The idea of a national caste census might be abhorrent when the stated policy is to strive for a casteless society, but it will be useful to establish statistical justification for preserving caste-based affirmative action programmes. It may also be a legal imperative, considering that courts want ‘quantifiable data’ to support the existing levels of reservation. Political parties with their base in particular social groups may find a caste enumeration useful, if their favoured groups are established as dominant in specific geographies; or they may find the outcome inconvenient, if the precise count turns out to be lower and has a negative bearing on perceptions about their electoral importance. In this backdrop, the Union government’s assertion in the Supreme Court that a census of the backward castes is “administratively difficult and cumbersome” may evoke varying responses. There are two components to the government’s stand. First, it asserts that it is a policy decision not to have caste as part of the regular census and that, administratively, the enumeration would be rendered so complex that it may jeopardise the decennial census itself. Second, it cites the difficulties and complexities inherent in getting an accurate count of castes, given the mind-boggling numbers of castes and sub-castes, with phonetic variations and similarities, that people returned as their caste in the Socio-Economic Caste Census (SECC) conducted in 2011.
The Government has said data from the 2011 SECC were not acted upon because of “several infirmities” that rendered them unusable. Even in the Censuses up to 1931, when caste details were collected, they were wanting in completeness and accuracy. Further, the data contained 46 lakh different caste names, and if subcastes were considered, the ultimate number may be exponentially high. These points do merit consideration, and even those clamouring for a caste census cannot easily brush them aside. However, these need not mean that an enumeration of the social groups in the country is impossible. A caste census need not necessarily mean caste in the census. It may be an independent exercise, but one that needs adequate thought and preparation, if its ultimate goal is not for political or electoral purposes, but for equity in distribution of opportunities. A preliminary socio-anthropological study can be done at the State and district levels to establish all sects and sub-castes present in the population. These can be tabulated under caste names that have wider recognition based on synonymity and equivalence among the appellations that people use to denote themselves. Thereafter, it may be possible to do a field enumeration that can mark any group under castes found in the available OBC/BC lists. A caste census may not sit well with the goal of a casteless society, but it may serve, in the interim, as a useful, even if not entirely flawless, means of addressing inequities in society.