Caste census: DMK chief Stalin urges Centre to avoid discrepancies, demands Tamil Nadu Assembly resolution
Concerns have been raised over the Centre's proposed open-ended format for recording caste details in the upcoming Census 2027, with warnings that it could produce inconsistent or unusable data
A state government has been urged to pass a special Assembly resolution pressing the Centre to instead use a standardised caste list or drop-down system, based on existing state Backward Classes (BC) and Most Backward Classes (MBC) lists
The Registrar General of India's pre-test, conducted in July 2026 across 16 states and Union Territories, used the open-column (write-in) method for caste, and this method has been retained for the final census schedule
The debate centres on data quality and comparability of caste enumeration data collected nationwide
Constitutional and Legal Basis of the Census
The population census is a Union subject under Entry 69 of the Union List in the Seventh Schedule, read with Article 246 of the Constitution, giving Parliament exclusive legislative competence over it. The Census Act, 1948 is the statutory basis for conducting census operations in India, including appointment of census officers, mandatory disclosure obligations, and confidentiality of individual data.
Because caste enumeration will now be part of the main constitutionally-mandated census (not a separate SECC-type exercise), the format chosen for recording caste has heightened legal and administrative significance — hence the dispute over open-ended versus standardised recording.
Open-Ended Format vs Drop-Down/Standardised List — Lessons from SECC 2011
An "open-ended" or open-column format records the caste name exactly as stated by the respondent, without enumerators categorising it against a pre-set list. A standardised or "drop-down" format requires enumerators to match responses against an official, pre-approved list of castes (e.g., state OBC/MBC lists). The 2011 Socio-Economic and Caste Census (SECC) used an open-ended approach and generated an unmanageably large and inconsistent dataset.
Key Details
- SECC 2011 (not conducted under the Census Act, 1948) recorded approximately 46 lakh (4.6 million) distinct caste names nationally, most attributable to surname/sub-caste variations rather than genuine caste categories
- By contrast, the last full caste census, in 1931, recorded only 4,147 castes, using more structured classification
- The Registrar General of India has opted for the open-column method again for Census 2027, based on a July 2026 pre-test across 16 states/UTs
- States favouring standardised drop-down lists argue this would ensure consistency with existing state-level OBC/MBC classifications used for reservation implementation
The core of the current dispute is precisely this open-ended-versus-standardised choice — a repeat of the SECC 2011 data-quality problem is the specific risk being flagged, since inconsistent caste data would complicate its intended use for backward-classes policy.
Caste Census, the Mandal Commission, and the Reservation Framework
Historically, caste-based reservation policy in India has relied on estimates rather than direct enumeration. The Mandal Commission (1980) estimated the Other Backward Classes (OBC) population at around 52% by extrapolating from the 1931 census and other survey data, rather than from a direct caste count, since no caste census had been conducted after 1931. A national caste census is expected to provide an evidence base for backward-classes policy and could inform "quota-within-quota" sub-categorisation debates.
Key Details
- Mandal Commission (1980) recommended 27% reservation for OBCs, based on extrapolated (not enumerated) population estimates
- The Supreme Court, in Indra Sawhney v. Union of India (1992, a nine-judge bench), upheld OBC reservation and laid down the general 50% ceiling on total reservations, along with the concept of a "creamy layer" exclusion within OBCs
- Caste census data is seen as providing an empirical basis for future decisions on reservation ceilings, sub-categorisation of OBCs, and targeting of welfare schemes
- Some states have already conducted their own caste surveys independent of the central Census, creating precedent for state-level standardised classification systems
The demand for the Centre to align caste enumeration with state BC/MBC lists reflects concern that non-standardised central data could conflict with, or fail to reinforce, existing state-level backward-classes classifications used for reservation implementation.
- Census is a Union subject: Article 246, Seventh Schedule, Union List, Entry 69
- Legal basis: Census Act, 1948
- SECC 2011 recorded approximately 46 lakh distinct caste names (not conducted under the Census Act)
- 1931 Census (last full caste count) recorded 4,147 castes
- Mandal Commission (1980) recommended 27% OBC reservation based on extrapolated ~52% OBC population estimate
- Indra Sawhney v. Union of India (1992): nine-judge bench, upheld OBC reservation, laid down 50% ceiling on total reservations and "creamy layer" exclusion
- Census 2027 caste enumeration pre-test conducted July 2026 across 16 states/UTs using the open-column method