Registrar-General reviews Census exercise in Ladakh

1. At a Glance

• Census 2027 is India's first digital census and the first since 2011. It runs in two phases: Houselisting & Housing Census (HLO) and Population Enumeration (PE). [2][3]
• Ladakh is the first UT or State to undergo the PE phase, ahead of the rest of the country, because of snow-bound access constraints. [1][4]
• It also revives caste enumeration, which was last done in a census in 1931. [5]
• Aspirants should know the Census machinery, the Census Act 1948, the reference dates and the administrative design (RG&CCI, CPCO, Charge Officers). [1][2]

2. Why in the News

• RG&CCI Mritunjay Kumar Narayan made a four-day visit to Ladakh to review the PE phase. [1]
• He focused on coverage in remote and snow-bound areas, the Census mobile app, and preparedness of field staff. [1]
• He was accompanied by Amit Sharma, Chief Principal Census Officer (CPCO) and Director of Census Operations for J&K and Ladakh. [1]

• Field inspections covered Nubra, Chumathang and Chushul (Nyoma). [1]

• The PE phase in Ladakh is due to conclude by 30 September 2026. [1]

• Print edition of 26 Sep 2026, page 5. [1]

3. Background & Evolution

• Census Act, 1948 is the statutory basis. The intent to conduct Census 2027 was notified in the Gazette on 16 June 2025. The notification was signed by RG Mritunjay Kumar Narayan. [2]
• Phase I (HLO) was scheduled for 1 April to 30 September 2026, with a 15-day self-enumeration window before house-to-house work. [2][6]
• 14 Aug 2026 (per the report's date code): the Centre notified the 40 PE questions. [5]
• Ladakh's PE began with self-enumeration in Aug 2026. Snow-bound areas of J&K, Himachal Pradesh and Uttarakhand were also included. [4]
• The PE phase for the rest of the country is scheduled for February 2027. [2]
• The previous census was in 2011, with a 29-question schedule. [1]

4. Core Static Facts

Item	Fact
Nodal authority	Registrar General & Census Commissioner of India (RG&CCI) [1]
Statute	Census Act, 1948 (data strictly confidential) [2]
PE questions (2027)	40, up from 29 in 2011; 13 new fields [1]
Time per household	30–60 minutes on average, using the app [1]
Reference date, Ladakh and snow-bound areas	00:00 hrs, 1 Oct 2026 [2][4]
Reference date, rest of India	00:00 hrs, 1 March 2027 [4]
Snow-bound areas covered	Ladakh, plus non-synchronous areas of J&K, Himachal Pradesh and Uttarakhand [2][4]
Field hierarchy	CPCO, Charge Officers, enumerators [1]

5. Multi-Dimensional Analysis

Administrative

• Asynchronous enumeration avoids winter inaccessibility. [4]
• Review focus areas were app functioning and staff readiness, which are the main bottlenecks in a first digital census. [1]
• Special arrangements cover military and paramilitary locations. [4]

Scientific / Technological

• The census is digital, with app-based data capture and self-enumeration. [3][6]
• Connectivity in remote terrain is the main risk. That is why the RG reviewed the app in the field. [1]

Social

• Caste enumeration returns after 1931. [5]
• Remote and tribal communities in Nubra and Nyoma risk undercounting without dedicated reach. [1]

Legal / Governance

• The Census Act 1948 guarantees confidentiality of data. [2]
• Central operations run through State and UT administrations, which reflects the federal-administrative split. [1]

6. Recent Developments (last 12-18 months)

• 16 Jun 2025: Gazette notification of intent to conduct the census. [2]
• Apr 2026: Phase I houselisting began, with a digital, self-enumeration option. [6]
• Aug 2026: the 40 PE questions were notified. [5]
• Aug 2026: the PE phase began in J&K and Ladakh. [4]
• 1 Sep 2026: second-phase inauguration across snow-bound areas in Udhampur. [1]
• Sep 2026: RG's four-day Ladakh review. The PE phase is to conclude by 30 Sep. [1]

7. Prelims Hooks

• Census 2027 is conducted under the Census Act, 1948. [2]
• Intent notified in the Gazette on 16 June 2025. [2]
• Ladakh is the first State/UT to undergo the PE phase. [1]
• The PE questionnaire has 40 questions, versus 29 in 2011. [1]
• 13 new data fields were added. [1]
• Reference date for Ladakh and snow-bound areas: 1 Oct 2026. [4]
• Reference date for the rest of India: 1 March 2027. [4]
• The census is the first digital one and includes self-enumeration. [3][6]
• The RG&CCI is Mritunjay Kumar Narayan. [1]
• Caste is enumerated in a census for the first time since 1931. [5]
• CPCO for J&K and Ladakh: Amit Sharma, who is also Director of Census Operations. [1]
• Ladakh sites inspected: Nubra, Chumathang, Chushul. [1]

8. The Cost of Counting Late: Ration Cards Still Fixed on 2011 Numbers

• The note treats 2011 as just "the last census". For welfare schemes, it is a live problem.
• Under the National Food Security Act, 2013 (NFSA) — the law that gives cheap grain to the poor — how many people each State can cover is worked out from census population [7].
• That calculation still uses the Census 2011 population of about 121 crore, because no newer census count exists [7].

• India's population has grown a lot since. So the number of ration slots stayed the same while the population rose.

• Who this leaves out

• One report counted over 9 crore people left outside PDS entitlement, purely because coverage is pegged to 2011 [14].
• The Supreme Court in 2021 told the Centre to re-work beneficiary numbers and not stay tied to Census 2011 [7].

• In August 2026 the Centre told States they could add about 2.27 crore more beneficiaries — an interim patch, not new census data [17].

• Exam point: a census is not only a statistics exercise. It is the denominator (the bottom number) for welfare coverage, grants and seat shares. A delayed census silently shrinks entitlements without any law being changed.

9. Why Caste Is Easy to Ask and Very Hard to Tabulate

• The note says caste returns after 1931. It does not say why the last attempt never produced usable tables.
• The SECC 2011 (Socio Economic and Caste Census) did collect caste. Its caste data was never published [8].
• The reason was classification, not politics alone: the field was open-ended, so people wrote whatever name they used [8].

• The result was about 4.7 million (47 lakh) distinct caste entries — caste names, sub-castes, synonyms, surnames, clan and gotra names, and spelling variants of the same caste [8].

• The error pile that followed

• States were asked to correct 8.19 crore flagged errors; after they fixed 6.74 crore, about 1.45 crore errors still remained [8].

• Once the same caste appears under 20 different spellings, you cannot add up its population. That is why the tables never came out [8].

• Why Census 2027 could repeat it

• The census app records caste as one of the 40 questions [1][5]. Unless the app offers a pre-loaded, State-wise list of caste names for the enumerator to pick from, the enumerator is again typing free text.

• A tablet does not solve a naming problem. It only makes a messy list faster to build.

• What good caste data can do: Bihar's 2022 survey found EBCs at about 36% and OBCs at about 27% of the State's population [16]. Numbers like these immediately change the argument over quota sizes.

10. Why the Ladakh Count Carries More Weight Than Its Population

• Ladakh is a tiny population but a high-stakes count, because almost every political demand there rests on tribal numbers.
• Over 97% of Ladakh's people are Scheduled Tribes [11].
• The National Commission for Scheduled Tribes (NCST) has recommended including Ladakh in the Sixth Schedule — the part of the Constitution that gives tribal areas their own autonomous councils with powers over land and local law [11].

• Sixth Schedule status, council seats and quota shares are all argued using tribal population figures. A fresh, official count becomes the evidence base for that argument.

• The concessions already made through talks

• Through the High Powered Committee dialogue with Ladakh leaders, reservation for Ladakh's Scheduled Tribes was raised from 45% to 84%, one-third of seats in the Ladakh Autonomous Hill Development Councils (LAHDC) were reserved for women, and Bhoti and Purgi were made official languages [10].

• Statehood and the Sixth Schedule remain unsettled [10].

• So an undercount in Nubra or Nyoma is not just a statistical error. In a UT where 97% are ST, the count feeds directly into how many jobs, council seats and protections a community can claim [10][11]. That is the real reason a Registrar General personally inspects three remote blocks [1].

11. The 40 Questions Ask for Your ID Numbers, and That Changes the Risk

• The note counts the questions. It does not say what the new ones are.
• Beyond caste, the 2027 schedule asks for Aadhaar, voter ID, mobile number, passport, driving licence, Covid-19 vaccination details, parents' details and permanent address [9].

• Adding these has raised privacy concerns, because a census record now contains the keys that link to many other government databases [9].

• The law protects against one risk, not the other

• The Census Act, 1948 makes individual answers strictly confidential: they are outside the RTI Act, cannot be used as evidence in court, and cannot be shared with any institution [3].
• That protects against disclosure. It says nothing about linking — using the same identifier to match a census record with other official records inside government.

• Exam framing: compare this with the DPDP Act, 2023, which builds consent and purpose-limitation duties. A statutory secrecy clause from 1948 was not written for a world of linked digital identifiers.

• Self-enumeration adds a quality question too

• Self-enumeration means the household fills its own form online, before any enumerator visits [3].
• No field officer sees the household at that moment, so entries about caste, age or household size are not observed by anyone. The enumerator's later visit becomes the only check.

12. The Strongest Argument Against a Caste Count, and What It Misses

• The case against (state it honestly):
• A newspaper editorial argued that a caste census will not address the basic issues of deprivation — schooling, jobs, health — and that counting caste can harden caste identity instead of weakening it [12].

• It also sets off a race for bigger quotas. After its own survey, Bihar passed a law in 2023 raising reservation in educational institutions [13]. Every published caste share becomes a claim for a larger share of seats.

• Where that argument is right: caste numbers alone tell you group size, not group deprivation. Size is not the same as backwardness. Without income, education and occupation data read alongside it, a bigger count can justify a bigger quota for a group that is not the worst off.

• Where it falls short:
• India currently fixes OBC quotas without knowing the actual OBC population — the last full caste count was 1931 [5]. You cannot test whether a quota is too large or too small with no denominator.
• The alternative that grew in the gap was State-run surveys of uneven quality and intent, which is exactly why caste was folded into the national census instead [2].

• The census carries a statutory confidentiality guarantee that an ad-hoc State survey does not [3].

• Balanced line for a Mains answer: the caste count is necessary but not sufficient. Its value depends on whether the caste tables are published together with the socio-economic columns, so size can be read against deprivation.

13. What Must Be Settled Before the February 2027 Round

• RG&CCI should load a standard caste directory into the app, not a blank text box
• SECC 2011 failed because 47 lakh free-text caste entries could not be added up [8].

• A State-wise picklist, with local synonyms mapped to one standard name, is the one fix that addresses the exact reason the last attempt produced no tables.

• The Ministry of Home Affairs should announce, in advance, a date for publishing the caste tables

• SECC's caste data was collected and then simply never released [8]. Silence after collection is the historical pattern, so a pre-committed release date is the safeguard.

• The Department of Food and Public Distribution should re-fix NFSA coverage as soon as 2027 figures arrive

• Coverage is still built on the 2011 population [7], and the Supreme Court's 2021 direction to stop using Census 2011 is still the standing instruction [7].

• Without this, a new census will be published while ration entitlements stay frozen on old numbers.

• States must finish district and boundary changes before the freeze bites

• All administrative units are frozen from 1 January 2026 to 31 March 2027 [15].
• Any district or tehsil created after the freeze will not appear as a separate unit in Census 2027 tables, so that area's planners work without its own data for years.

14. Anchors for Answers

• Data: Over 9 crore people excluded from PDS entitlement because coverage is still pegged to Census 2011 [14]
• Data: 4.7 million (47 lakh) distinct caste entries recorded in SECC 2011; 8.19 crore errors flagged, 1.45 crore still uncorrected [8]
• Data: Over 97% of Ladakh's population is Scheduled Tribe [11]
• Data: 2027 PE schedule has 40 questions including Aadhaar, voter ID, mobile, passport and driving licence [1][9]
• Report/Committee: NCST recommendation to include Ladakh in the Sixth Schedule [11]; High Powered Committee on Ladakh, which raised ST reservation from 45% to 84% [10]
• Law/Case: Census Act, 1948 — statutory confidentiality, data outside RTI and inadmissible in court [3]; Supreme Court 2021 direction to re-determine NFSA beneficiaries beyond Census 2011 [7]; Sixth Schedule of the Constitution [11]; DPDP Act, 2023 for digital data duties
• Comparison: SECC 2011 — same caste question, collected but never published because of classification failure; the direct precedent Census 2027 must avoid [8]
• Comparison: Bihar caste survey 2022 — EBC about 36%, OBC about 27% of State population [16], followed by a 2023 State law expanding reservation in education [13]
• Scheme: National Food Security Act, 2013 — coverage still calculated on the 2011 population of about 121 crore [7]; interim addition of about 2.27 crore beneficiaries allowed in August 2026 [17]

15. Mains Relevance

• GS-I: Population and associated issues. GS-II: Government policies and interventions; Statutory bodies; Federal structure. GS-III: Technology in governance.
• Likely questions:
• Asynchronous census enumeration in snow-bound areas raises both administrative and data-consistency issues. Discuss.
• Evaluate the significance of the first digital census for policy design and data governance.
• Examine the implications of reintroducing caste enumeration in the census.

16. Related Topics to Study Next

• Census Act 1948 and Article 246 (Union List entry 69): the legal basis for the census.
• SECC 2011 versus caste census: comparison of past caste data efforts.
• Delimitation: it is linked to census population data.
• NPR and NRC: related population registers.
• Women's Reservation Act: implementation depends on the census and delimitation.
• Data protection (DPDP Act 2023): confidentiality and digital data handling.
• Ladakh's governance (Sixth Schedule demands, UT status): context for the region.
• Digital India and connectivity in border areas.

17. Common Errors / Trap Areas

• Confusing HLO (Phase I) with PE (Phase II). Ladakh is only ahead in Phase II. [1][2]
• The reference date is 1 Oct 2026 for Ladakh, not 1 March 2027. [4]
• 40 questions is the 2027 PE count. The 2011 schedule had 29. [1]
• The RG&CCI works under the Ministry of Home Affairs. This is not stated in the sources; verify it.
• The caste count is the first since 1931. The SECC 2011 was not a census under the Census Act. [5]