Repeating a gigantic exercise does not make administrative, financial or practical sense.

The government's decision to collect caste data in a separate census threatens to nullify the potential benefits of the historic and progressive policy shift towards enumerating caste announced earlier. The Group of Ministers' recommendation to enumerate caste during the house-to-house census enumeration phase in February 2011 and not at the biometric camps was a welcome one. However, the Union Cabinet's present decision to schedule the caste census as “a separate exercise” is deeply disturbing. In its 140 year-history, the census has never collected caste data in a separate exercise.

If caste data collection is done as a stand-alone exercise in June 2011, it will generate only a headcount of castes. It will not be possible to integrate the caste data with the socio-economic, educational and demographic data like literacy, education, marital status, life expectancy, occupation, etc gathered during the census headcount. If all we get is a headcount of castes, then the entire exercise is futile. The main reason to enumerate caste is to enable the distribution of national resources and opportunities to be informed by reliable empirical evidence on the socio-economic levels of different communities now and in future. Without such evidence, all the problems blocking the implementation of social justice policies will remain unsolved.

Other reasons

There are other strong reasons against collecting caste data in a separate census from June 2011. June-September is not a suitable time for conducting a nation-wide census, since summer will be intense in most of India and monsoon will be on in the South. Since schools in most states will have reopened, the 21 lakh teacher-enumerators will not be available. Further, it will not be practically possible to mobilise the gigantic field organisation once again in June, within two months of the close of the census enumeration in February-March 2011. The whole exercise will be patchy and unsatisfactory.

Further, the separate caste census in June 2011 is expected to cost Rs. 2,000 crore as against the entire 2011 Census reported cost of Rs. 2,240 crore. When the caste data can be collected simultaneously in the census enumeration phase without added cost, duplicating this effort at such a high cost and with doubtful success is not the right decision.

If the government does persist with a separate caste census — and we hope it will not — it is imperative that the Office of the Registrar General of India ensures the integration of the caste data with the other indices collected during the census enumeration in February 2011, because only this will allow for correlating the data gathered by the separate caste census with the socio-economic, educational and demographic data collected in Census 2011. But this will still be less than satisfactory since a sizable section of the population (about 20 per cent) will be non-comparable due to shifting dwellings/seasonal migration.

The Cabinet's decision to constitute an Expert Group to “classify the caste/tribe returns after the enumeration of caste” does not address the real concern. A group of empowered real experts is indeed needed to monitor the process of caste enumeration and to safeguard it against possible confusions and derailment, but it should be appointed right now to oversee the entire process of caste enumeration and not merely for tabulating the collected data.

It has been alleged that collecting caste data will “compromise” the “integrity” of the census and “distort” the “population count itself.” The allegation that every family will inflate its numbers just to increase their caste strength is baseless because the census collects full details (including name, sex, education, marital status, occupation, etc) of every member returned and non-existent persons cannot be simply added at will. Further, the legal safeguards as well as verification checks built into the system will prevent the collection of false data. Also, in its experience of collecting and publishing caste data from 1871 till 1931, and the data on the Scheduled Castes (SC) and Scheduled Tribes (ST) from 1951 till 2001, the census has not reported any instance of the numbers being inflated or distorted.

The way out

The collection of caste data can be easily done and without any extra expenditure through the schedule used in the house to house census enumeration in February 2011 by just rewording the column for SC-ST as ‘Caste.' This will straightaway ensure correlation with all the other data and make for quick and simultaneous processing and publication of caste data along with the other 2011 Census tables. The advance enumeration scheduled in the states of Jammu & Kashmir, Himachal Pradesh and Uttarakhand in October 2010 can be postponed to the summer of 2011, giving enough time from now till next February to finalise, print and distribute the census schedules and manuals.

The arguments against conducting such a gigantic exercise twice within a few months are so many and so compelling that one wonders how they can be ignored. A separate census for caste in June 2011 cannot be defended in administrative, logistical, or financial terms. If the intended policy benefits of the proposed caste data collection are to be achieved, a decision to effect a course correction is required urgently.


Dr. M. Vijayanunni, Former Census Commissioner & Registrar General of India; Prof. Satish Deshpande, Department of Sociology, Delhi School of Economics; Prof. Yogendra Yadav, Political Scientist, Centre for the Study of Developing Societies; Prof. Sukhadeo Thorat, Chairman, University Grants Commission; Prof. S. Japhet, Director, Centre for the Study of Social Exclusion, National Law School of India, Bengaluru; Dr. Chandan Gowda, Associate Professor, Centre for the Study of Social Exclusion, National Law School of India, Bengaluru; Prof. Ravi Varma Kumar, Senior Advocate and former Chairperson, Karnataka Backward Classes Commission.

More In: Comment | Opinion