The case for a caste Census

The Census Act, 1948 should be amended to make enumeration by caste mandatory, instead of leaving it to the whims of the Union executive. Such data can be collected as part of the regular Census, with a few pertinent queries added to the questionnaire

Published - July 10, 2024 11:01 pm IST

JD(U) workers and supporters celebrate after the declaration of Bihar’s caste-based census report, in Patna, on October 4, 2023.

JD(U) workers and supporters celebrate after the declaration of Bihar’s caste-based census report, in Patna, on October 4, 2023. | Photo Credit: PTI

Peter Drucker famously said, “Only what gets measured gets managed.” The problems of social groups that have been historically discriminated against (be it by caste, race, religion, gender, disability etc.) cannot be resolved without collecting data group-identity wise. Doing so is not a capitulation to identity politics but a vital step towards informed policy making and inclusive development.

For instance, Germany’s census does not enumerate people by race. This has worked to the disadvantage of its Black people who started a private, country-wide, online survey called Afrozensus in 2020. Its results showed that anti-Black racism is widespread and institutionally entrenched in Germany.

Applying Cicero’s test of cui bono (who benefits?), it can be said that the demand for enumeration generally emanates from the victims of discrimination and is resisted by vested interests.

Why a caste Census?

A caste Census is crucial for four reasons — one, it is a social imperative. Caste continues to be a foundational social construct in India. Only about 5% of Indian marriages were inter-caste as of 2011-12. The use of caste surnames and caste marks is still widespread. Residential segregation by caste persists. Choices of candidates for elections and ministers for Cabinets continue to be dictated by caste considerations.

Two, it is a legal imperative. Constitutionally-mandated policies of social justice which include reservations in electoral constituencies, education and public employment cannot be pursued effectively without detailed caste-wise data. Even though the Constitution uses the word class instead of caste, various rulings of the Supreme Court have held caste as a ‘relevant criterion’, ‘sole criterion’ or ‘dominant criterion’ for defining a backward class, and have demanded detailed caste-wise data for upholding reservation policies.

Three, it is an administrative imperative. Detailed caste-wise data is necessary to avoid/correct wrongful inclusions of undeserving castes and exclusions of deserving castes, and to guard against a few dominant castes in a reserved category crowding out others. It is also needed for sub-categorising castes within a reserved category and to determine the income/wealth criterion for the creamy layer.

Four, it is a moral imperative. The absence of detailed caste-wise data has helped a coterie of elites, among upper castes and dominant Other Backward Classes (OBCs), to corner a disproportionate share of the nation’s assets, incomes, and positions of power.

Censuses in British India between 1881 and 1931 enumerated all castes. In the first Census conducted after Independence in 1951, the Government of India (GOI) ordered that caste should not be enumerated. However, an exception was made for Scheduled Castes (SCs) and Scheduled Tribes (STs) which have been enumerated in every Census since 1951. In 1961, the GOI advised States to conduct their own surveys and draw up State-specific OBC lists if they so desired. There was no reservation for OBCs in the Centre and its undertakings then.

The arguments against caste Census

There are multiple views against the caste Census. These include:-

First, that it is socially divisive. India’s social divisions predate Census efforts by nearly 3,000 years. The Census counts of the SCs and STs since 1951 have not led to any conflicts among these castes or tribes. Further, India’s Census enumerates religion, language, and region which are as divisive as caste, if not more. Casteism will not wither away by not counting caste in the Census, any more than communalism, and regionalism will disappear by not enumerating religion, language and region.

Second, that it is an administrative nightmare. Unlike race which is a fuzzy concept, but is still enumerated in many countries such as the U.S., there is little or no ambiguity about anyone’s caste. The GOI has been able to smoothly enumerate 1,234 castes in the SC category and 698 tribes in the ST category. Therefore, it is difficult to understand why the enumeration of the 4,000-odd other castes, most of which are State-specific, should pose an intractable problem.

Third, that it would fuel demands for increased reservations. On the contrary, the availability of caste-wise Census data would help curb arbitrary demands from caste groups and capricious decision-making by governments. Policy makers would be able to objectively debate and address the claims of, say, the Marathas, Patidars, Jats, or any other groups for reservations. But governments prefer fuzzy data because it gives them the latitude to implement reservations arbitrarily for electoral considerations.

The case for OBC inclusion in Census

Like the SCs and STs, the Constitution permits reservation for the OBCs in education (Article 15(4)) and public employment (Article 16(4)). After the implementation of the Mandal Commission recommendations, the OBCs enjoy reservations in the Central government and its undertakings as well. In the Indra Sawhney case (1992), the Supreme Court ruled that the OBC list, based on the 1931 Census, should be revised periodically.

The OBCs do not have reservation in electoral constituencies for MPs and MLAs like the SCs and the STs. But after the 73rd and 74th amendments (1993), the Constitution provides for reservations in electoral constituencies in panchayats and municipalities not only for SCs and STs but also for OBCs (Articles 243D(6) and 243T(6)). For this, caste-wise, area-wise Census data of the OBCs is essential. Therefore, the GOI should have enumerated the OBCs at least in the 2001 Census. But it did not.

Whenever States like Uttar Pradesh, Madhya Pradesh, Gujarat, Maharashtra, Karnataka, Odisha and Jharkhand tried to implement reservations for the OBCs in elections to local bodies, the High Courts and the Supreme Court stayed the same, on the ground that there was no caste-wise data of OBCs. While one arm of the State — the judiciary —demands caste-wise data to uphold reservations, another — the executive — has avoided enumerating the very same data.

However, the 10% reservation for economically weaker sections (EWS) among those other than the OBCs, SCs and STs (effectively, the upper castes) was upheld by the Supreme Court in 2022 despite the absence of any supporting empirical data. In view of the EWS reservation, the Census should now enumerate all castes including the upper castes as it did till 1931.

While Census is a Union subject, the Collection of Statistics Act, 2008 empowers States and even local bodies to gather the necessary statistics. So, individual States can always do Caste surveys like Karnataka (2015) and Bihar (2023) did. But Census data carries more authority and is less contested. The government’s reluctance to enumerate caste as part of the Census is legally indefensible and administratively unwise.

How an attempt at caste Census failed

After considerable lobbying by OBC leaders, in 2010, the Parliament passed a unanimous resolution (with both Congress and BJP on board) calling for caste to be enumerated as part of the 2011 Census. As per the 1931 Census when caste was last enumerated, there were 4,147 castes in India apart from the depressed classes/untouchables (as they were called then). Unfortunately, the Socio Economic and Caste Census (SECC)-2011 was poorly designed and executed, throwing up a ludicrous figure of 46 lakh castes and the results were never released.

The SECC-2011’s failure was because of the fact that it was not conducted under the Census Act, 1948 as the Act was not amended to include caste as a parameter. It was conducted through the Union Ministries of Rural Development and Urban Development which did not have prior experience of conducting sociological/anthropological surveys. Additionally, the questionnaire was poorly designed and asked open-ended questions about caste. The enumerators couldn’t distinguish between genuine castes, alternative caste names, larger caste groups, sub-castes, surnames, clan names, gotras, etc. In contrast, the Bihar government’s Caste Survey in 2023, provided enumerators the list of 214 caste names specific to Bihar, with the 215th category labelled “Other Castes” and came up with better results.

Despite the unanimous Parliamentary resolution of 2010, the Central government announced in 2021 that it would not enumerate caste as part of the next Census. It reiterated this stand before the Supreme Court in a case filed by the Maharashtra government seeking a direction to the Centre to enumerate OBCs in the 2021 Census. The Supreme Court judgment dismissing the Maharashtra government’s plea in December 2021 is questionable, considering its own past rulings. Being a legal imperative, this is not ‘a policy matter in which Courts should not interfere’ as claimed by the Centre in that case.

What is the way forward?

Learning from the SECC-2011 failures, the Census Act, 1948 should be amended to make enumeration by caste mandatory instead of leaving it to the whims of the Union executive. Caste should be enumerated as part of the regular Census by the Census Commissioner only, with a few pertinent questions added to the questionnaire. Additionally, the government should enlist sociological/anthropological experts to draw up a draft list of castes specific to each State, publish the draft list online, inviting suggestions and comments from the public before finalising it, and give only that list to the enumerators. The questionnaire should be so designed as to ask the names of the sub-caste, caste, larger caste group, and the caste surname of the respondent. Internet-enabled hand-held devices preloaded with these details and limiting the enumerator’s role to one of choosing the correct option will make the task easy and foolproof.

Interested States must move the apex Court to review its 2021 judgment. It is farcical to implement OBC reservation based on 1931 Census data and EWS reservation with no empirical data. The next Census must enumerate caste.

The writer is former IAS officer of Tamil Nadu cadre and former Vice Chancellor, Indian Maritime University, Chennai.

0 / 0
Sign in to unlock member-only benefits!
  • Access 10 free stories every month
  • Save stories to read later
  • Access to comment on every story
  • Sign-up/manage your newsletter subscriptions with a single click
  • Get notified by email for early access to discounts & offers on our products
Sign in


Comments have to be in English, and in full sentences. They cannot be abusive or personal. Please abide by our community guidelines for posting your comments.

We have migrated to a new commenting platform. If you are already a registered user of The Hindu and logged in, you may continue to engage with our articles. If you do not have an account please register and login to post comments. Users can access their older comments by logging into their accounts on Vuukle.