THE past year (2010) saw considerable heat generated by the debate on the issue of a caste census. The present author too participated in that debate writing in favour of a full caste census,¹ but with an emphasis slightly different from others who were demanding the same. In continuation, this essay argues that the issue needs to be seen in a larger context of the Indian state’s approach to the caste question. Further, that a plea for a full caste census needs to be followed up by an overhaul of our approach to the caste issue and its relevance to public policy-making.

Controversies about the caste question abound despite an overall consensus that caste inequalities should be corrected, that discrimination on grounds of caste should be disallowed, and that permanent identification of individuals with the ascriptive identity of caste is incongruous with norms of individual freedom and the evolution of a modern society. Controversies focus on ‘how to do’ all this. Broadly, there are two diametrically opposite positions, one believing that by not recognizing caste in the public domain we can address most of these objectives; the other equally firmly believes that only by approaching most, if not all, issues of public policy through the prism of caste can we address the objectives and move towards a casteless social order.

Thus, the one tends to begin with a ‘casteless’ approach to effect caste-lessness in reality and the other adopts a ‘caste approach’ to ultimately arrive at a casteless situation. The former does not realize that in a caste-ridden social order, castelessness cannot be an advisable option for public policy formulation, while the latter refuses to recognize that the prime objective of public policy cannot be to enforce a casteless social order, that public policy can only ensure prohibition of discrimination and redressal of inequalities emanating from the caste situation – thus facilitating/encouraging castelessness but pending that, thwarting the march of caste based inequalities.

 

While this positioning and resultant public debate over how best to formulate appropriate policy continues, actual public policy in India has often been confused and marked by equivocation on the question of caste. To be sure, we take pride in the regime of affirmative action and the many schemes directed towards the backward sections, but their overall journey remains complex, often lacking in direction and purpose, and somewhat ad hoc. This is mainly because of the reluctance to adapt our ‘modern’ public policy frameworks to existing structures of social discrimination.

Having adopted the affirmative action regime for the SCs and STs, the Indian state prevaricated and did not ensure effective implementation of these policies for the first two decades.² It also refused to take on board more systematically the issue of backwardness resulting from the caste system – evident in both a lackadaisical treatment of the caste question by the first backward classes commission and subsequent response of the government. Then, we witness some effort at effective implementation of the reservation policy for the SCs and STs, followed by sidelining of the report of the second backward classes commission for almost a decade. That it was resurrected by the contingencies of competitive politics is another matter.

 

The fact remains that the Indian state lacked any concrete policy approach to the issue of caste based backwardness and inequalities arising from that. The nineties did witness a momentary spurt to caste based policies in response to the rising political relevance of the groups that constellated as OBCs in many parts of the country. It appeared that finally the state was waking up to the need to evolve a policy approach taking into consideration social facts rather than responding on a piecemeal basis to political pressures.

However, as was the case with the NF government in 1990, the UPA government too adopted a narrow and instrumental approach to the caste question when it ushered in what came to be popularly called ‘Mandal II’ (introduction of OBC reservations in institutions of higher education under central government). Therefore, the argument that the Indian state has often adopted strategies of accommodation/manipulation applies well to the policy on caste.³ The Arjun Sengupta Commission Report⁴ (National Commission for Enterprises in Unorganized Sector – NCEUS) provided yet another opportunity to both take the caste question seriously as also to broaden the scope of the debate over underprivilege associated with caste status. Of course, the NCEUS restricted itself to existing administrative categories of SC, ST and OBC rather than go into a more sophisticated categorization and/or examination of the social composition of the workforce and social distribution of the underprivileged. Here was a clue to a robust policy approach towards the caste question, but the government let the opportunity slip.

The real test of the UPA’s approach to the caste issue came in the wake of the question of a caste based census. The confusion that the government has managed to create around the question of caste census is part of a six decades old muddle: we have often adopted caste in our public policy considerations only as an apologetic addition. (The Mandal moment was an exception, but it failed to sustain. And that moment too was marked more by manipulation, accompanied by unexpected political circumstances, rather than a conscious policy shift.)

 

The debate about a caste census symbolically marked the dissipation of that moment, represented by frantic efforts by one section to insist on ‘numbers’ and an uncanny response devoid of ability to grapple with the issue. That we as a nation have (almost) lost the third opportunity in quick succession to shape a public policy approach to the caste question (first two being the ‘Mandal II’ episode and the Sengupta Commission Report) indicates the confusion surrounding the caste question as also the possible dissipation of our imaginative energy.

 

The proposal to count castes is not new: the Mandal Commission had asked the Home Ministry to undertake the task in the census of 1981 and that was turned down. In 2001, too, the Home Ministry rejected a similar proposal by the then Registrar General of Census. This time around, arguing for a caste census, the political class by and large limited itself to the ‘numbers argument’: that we need information on the number of OBCs in order to defend the existing quota if not perhaps an expansion.

On the one hand, this argument was a response to the repeated suggestion by the court that government should come up with data on proportion if it expects the judiciary to rule on the issue of crossing the fifty per cent threshold, as many states hope to do. The anxiety about numbers was also a response to the reported figures (often in distorted or uninformed manner in the media) from the NSSO (National Sample Survey Organization) that OBCs account for around 30 per cent as against the received (adopted) wisdom that they account for 52 per cent of the population.

But more than these concerns are the fallouts of the fact that the OBC terrain is becoming increasingly competitive. No single party can reap the harvest of OBC mobilization. Besides, there is also a marked fragmentation of the OBC constituency. These developments created some concern among political actors with a social base primarily among the OBC sections. Hence, their emphatic demand for an enumeration of the OBCs as a potential rallying point for those sections currently grouped under the OBC category.

 

The divisions among academic circles too are not new. Since the days of ‘Mandal’ there has been an oversimplified division. One set sees caste based measures (and therefore caste census) as deeply divisive, counterproductive and as legitimating traditional ascriptive markers (and hence antithetical to the modern democratic project). Then there are those who welcome the mobilization based on caste, see the necessity of caste based policy measures and, therefore, recognize the importance of caste as a factor in public policy. It would be somewhat grotesque to claim that all those opposed to caste based measures and caste census are casteist reactionaries. For instance, even some with sympathies for the anti-caste project have came forward in opposition to caste census, notable among them, Anand Teltumbde.⁵ 

But the debate over the issue of counting castes this time around also had another twist: the supporters of caste census themselves are sharply divided. One section took the view that as in the case of SCs and STs, the exercise should be restricted only to counting the OBCs rather than carry out a full caste census (Yogendra Yadav and D.L. Sheth). Others argued for a ‘full’ caste census.⁶ Yogendra Yadav, though, does not raise any fundamental objection to a full caste census but probably sees that as unnecessary and perhaps unviable at this point of time⁷ while Sheth launches a strong criticism of the idea of a full caste census calling it an ‘absurd proposal which is …least desirable.’⁸ 

In a tame anti-climax pertaining to a very important policy question, it has now been decided that a separate enumeration exercise will be done wherein the caste of all citizens will be recorded. This means that supporters of a full caste census have won a pyrrhic victory – caste will be counted but that information will be practically useless for purposes of arriving at any informed policy on the issue of caste based inequalities. We will surely get numbers, but without the benefit of collating those numbers with a host of social demographics in order to be able to talk about issues of structural disadvantage and backwardness.⁹ 

 

On the sidelines of serious debate we also come across occasional efforts to trivialize the idea. Dipankar Gupta’s criticism is a case in point.¹⁰ He seems to hold the view that caste differences are similar to differences between the bald headed and the non-bald. As such, there is no need to count caste and if anyone wanted to count it, they should do the hard work themselves. Such a casual treatment of an important social phenomenon can arise only from a lack of understanding if not caste bias. Moreover, Gupta misses the main point when he accuses some scholars of trying to get the Census do what (collection of data) they should be themselves doing. It is not the fancy of some academics that drives them to ask for caste census; caste constitutes an important structural factor affecting inter-group relations and asymmetries of power.

 

Second, analyzing this data is not the pastime of some intellectuals but a crucial input for policy (even if one wanted to do away with caste based policies, such data would still be necessary for justifying that shift). Caste is an important factor that is taken into consideration either in defining backwardness, for identifying the backward sections or for targeting specific schemes. Therefore, it becomes necessary that the government puts in place a mechanism to enumerate caste and also periodically monitor the material movement of the various caste communities. Caste census is one preliminary step in this direction.

But won’t an ‘OBC census’ alone be sufficient (as Yogendra Yadav suggests), since we have affirmative action policies only for OBCs, apart from SC and ST? Unlike the SCs and STs, the administrative nomenclature ‘OBC’ has not achieved full resonance among the groups included in that category. This is partly because castes included in this category are materially differentiated and mostly without a common history of political mobilization. Besides, many castes included in the OBC category by the Mandal Report are differently sub-categorized in many states and in so far as this sub-categorization facilitates reservations at the state level, acquaintance with that identification is much more in actual practice (for instance, DNTs, MBCs, and so on). Therefore, even if we wanted only an OBC census, doing that by first asking citizens if they belonged to the OBC category and then getting this cross-checked will leave out a significant portion just because they might not, in the first instance, report being OBCs and hence will not be counted.

By counting caste, do we run the risk of entrapping individuals in caste identities? Enumeration may strengthen caste identity. However, there is a difference between the identity that may temporarily consolidate due to enumeration and the identity that entraps. Political mobilizations, socio-cultural universes in which individuals operate, and the existing regime of affirmative action, all contribute to ‘entrapment’. Counting OBCs is sure to strengthen a so-called (and non-existing) OBC identity that might entrap. On the other hand, counting caste could only boost a localized, caste-specific identity which anyway gets sustenance from cultural and family practices. In this sense, the meaningful alternatives are ‘practiced identity’ vs. ‘professed political or bloc identity’.

 

It must however be conceded that caste is a complex and slippery phenomenon and counting it is no simple affair. Caste names pose a challenge and determining which caste name denotes what status is also difficult. In the case of religion, since the core social division is around ‘majority’ (Hindu) and various larger ‘minority’ religious groups, it matters less how some persons define themselves by naming identities that are beside the main religions. In the case of caste, we shall be much more interested in the social situation in which that caste lives (social hierarchy); whether it belongs to traditionally received categories (Varnic hierarchy) and, above all, whether it fits into an existing administrative category. This would surely be a challenging task for students of Indian society and for policy-makers.

Why should we take up this challenge? Anand Teltumbde has a point when he argues that the post-independence political economy has resulted in certain advantages to some sections of the OBCs.¹¹ We also need to accept that the category of OBCs is marked by tensions and contradictions. Within the framework of identifying backwardness established by the Mandal Report, inclusion in the OBC category is fraught with political contestation as there is constant pressure for new inclusions.

 

Therefore, a full caste census can open up possibilities of a larger debate, as also more reliable data for such a debate. Nevertheless, while caste continues to be a crucial factor in understanding inequalities in the Indian context, governments have always adopted ad hoc policies under political pressures rather than arrive at an overall approach to address this question. Therefore, data collection and decision-making always remain haphazard and hence debatable.

Post Mandal, most political parties now agree with the proposition that caste is important in understanding backwardness. In this sense, the present moment (and the opportunity to do a caste census) offers the possibility of shifting the policy approach away from casualness, and provides an opportunity for critiquing and modifying the status quo on this issue. The status quo favours a situation where the politicians capitalize on the OBC category as a resource for bloc politics; it also favours those sections of OBCs that have benefitted from state policies and capitalist development. And this status quo has left many sections both deeply dissatisfied and left unattended by state policies.

Many groups continue to claim inclusion in the OBC category, but this process is rarely fair because of the politicization of the issue of inclusion. As the Supreme Court mandated in the Indra Swahney ruling, in 2003 we should have reviewed which of the groups included in OBC category deserved to be excluded, but this did not happen. Many communities demand a change from OBC to SC or ST category; both among SCs and OBCs, there are groups that complain of a raw deal and ask for a regrouping or sub-classification. All these complications call for a larger investigation into the questions not only of the procedure to include groups into the OBC category, but also of defining and detecting backwardness.

 

Contemporary competitive politics and the movement sector are unlikely to be ready to handle these emerging issues mainly because the strategy of affirmative action has over the years also become a strategy of political mobilization on grounds of backwardness and a common interest in the administration of affirmative action. Thus, each administrative category adopted by the affirmative action regime tends to spill over into the political domain and acquire the status of a political category. They become instruments of political mobilization and hence get fossilized – leaders of each beneficiary group and the political parties claiming to represent the backward communities treat the administrative categories as more or less permanent, unchallengeable and sacrosanct.

But a two decades long politics of Mandal has also shown that such political consolidation is bound to be weak, tentative and does not necessarily strengthen democratic politics because it traps persons into artificially constructed blocs. Thus, we are confronted with the twin reality of attempts to build political blocs and the constraints such an exercise faces in the long run.

 

This impasse can be resolved by opening up the caste question to a larger and comprehensive inquiry. A full caste census can be the beginning of such an inquiry. Over the years, this exercise of caste census (if done in the manner in which collation of caste with other data on demographics is possible) can help us map out the characteristics of our collective self-portrait,¹² indicating the correlates of inequality and pockets of underprivileged sections. It is not argued here that such a collective self-portrait in itself and alone can correct our policy direction. It will only be the beginning of a long and tedious process. In the long run, a full caste census can be useful if three logical steps are taken following it or along with it.

The first and perhaps the most crucial step will be to appoint a third backward classes commission with a comprehensive brief to report on backwardness arising from our social structure. It should not be seen only as an OBC commission, but a body debating and recommending a policy approach to both – the overall social character of backwardness and the measures for neutralizing the handicaps. Needless to say, such a commission should address the issue of what benefits have accrued to which sections of the backward communities through reservations, how far the political economy has helped them fight free of traditional structural deficits (as for instance the recent study by Devesh Kapur, Chandra Bhan Prasad and Shyam Babu¹³ ) and so on.

Second, small scale case studies for in-depth investigations into the issues of caste-occupation linkage, occupational mobility among castes¹⁴ and so on will also be required and it is here that the academic community can extend help by designing and executing such studies. Third, it is high time that the Planning Commission awakens to the responsibility of investigating the caste question. After all, so many of our public policies take into account the caste factor. But we have so little data on caste (at macro level) that can form a reliable input in policy-making and more than that in framing an approach to policy. This can happen only if the Planning Commission sheds its ostrich-like attitude and takes up the responsibility of designing a large scale, nationwide and long term research programme for a comprehensive study of contemporary caste.

 

Given the flip-flop on the question of a caste census, one doubts whether this government has the capacity to comprehend an opportunity far less the courage to seize it. But in the absence of such a comprehensive investigation into the caste question, a caste census, even if done in the manner in which many social scientists are pressing for, may at best yield only limited benefits for policy-makers.

 

Footnotes:

1. Suhas Palshikar, Enumerating the Backward: Need to Broaden the Debate and the Inquiry, in S. Japhet and Chandan Gowda (compiled and edited), Caste Census: Towards an Inclusive India. CSSEIP, NLSIU, Bengaluru, 2010, pp. 63-69.

2. Marc Galanter, Competing Equalities. OUP, Delhi, 1984 (pb, 1991), pp. 93-102.

3. See Rajeshwari Deshpande, State and Democracy in India: Strategies of Accommodation and Manipulation. Occasional Paper Series (III) 4, Department of Politics and Public Administration, University of Pune, 2005, see especially, pp. 36-41.

4. Report on Conditions of the Work and Promotion of Livelihoods in the Unorganized Sector, 2007, NCEUS, New Delhi, http://nceus.gov.in/Condition_of_workers_sep_ 2007.pdf

5. Anand Teltumbde, ‘Counting Castes: Advantage Ruling Class’, Economic and Political Weekly XLV(28), 10 July 2010, pp. 10-11.

6. A comprehensive argument on this behalf can be found in Satish Deshpande and Mary E. John, ‘The Politics of Not Counting Caste’, Economic and Political Weekly XLV(25), 19 June 2010, pp. 39-42.

7. Yogendra Yadav, ‘Why Caste Should be Counted in’, The Hindu, 14 May 2010.

8. D.L. Sheth, ‘Caste in the Right Mould’, Indian Express, 23 June 2010.

9. For a cogent criticism of the decision of the Cabinet, see M. Vijayanunni, Satish Deshpande and Others, ‘Caste Census: Senseless Separation’, The Hindu, 13 September 2010.

10. Dipankar Gupta, ‘And Now They Want a Caste Census’, India Today, 4 May 2010.

11. Teltumbde, op cit., p. 11.

12. Deshpande-John, op cit., p. 40.

13. Devesh Kapur, Chandra Bhan Prasad, Lant Pritchett, D. Shyam Babu, ‘Rethinking Inequality: Dalits in Uttar Pradesh in the Market Reform Era’, Economic and Political Weekly XLV(35), 28 August 2010, pp. 39-49.

14. Some illustrative studies are: Rajeshwari Deshpande and Suhas Palshikar, ‘Occupational Mobility: How Much Does Caste Matter?’ Economic and Political Weekly XLIII(34), 23 August 2008, pp. 61-70; and Divya Vaid and Anthony Heath, Unequal Opportunities: Class, Caste and Social Mobility, in Anthony F. Heath and Roger Jeffery (eds.), Diversity and Change in Modern India. Proceedings of the British Academy, 159. The British Academy and OUP, Oxford, 2010, pp. 129-164.