Note on meeting over caste census rewritten after Ministry pushback

Why in the News

The Office of the Registrar General of India has rewritten its record of a meeting on caste enumeration after the Ministry of Social Justice and Empowerment objected to how its position had been minuted. The original note recorded the Ministry as stating that it holds no caste list at all, and that enumeration using a pre defined list in Census 2027 may therefore not be possible. The Ministry read that wording as shifting onto it the responsibility for dropping a pre defined list, and returned an amended version, which was accepted. Census 2027 will ask caste as an open ended question rather than offer a drop down menu of names. The tension is between a state unwilling to become the arbiter of who belongs to which caste and a count that must still produce standardised numbers on which social justice policy can be built.

What is the open ended caste question, and how does it differ from a drop down?

1. An open ended question records what the respondent says: The enumerator writes down the caste name as reported, without matching it to any approved list at the point of collection.
2. A drop down constrains the answer to a fixed list: The Census already uses drop down lists for the notified Scheduled Castes and Scheduled Tribes, where a legally settled list exists.
3. The stated reason for the choice: The government’s position is that an open ended response is the only way to run the exercise without the state itself becoming an arbiter of caste identities.
4. The cost is deferred, not avoided: The work of converting millions of individual responses into reliable, standardised caste data moves from the field to the stage after enumeration.

How was the record of the June meeting rewritten?

1. The Ministry offered the OBC list at the meeting: Ministry officials expressed willingness to share the list of Other Backward Classes (OBCs) maintained by the National Commission for Backward Classes (NCBC) for the caste enumeration exercise.
2. The first draft minutes did not carry the offer: The initial draft of the minutes prepared by the Registrar General’s office did not reflect that offer accurately.
3. The circulated note recorded a flat denial instead: A four paragraph note circulated on 8 June stated that Ministry officials had informed the meeting that there is no constitutional mandate for the Ministry to maintain a caste list, and that no such list is available with it.
4. The note then drew the operative conclusion: It recorded that in view of the absence of any list of castes, enumeration of castes in Census 2027 may not be possible using any pre defined list in the Census question.
5. The Ministry read the wording as a trap: Ministry officials said the onus for not going forward with a pre defined list was being shifted onto them.
6. The amended note was accepted: The Ministry sent a revised version on 11 June, and the Registrar General’s office accepted the revised notes on 30 June.

What lists does the Ministry actually maintain?

1. Two statutory lists, not a general caste list: The revised note recorded that the Ministry maintains the List of Scheduled Castes and the Central List of Other Backward Classes as mandated by the Constitution, and that no separate list of castes is required to be maintained.
2. The Scheduled Caste list: It comprises 1,258 entries and had already been shared with the Registrar General’s office by a letter dated 25 September 2025.
3. The Central OBC list: It comprises nearly 2,483 entries, and the Ministry said it would be provided to the Registrar General’s office if required.
4. Where the assistance stopped: The note recorded that officials of the Department of Social Justice and Empowerment were unable to assist further, given the absence of any mandate to maintain a list of castes beyond those two.

Why does the 2011 experience shape this decision?

1. The last attempt produced unusable variety: Respondents in the 2011 Socio Economic and Caste Census (SECC) reported 46.7 lakh distinct caste names, against 4,147 recorded in 1931.
2. The cost: That exercise cost about ₹4,900 crore.
3. The caste data was never released: The SECC data was published by the Ministries of Rural Development and Urban Development in 2016, and the dataset on caste was excluded.
4. Classification was outsourced and never completed: The raw caste data went to the Ministry of Social Justice and Empowerment, which formed a group of experts under the then Vice Chairperson of NITI Aayog for classification and categorisation, and that data is yet to be made public.
5. The government told the Supreme Court the data was unusable: Affidavits filed on 21 September and 14 December 2021 said the SECC 2011 data was inaccurate and marked by technical flaws.
6. The Court recorded that position: On 15 December 2021 the Supreme Court noted that the affidavit before it states that the data as collated “is not accurate and is unusable for any purpose whatsoever”.
7. The same failure is expected to recur: The assessment recorded inside the Ministry is that the 2027 count is set to repeat the failure of the previous exercise.

How did the 2027 caste count come about?

1. Cabinet approval reversed a standing position: The Cabinet Committee on Political Affairs approved caste enumeration in Census 2027 on 30 April 2025, conceding a long pending demand and reversing the government’s earlier position.
2. The demand was first conceded in 2010: The then government promised caste enumeration alongside Census 2011 in May 2010 and referred the question to a Group of Ministers.
3. It was then diverted out of the Census: Acting on that Group’s recommendations, the Union Cabinet decided in September 2010 to conduct a separate SECC instead of counting caste in the Census itself.
4. The field timetable is already running: Population enumeration together with the caste census is scheduled across the country in February 2027, and it began on 1 September 2026 in hilly and snow bound areas to avoid winter difficulties.
5. The Opposition objection is about usability, not about counting: The Leaders of the Opposition in the Lok Sabha and the Rajya Sabha argued that social justice policies cannot be effectively implemented without accurate data.

Challenges to caste enumeration in Census 2027

1. Open responses multiply into synonyms, sub castes and surnames: One community reports itself under several names across districts, and no field rule decides which of those names denote the same group. Eg. The Bihar caste survey of 2023 enumerated against a pre defined list of castes rather than an open field, precisely to avoid that outcome.
   The Fix: Publish the standardisation methodology and the mapping rules before enumeration begins, so the conversion is auditable rather than discretionary.
2. The classification stage carries no statutory deadline: Raw responses become policy relevant only after grouping, and nothing fixes when that grouping must be completed or released. Eg. The Census Act, 1948 and its rules set the enumeration schedule and say nothing about publishing a caste classification.
   The Fix: Fix a statutory deadline for publishing classified caste tables, as already exists for the primary Census abstracts.
3. Self reported caste can be reported strategically: Where a count is known to feed benefit eligibility, a respondent has an incentive to report the category that maximises entitlement. Eg. Reservation litigation regularly turns on disputed community claims to Other Backward Class or Scheduled Tribe status.
   The Fix: Delink the enumeration record from individual entitlement, using it only for aggregate policy design.
4. Numbers alone do not satisfy the constitutional test: Population share does not establish that a group is socially and educationally backward, which is what the Constitution requires before reservation follows. Eg. Indra Sawhney v. Union of India (1992) fixed a ceiling on reservation and required backwardness to be demonstrated rather than assumed.
   The Fix: Collect the socio educational indicators the constitutional test needs in the same schedule as the caste question.
5. Two arms of government own different halves of the exercise: The Registrar General’s office runs the enumeration and the Ministry of Social Justice and Empowerment holds the statutory lists, and neither is answerable for the output the other needs. Eg. The rewriting of the meeting note turned entirely on which of the two would be recorded as having ruled out a pre defined list.
   The Fix: Name a single accountable authority for the caste module, with the other body’s inputs recorded as statutory obligations rather than as offers.

Conclusion

The design of the caste question has been settled by default rather than by decision. Neither arm of government was willing to own the list a closed format needs, and the open format was what remained. The methodological problem that broke the last attempt is therefore untouched: responses will be collected freely and must still be grouped afterwards by someone. What to watch is whether the standardisation rules are published before enumerators go into the field, or arrive only after the data is already in.

Back2Basics

1. What it is: The National Commission for Backward Classes examines requests for inclusion in and exclusion from the Central List of Other Backward Classes and advises the Central government on them.
2. Constitutional status: The Constitution (One Hundred and Second Amendment) Act, 2018 inserted Article 338B and gave the Commission constitutional status, replacing its earlier statutory basis.
3. Composition: It consists of a Chairperson, a Vice Chairperson and three other members appointed by the President.
4. States retain their own lists: The Constitution (One Hundred and Fifth Amendment) Act, 2021 restored the power of States and Union Territories to prepare and maintain their own lists of socially and educationally backward classes.

Matching Previous Year Question

“No direct PYQ traced in the provided files. Closest microtheme: Census.”