Caste count revisited

By Manabi Majumdar

ABOUT FIVE decades ago, to get beyond casteism, the framers of our Constitution first took account of caste while delineating the building blocks of social justice. Today, after 50 years of Independence, is caste still relevant to our social, political and economic life either as a `measuring rod' for determining social handicap and/or as a `social unit' providing the basis for identity formation and community aspirations, political mobilisation and representation, and, more generally, struggle for power? If, on the ground of relevance to India's past and present, caste is an inescapable focal variable for analysis, understanding and policy action, is it not desirable to furnish a caste-wise demographic account of the nation, similar to the currently available profiles by language, religion, location and gender? The desirability question pushes the enquiry further back to the attendant issue of feasibility which, in turn, has to be approached on two different levels. The first one is about method: should we conduct a census, i.e., a complete caste enumeration of the billion-plus, or should a decentrally designed, context-sensitive, sample survey be more worthwhile? The second one is about agency: given that caste is a `nebulous' category, is the Population Census or an independent agency drawing on multidisciplinary resources more equipped to capture its complexities and nuances?

These questions brought together a group of people from diverse disciplinary perspectives and professional backgrounds - administrators, anthropologists, constitutional experts, economists, judges, lawyers, policy advisers, political scientists, sociologists, statisticians, writers, present and past chairpersons of Backward Classes Commissions - in a seminar on `Caste Enumeration in the 2001 Census'.

The Madras Institute of Development Studies conceived of the programme and took the initiative in organising it. The University of Mysore, the Institute for Social and Economic Change, and the National Institute of Advanced Studies, Bangalore, were co-sponsors.

While the original Government proposal to recommence caste returns in the Decennial Census constituted the proximate backdrop for the meeting, the purpose of the seminar was to go beyond this momentary furore - with a likelihood of generating `academic light' rather than rhetorical heat. The idea was to confer on the desirability and feasibility of collecting reliable data on caste which would likely resolve the conceptual and policy riddle that besets the identification of backward castes and communities. The guiding premise was that even with authentic caste particulars the political settlement of the contemporary contentions vis-a- vis compensatory discrimination would not be easy to come by; but in the absence of useful data there is very little hope. Hence it is the `poverty of information', the group felt, which needs to be addressed squarely.

The endurance of caste is manifest in three distinct ways: (1) caste as a pernicious vestige of our discriminatory history; (2) caste as a positive and empowering identity for groups traditionally consigned to the bottom of the social hierarchy and (3) castes as `pressure groups' jostling for political representation and power. Put simply, in its twentieth century incarnation, caste has emerged as `a major recruiting ground of electoral politics'. This is anything but a testimony to a decline in the salience of caste. As a corollary, the group reasoned that it is time we revisited the rationale behind discontinuing caste enumeration in the Census. Caste tabulations were integral to Census operations between 1872 and 1931. But caste groups have not been counted in the Decennial Census since 1931. In the absence of required figures, several academicians and Backward Classes Commissions have extrapolated population figures of various communities based on the 1931 estimates. But the underlying assumption that all communities experienced equal growth rates renders suspect many of these estimates. Such shaky foundations for analysis and policy choice, the majority argued, should be questioned.

There was a view that collection of caste information would reinforce rather than weaken caste divisions and lead to large- scale caste conflicts. Also, it might result in legitimising caste. But several speakers countered that the society was no better for the lack of caste details as caste conflicts were none the fewer for it. Moreover, if relations between caste groups remain strained due to genuine or manufactured reasons, lack of information will not resolve the problem, but only mask it. Discussions on these issues are often uninformed, provocative and rhetorical precisely because of shortage of data. What are the potential gains from collecting caste data? It can help identify the numbers of those which require target group oriented programmes of social and economic amelioration. Policy can then be anchored in a reliable data base. Above all, it would enable authentic identification of the backward classes.

Having arrived at a clear `yes' to the question whether caste counts, the participants confronted the inevitable: can it be counted? Through analyses of past censuses, the speakers showed how localised, fragmented and fluid caste groups are; how social political movements effect changes in caste nomenclature; how names of distinct castes phonetically resemble; how two or more communities/groups adopt the same name; how some groups disguise into the identity of another community and so on, creating, in turn, the possibility of a schism between the self- representation of the respondent and the judgment of the enumerator.

Another set of complications arises out of the unintended effects of compensatory measures. The potential beneficiaries of welfare measures are not just passive `targets'; they think, choose and respond to policies. And their response may not be immune to incentive effects. People may willfully misreport their caste origin on tactical grounds to get a share of reservation benefits which they do not otherwise deserve. But do all these complexities indicate a case for dropping the idea of collecting caste particulars?

Some participants endorsed the proposal to reintroduce caste enumeration in the Decennial Census, since the latter is a statutory, all- India operation. A more forceful plea was made in favour of caste enumeration on the ground that it is a constitutional imperative to do so. However, the legal position on caste enumeration being unclear (the recent Supreme Court order is not categorical on the issue), the `Census-sceptics' focussed on the tractability of collecting data on caste on a national scale in a centralised manner - a generic, context- insensitive, all-purpose caste schedule is quite unlikely to be workable.

An alternative suggested was that the responsibility for gathering caste details should be entrusted to an independent agency. It was argued that only a decentralised agency, with an awareness and sensitivity to local contexts, can yield worthwhile socio-economic data on caste. The majority felt, that a well- designed sample survey, repeated periodically, will be a much better way of building an adequate data base for facilitating policy than a complete census.

The general opinion was that the Government should initiate steps to gather caste details and this should be completed before the end of 2003. The exercise should have legislative backing, maximum transparency, and the information collected should be made public.

Send this article to Friends by E-Mail

Section  : Opinion
Previous : Hamlet with the prince
Next     : A hidden agenda?