of the conversion of traditional caste capital into secular modern casteless capital that previous generations effected. It is objectively true that in the life-experience of such individuals – who, it must not be forgotten, may still constitute a minority within their own caste group – caste-qua-caste plays no direct role, or only a minor one. It is for this group – and this group alone – that family seems to have replaced caste as the source of social capital (Beteille 1991). Long accustomed to a comfortably homogeneous environment populated almost entirely by people like themselves, this group is unsettled by the recent arrival of hitherto excluded and therefore strange and unknown social groups in their vicinity. It is the double coincidence of the maturation of a sense of castelessness and the arrival of caste-marked strangers in hitherto upper caste social milieux that confirms and amplifies this response.


This is a good time, then, to be working towards a biography of the general category. The problem of false universals
is already known to us from feminist theory and from critical race theory. We can use the insights of this literature to understand how the general category has fared as a universal in our context. We can also examine the possibility
of reclaiming and repositioning this category in the light of what we have learnt. Can we imagine a different avatar of the general category as a
“true universal”?

It is in the nature of utopias to be ill-defined. So it is hardly surprising that B R Ambedkar’s cherished ideal, “the annihilation of caste”, remains to this day an inspiring but vague destination without a reliable route map. But located as we now are within the postnational condition, with the Nehruvian naivetes of yesteryear a distant memory, it is time perhaps to interrogate this utopia more closely. If one meaning of annihilation must be to render caste irrelevant as a determinant of life chances, then it is necessary to understand not only how a particular caste habitus might block or limit these chances, but also how another such habitus might enlarge or amplify life chances. Recent social science research offers us many accounts of the former but almost nothing on the latter. When it comes to the positive and productive facets of caste we have only broad correlations between outcomes; we lack detailed accounts of processes and modalities, the concrete ways in which an upper caste identity secretes and synergises the dispositions and embodied competences that add up to that abstract term: “merit”. To understand the productive side of caste we need not one but many detailed biographies of the “general category”. In the last analysis, then, the call to interrogate the upper caste self is not about the end of illusion as it might first seem, but about the revitalisation of what is perhaps our most intimate as well as our most elusive utopia.


This is a slightly edited – and very lightly referenced – text of the 12th Malcolm Adiseshaiah lecture delivered at the Asian School of Journalism, Chennai on 21 November 2012. As it forms part of a larger book project where there will be more space for proper footnoting and referencing, I have let this version remain almost exactly as presented in Chennai. This means that there are numerous fellow-travellers whose work I have benefi ted from even though I have been unable to cite them explicitly here. I am grateful to the Malcolm and Elizabeth Adiseshaiah Trust, and especially to C T Kurien, Rama Melkote, V K Nataraj and M N Shetty, for this honour. Those familiar with this fi eld will instantly recognise that my greatest intellectual debt is to Marc Galanter. For the critical engagement and encouragement that they have generously provided, I thank Mary John, Kalpana Kannabiran, V K Nataraj, Madhava Prasad and A Vaidyanathan.
Caste has been at the centre of public attention for a long time, especially in the last two decades. Despite being at the centre of our attention, however, caste continues to elude us in fundamental ways – or at least so it would seem. In this article I would like to explore some of the ways in which caste has proved to be elusive, and the reasons why this has happened.



1 First published in 1945 and 1943 respectively, both works are included in Ambedkar (1990).

2 The quotations in the following discussion are taken from Lindley (2002), who is himself citing various sources, including (as in this immediate instance) Ambedkar, the Harijan, the 100-volume Collected Works of Gandhi and other texts in the enormous literature on Gandhiana. I have verified that this particular quote is from Ch XI of Ambedkar’s What Congress and Gandhi Have Done to the Untouchables, and is to be found on p 276 of Ambedkar (1990).

3 For more detailed treatments of this famous case, see, inter alia, Galanter (1984:164ff) and Kannabiran (2012: 166-73).


Ambedkar, B R (1990): Dr Babasaheb Ambedkar: Writings and Speeches, Vol 9, edited by Vasant Moon, Education Department, Government of Maharashtra, Mumbai.

Bailey, Susan (2008): Caste, Society and Politics in India from the 18th Century to the Modern Age (Cambridge: Cambridge University Press). (Online version of 1999 print edition).

Beteille, Andre (1991): “The Reproduction of Inequality: Occupation, Caste and Family” in Contributions to Indian Sociology, n s, Vol 25, No 1, January-June.

Burghart, Richard (1983): “Renunciation in the Religious Traditions of South Asia” in Man, New Series, Vol 18, No 4, December.

Deshpande, Satish and Mary E John (2010): “The Politics of Not Counting Caste” in Economic & Political Weekly, Vol 45, No 25, 19 June.

Dirks, Nicholas (2001): Castes of Mind: Colonialism and the Making of Modern India (Princeton: Princeton University Press).

Galanter, Marc (1963): “The Law and Caste in Modern India” in Asian Survey, Vol 3, No 11, November.

– (1968): “Changing Legal Conceptions of Caste” in Milton Singer and Bernard Cohn (ed.), Structure and Change in Indian Society.

– (1969): “Untouchability and the Law” in Economic & Political Weekly, Vol 4, No 1, 1 January.

– (1984): Competing Equalities: Law and the Backward Classes in India, Oxford India Paperbacks (New Delhi: Oxford University Press).

Kaviraj, Sudipta (2011): “Democracy and Social Inequality”, Chapter 8 in The Enchantment of Democracy and India (Ranikhet and New Delhi: Permanent Black).

Kannabiran, Kalpana (2012): Tools of Justice: Non Discrimination and the Indian Constitution (New Delhi: Routledge), pp 166-73.

Lindley, Mark (2002): “Changes in Mahatma Gandhi’s Views on Caste and Intermarriage” in Social Sciences Journal (Ankara: Hacettepe University), Vol I, accessed at: http://www. Gandhis_views_on_caste_and_intermarriage

Madras High Court (1950): Srimathi Champakam Dorairajan and Anr vs State of Madras, Judgment delivered on 27 July 1950, Chief Justice Rajamannar presiding, with separate (concurring) judgments by Justices Viswanatha Sastri and Somasundaram.

Pandian, M S S (2007): Brahmin and Non-Brahmin: Genealogies of the Tamil Political Present (Ranikhet and New Delhi: Permanent Black).

Prasad, Madhava (2011): “Language, Education and Political Existence” in Seminar, No 624, August.

The Grid

In the context of the recent sexual molestation of an actor in a public space in Kerala, this article analyses Malayalam cinema’s language of neo-liberal governmentality that seeks to police gendered...
Click through the timeline above to explore a history of the Right to Information Act 2005 through the EPW Archives.
As a modern republic, India felt duty-bound to "abolish" caste, and this led the State to pursue the confl icting policies of social justice and caste-blindness. As a consequence, the privileged...
Read Economic and Political Weekly's special review edition featuring articles on environment and development. 
Since the 1970s, urbanisation across the globe has been shaped by corporate capital under the neo-liberal policies of the state. Cities are treated as consumer products with massive private...
Back to Top