34 | The Last Caste Census of India and Its Long Shadow

By Nishitha Mandava, Gaurav Kalyani and Shivakumar Jolad

Published on: 31 August 2026

The year is 1930. M. K. Gandhi, by then already bearing the mantle of the Mahatma, led the Civil Disobedience Movement, inaugurated by his Salt March. At the same time, preparations for the 1931 census had begun. But across several cities, most notably in Gujarat, such as Ahmedabad, Broach, and Surat, the Indian National Congress (INC) declared a partial boycott of the census.

In Ahmedabad, Gandhian non-cooperation took the form of householders locking the gates of their pols (residential lanes of the city), shutting census enumerators out. This forced the officials to make mathematical estimations for approximately 75,000 to 150,000 uncounted urban dwellers. However, beyond this, John Henry Hutton, the Census Commissioner, writes that there was little impact of the boycott on the census operations. 

Simultaneously, communal tensions flared. As the Roundtable talks on constitutional reforms were underway, the proposed expansion of electoral franchises based on communal quotas generated a new consciousness among caste and religious associations to maximize their numerical representation. Punjab witnessed the worst of these tensions taking the form of intense competition between Hindu, Sikh, and Muslim organisations that claimed lower-caste adherents. This led to physical affrays and homicides. 

Within this charged atmosphere, Hutton was tasked with compiling data for the caste census, unaware that this would be the last time a comprehensive caste enumeration would be undertaken. The costs and logistics of the Second World War led the 1941 census operations to abandon caste enumerations, and postcolonial India would go on to record only the castes of those considered Scheduled Tribes and Scheduled Castes. 

The 1931 data would be used for estimating categories such as Other Backward Classes (OBCs) and persists as the only source for estimating them. Nearly a century later, the caste data generated by the 1931 census continues to serve as a reference point for caste-specific welfare policies, demographic estimates, and debates over social representation in India. 

In this essay, we explore the key findings of India’s last caste census and attempt to make a dent in tracing its legacies and afterlives. These insights take on renewed significance given that the 2027 census of India will, for the first time in postcolonial India, undertake a caste census.

The Scope 

The caste enumeration exercise in 1931, conducted against the background of the Great Depression, recorded 4,100 distinct castes. More than double the 1,646 recorded in 1901 (Bhagat, 2006). This reflected how caste categories were multiplying rather than settling under the pressure of enumeration, education, and political mobilisation. It was in stark contrast to what even the census officers had hoped that rigid caste distinctions were breaking down (Hutton, 1933).

Owing to the financial stress caused by the Great Depression, the 1931 census abandoned complete tabulation for all minor sub-castes. Tabulation was concentrated on major castes, groups requiring administrative protection, and primitive tribes. 

Caste had been recorded in every Indian census since 1872, but the exact wording of the question, and the categories into which answers were sorted, shifted nearly every decade. By 1931, Column 8 of the general schedule asked simply for “caste, tribe or race.” Hutton was aware of its limitations (Hutton, 1933).

As our previous articles have shown, the census officials either refused to define caste or gave it a very broad definition given its dynamic nature. Hutton on defining caste writes: 

The census recorded a general decennial population growth rate of 10.6% across India. However, individual castes followed distinct growth trajectories shaped by public health conditions, geography, social practices around widow remarriage, and religious conversion.

Table: Population and Decennial Variation of Selected Major Castes (1911–1931)

(Source: Compiled from Chapter XII and Imperial Table XIV)

Exterior and Depressed Castes 

The 1921 census had included a category ‘Depressed Classes’ which was to refer to castes that occupied a low ritual status within the Hindu caste order, meaning that economic deprivation was not the only criterion. Social and cultural deprivations faced by these castes were the primary criterion, but this term lacked a precise definition. Hutton uses the term ‘Exterior Castes’ instead of terms such as ‘Depressed Classes,’ ‘Untouchables,’ or ‘Outcastes.’ 

He defined Exterior Castes as communities residing within the Hindu fold who were subjected to severe social, civil, and religious disabilities imposed by ‘Clean’ (Caste) Hindus. To systematically identify these groups across different provinces, Hutton formulated a nine-point diagnostic test (Hutton, 1933):

1. Barrier to Temple Entry: Whether the caste was prevented from entering Hindu temples or places of public worship.

2. Priestly Refusal: Whether Brahmin priests of good standing refused to perform religious ceremonies for the caste.

3. Occupational Service Restrictions: Whether traditional occupational castes (such as barbers, tailors, or washermen) refused their services to the community.

4. Public Access Restrictions: Whether the caste was barred from using public wells, schools, roads, or public offices.

5. Pollution by Touch or Proximity: Whether physical contact—or in some southern regions, mere proximity (distance pollution)—was believed to cause ritual pollution to caste Hindus.

6. Social Separation: Whether the community was forced to live in segregated hamlets (Paracheris, Wadas) outside the main village site.

7. Educational Barriers: Whether children of the caste were forced to sit outside schoolrooms or denied admission.

8. Eating and Drinking Taboos: Whether clean castes would accept water or cooked food from their hands.

9. Economic and Civic Degradation: Whether the group was subjected to severe customary restrictions regarding clothing, ornaments, or land ownership.

Hutton later clarifies that an exterior caste can raise its position over time and that this has happened in the past. In other words, “exterior” is not being presented as an unchangeable status; a caste could potentially move from being “exterior” to what he calls an “interior” caste (Hutton, 1933). 

Hutton discusses the main criterion traditionally used in Assam to distinguish what he calls “clean” and “unclean” castes. It depended on whether members of higher castes would accept water from a particular caste. A caste whose water is accepted is called “jal-chal” (the passage also gives the local terminology “pani chak”), while one whose water is not accepted is called “jal-achal” (“pani na chale”).

The author says that all exterior castes are jal-achal, but not all jal-achal castes are exterior castes. Being on the wrong side of the “water” boundary is therefore not enough to put a caste into the category of “exterior caste.”

He gives the Shahas of Sylhet as his example. They are technically a jal-achal caste because their water is not accepted by upper castes. But they are wealthy and influential, are treated with considerable respect, and have even acquired the unusual privilege of being able to purchase brides from higher castes. Because of their wealth and influence, the author does not classify them as an “exterior caste” (Hutton, 1933). 

Interestingly, Hutton notes a disagreement to this advanced by Babu K. C. Dutta, Extra Assistant Commissioner: 

“the Shahas as an exterior caste simply because of their education and the wealth and influence they command. Yet they are not jal-chal, they are not allowed entrance into the Thakurghar of the cleaner castes and the disability is not likely to be relaxed in the next decade. [...] These exterior castes are suffering from want of education, wealth, culture, in fact, all that contributed to social influence. As soon as they have achieved these, their position will be akin to that of the Shahas–neither clean nor depressed.” (Hutton, 1933)

Dutta here makes an important argument that the conditions producing the marginality of “exterior castes” are themselves partly social and material. These groups suffer from a lack of education, wealth, and influence, and when they acquired these, their position could become similar to that of the Shahas: “neither clean nor depressed.” Put simply, social advancement could weaken the practical consequences of ritual exclusion, but the ritual boundary itself persists, and these groups continue to be subjected to social exclusion.

Economists and historians have since treated this nine-fold test as the single most consequential piece of technical apparatus to emerge from the 1931 census. Ashwini Deshpande terms it “the most elaborate effort to identify the untouchables” attempted anywhere in the colonial record (Deshpande, 2023). 

Ambedkar, who had to work with the resulting lists for the rest of his career, later warned against reading too much precision into any of the nine criteria individually: “it is a fatal mistake to suppose that differences in tests of untouchability indicate differences in the conditions of untouchables” (Ambedkar, 1932).

Hutton treated the right to use public conveniences, roads, wells, schools, as the primary test from the state's point of view, with religious disability as a contributory factor (Hutton, 1933). He was candid that the tests conflicted with one another in application: the Cochin Superintendent excluded the prosperous, well-organised Iruvans from his list on these grounds, while acknowledging that groups such as the Shabas, Telis, and Mahishyas of Bengal and Assam would themselves “protest at any such inclusion” were they listed (Hutton, 1933). 

Because no uniform definition existed, provincial lists diverged. In the Central Provinces, Hutton noted, the castes counted as depressed varied even from district to district within the same province (Hutton, 1933).

The resulting province-by-province lists, reproduced in the appendix with population figures down to individual castes, such as 987,570 Bagdis and 2,094,957 Namasudras in Bengal, or 76,816 persons across fourteen listed castes in Ajmer-Merwara, were aggregated into an all-India total (Hutton, 1933). 

The appendix itself records the sum of provincial tables at “a little over fifty million,” while the separate Imperial Table XVII, compiled on a slightly different basis, gave 39,137,405, a discrepancy Hutton attributed to differing provincial compilation methods and one he did not attempt to fully reconcile (Hutton, 1933: 472, 474). Set against earlier, cruder estimates, the 1931 figure sat closer to the higher end of a wide and contested range (see below).

Table: Competing estimates of India's Depressed/Exterior Classes population before 1931

Source: J. H. Hutton, Census of India 1931, Vol. I, Appendix I.

The appendix ends with a long extract from Mullan’s note on Assam, which Hutton presented as an example of how the tests were meant to work. He chose Assam because its large tribal and hill population meant that “interior Hindus” were a smaller proportion of the population, making the distinction between “depressed” and “exterior” groups especially clear (Hutton, 1933). But the debate did not end with Hutton’s report. 

The Lothian Committee later reduced the tests to just two (Section 4), and these were formally adopted in the 1936 Order. By the 1941 census, the next census to tabulate these groups, they accounted for 19 percent of India’s Hindu population and 12.6 percent of the population of undivided India as a whole (Deshpande, 2023). These figures give a sense of the scale of the population that Hutton had estimated, a decade earlier, at roughly 50 million. 

Hutton’s formulation of the Exterior Castes list provided the direct empirical framework for the Government of India (Scheduled Castes) Order of 1936, issued under the Government of India Act 1935. 

This Order established the term Scheduled Castes, which we continue to use and was defined in the following manner: 

‘such castes, races or tribes or parts of groups within castes, races or tribes, which appear to His Majesty in Council to correspond to the classes of persons formerly known as the “Depressed Classes,” as His Majesty in Council may prefer.’ 

This administrative schedule was subsequently adopted into Article 341 of the Constitution of India in 1950.

Mired in Petitions 

Enumeration relied predominantly on local knowledge. There was no centrally issued dictionary of castes. Enumerators asked respondents to name the caste, sub-caste, or endogamous circle outside which they could not marry. The answer was recorded as given, with no attempt by census officers to verify, adjudicate, or standardise it. 

This self-reported, unverified method had an unavoidable consequence of claims made by various communities. Hutton observed that “a caste which had applied in one province to be called Brahman asked in another to be called Rajput, and there are several instances at this census of castes claiming to be Brahman who claimed to be Rajput ten years ago” (Hutton, 1933). 

The Superintendent for Madras critiqued that caste sorting was “really worthless unless nomenclature is sufficiently fixed”. In his province alone, 227,000 people who had returned themselves as Ambattans in 1921 had shrunk to 10,000, while entirely new labels (Navithan, Nai Brahman, Pariyari) absorbed roughly 140,000 people (Hutton, 1933).

Caste associations petitioning formally to Census Superintendents had become a recurring occurrence. These associations sought to utilise the official government census gazetteer to validate claims of higher ritual status. In the face of these demands, Hutton instructed Census Superintendents to record the name returned by the individual to avoid local friction, but maintained strict criteria during the tabulation phase to preserve longitudinal statistical comparability with the 1901, 1911, and 1921 censuses.

Some of the artisanal groups such as the Blacksmiths (Kammara), carpenters (Kamsala), goldsmiths, and coppersmiths made the demand for a collective title of Vishwakarma or Viswa-Brahmin, claiming Brahminical status. This broader tendency of the groups towards consolidations that were previously separated by caste distinctions can also be observed among the grazier castes such as Ahirs, Goalas, Gopis, Idaiyans and possibly other groups of milkmen together under the common designation of ‘Yadavas.’ (Hutton, 1933).

These groups sometimes also sought to attach a designation suggesting membership in one of the two highest varnas, Brahman or Rajput. The report notes that the preferred designation at the 1931 Census was usually Brahman, although there were some cases in which a caste that had sought to be called Brahman in one province asked to be called Rajput in another.

The census officials explain these changes partly as a desire to ‘rise in the social estimation of other people.’ They also suggest that, in some cases, adopting a broader or higher-status designation could provide the backing of a larger community and thereby enable a group to ‘count for more’ in political life  (Hutton, 1933).

There was little evidence that the newly consolidated groups had begun to intermarry across their previous caste boundaries, barring a few instances between sub-castes of Brahmans and Kalars. Along with the consolidation of castes, there were also cases of internal differentiation. Reddies, Velamas, and Kammas were initially classified as part of the broader ‘Kapu’ category, but internal differentiation within the group led them to acquire distinct and higher-status identities by the 1931 census (Thirumali, 1992).

These claims by caste associations created a considerable problem for the census officials. The officials complained that every census generated representation from castes seeking recognition of particular claims about their status or history, often accompanied by what they regarded as questionable accounts of caste origins. Moreover, one caste's attempt to improve its social position could provoke objections from another caste making a similar claim (Hutton, 1933).

More importantly, sections of the Depressed Classes rejected the Hindu fold altogether. In Punjab, 418,789 Chamars and other Dalit groups, a number comparable to the province's Christian population, returned themselves as “Ad-Dharmis.” They professed an “original,” pre-Aryan religion distinct from Hinduism. This movement, absent in 1921, was disproportionately led by the literate (Pathak, 2026). Similar assertions of a separate “Adi” identity appeared as Adi-Dravida, Adi-Andhra and Adi-Karnataka in the south (Pathak, 2026).

These claims received more traction when Gandhi and Congress-aligned opinion began to press for non-cooperation with caste enumeration on the ground that officially labelling people by caste entrenched the system. A number of respondents accordingly returned “no caste” (Denault, 2023).

Hutton admitted that census officers would be relieved to stop tracking caste, as the officers were overwhelmed by a flood of petitions and doubtful histories from different groups, each fearing that another caste's social rise would threaten their own status.

The volume of contested claims led the 1931 census to abandon varna as an organising category altogether. Hutton noted that “practically every Hindu who claims to be a Hindu at all would claim to be either Brahman or Kshattriya” (Hutton, 1933). He proposed to revert to the pre-1901 method of grouping castes by traditional occupation (broadly following the Bertillon occupational scheme) rather than by alphabetical order or notional rank (Hutton, 1933). 

Tabulation thresholds were also relaxed. Only castes above four persons per thousand of population needed to be listed at all (up from two per thousand in 1921), except for Exterior Castes and “primitive tribes,” which were tabulated in full regardless of size, given their bearing on the franchise question (Hutton, 1933).

Socio-Political Implications

The Exterior Castes Index was compiled in the immediate aftermath of the Simon Commission’s 1930 report. The Commission was formed to suggest constitutional reforms for the 1935 Government of India Act. It flagged the “pitiable” disabilities of the Depressed Classes, and simultaneously, as B. R. Ambedkar was making a case for separate political recognition before the Round Table Conferences (Fuller, 2016). 

The Indian (Lothian) Franchise Committee, on which Ambedkar sat, drew directly on the 1931 figures when it came to define who should count as “Depressed” for franchise purposes, but it discarded Hutton's nine-fold test in favour of just two, temple entry and pollution by touch or proximity (Fuller, 2016).

Table: From the 1931 Exterior Castes appendix to the Scheduled Castes Order: a political-legal timeline

Source: compiled from Fuller, 2016; Ram B. Bhagat, 2006.

The 1936 Order was the starting point for what would later become the system of Scheduled Castes. After independence, this system was carried forward through Article 341 of the Constitution, which gives the President the power to officially notify Scheduled Castes in each state (Bhagat, 2006).

In this sense, one of the most lasting legacies of the 1931 census was the categories it helped create. The categories of “Scheduled Castes” and, through a parallel process, “Scheduled Tribes” continue to shape the distribution of reservations in education, government employment, and legislatures in contemporary India.

The census also demonstrably shaped caste identity itself. The Ad-Dharmi movement in Punjab, the Adi-Dravida, Adi-Andhra, and Adi-Karnataka movements in the south, and the coming together of grazier castes under the name “Yadava” were all political projects that used the census to make their claims. The census offered communities a chance to define themselves officially, knowing that these classifications could later be used by legislators, franchise committees, and rival caste organisations (Pathak, 2026).

This should not, however, be mistaken for the census having invented status competition between castes out of nothing. Julie Marquet's study of caste litigation in colonial south India shows disputes over such honours running continuously from the seventeenth century onward, well before the census existed to record them. From the 1880s these petitions found a new site, the census, to make these claims (Marquet, 2023).

Ambedkar's own verdict, given in 1947, diagnosed the politicisation of the census: “The Census of India has over a number of decades ceased to be an operation in demography. It has become a political affair. Every community seems to be attempting to artificially augment its numbers at the cost of some other community for the sake of capturing greater and greater degree of political power… The Scheduled Castes seem to have been made a common victim for the satisfaction of the combined greed of the other communities” (Ambedkar, 1947, quoted in Denault, 2023).

The 1941 census collected caste data but, at the direction of Commissioner M. W. M. Yeatts never tabulated it. He wrote, “the time is past for this enormous and costly table,” given the financial constraints that the Second World War had imposed on them (Pathak, 2026). 

Independent India's 1951 census went further, dropping general caste enumeration altogether, retaining only counts of Scheduled Castes and Scheduled Tribes. Caste, in Nehruvian state ideology, was interpreted as a “feudal hangover,” and the new republic would not participate in officially reinforcing it (Bhagat, 2006; Jodhka, 2026).

The clearest shortcoming of such an approach is evident in the case of the 1980 Mandal Commission. Tasked with identifying India's “socially and educationally backward classes” for a new tier of quotas, the Commission needed a population share for the group it was defining and, in the total absence of any post-1931 caste count, it proposed an arithmetic solution. 

It calculated the share of the population that was not classified as Scheduled Caste or Scheduled Tribe, not recorded as non-Hindu, and not considered Hindu “upper caste.” The share of upper-caste Hindus was estimated using figures from the 1931 census and then projected forward. 

The residual came to 43.7 percent, to which half of the non-Hindu population share was added, yielding a final estimate that 3,743 caste groups, comprising 52 percent of India's population, were “backward” (Deshpande, 2023). 

When V. P. Singh's government implemented the Mandal recommendations in 1990, the OBC quota was fixed at 27 percent, not because that was 52 per cent of anything, but because 27 was the residual left after the existing 22.5 per cent SC/ST quota under the Supreme Court's 1963 Balaji ceiling of 50 per cent on total reservations (Deshpande, 2023). 

Deshpande observes that,

Table: Competing estimates of India's “backward classes” (OBC) population share

Source: Compiled from Deshpande, 2023; Meshram, 2023.

Writing from within the OBC rights movement, Tanoj Meshram highlights why these numbers matter. He notes that the often-cited figure of 52 percent, still disputed in relation to the NSSO’s estimate of roughly 46 percent, is striking when compared with OBC representation in state institutions. Figures obtained through Right To Information Act (RTI) show that OBCs make up only around 12 percent of central government staff, while there were no OBC officers in the President’s Secretariat and only 9 percent among the roughly 650 officials of the Union Public Service Commission (Meshram, 2023). 

He, arguing for a caste-inclusive census, puts the underlying stakes plainly. “A census”, he writes, “plays an important role in the construction of social reality and collective identities... in the way that a map is essential to understand the geography of a nation” (Meshram, 2023). Hutton's 1931 census drew many such maps, which continue to affect the socio-economic, cultural and political domains in India today.

Contemporary Reflections and Critiques

Arjun Appadurai's related notion of numbers as a technology of colonial and national governmentality, and his description of “self-consciously enumerated communities” (Appadurai, 1993, quoted in Denault, 2023), captures precisely what the Ad-Dharmi and Yadava mobilisations of 1931 were doing. 

C. J. Fuller, examining the later careers of Hutton, Guha, and their contemporaries, has argued that colonial ethnography's usefulness to the state was already declining by the 1930s even as its bureaucratic weight grew. Official anthropology “could no longer strengthen the colonial state” even as its categorical outputs (Exterior Castes, Scheduled Castes) became more, not less, central to governance (Fuller, 2016; discussed in Denault, 2023).

The sharpest and most immediate critique, however, came from Ambedkar, who had used the 1931 data in the very franchise negotiations it fed, but who by 1947 dismissed decades of census ethnography as demographically worthless political theatre (Ambedkar, 1947, quoted in Denault, 2023).

A related but distinct historiographical debate concerns how much of caste as India knows it today was, in a stronger sense, produced by exercises like the 1931 census rather than merely recorded by it. Nicholas Dirks, in Castes of Mind (2001), pushed the “objectification” thesis furthest. Using revenue settlements, ethnographic surveys, and the census he demonstrates that exercises like the census did not simply describe a pre-existing caste system but textualised and rigidified fluid, locally negotiated, and politically contingent set of hierarchies. In short, for Dirks, “caste” as a single pan-Indian system is substantially a colonial construction. 

Susan Bayly has articulated her reservations against this thesis. Bayle suggests that hierarchical, endogamous, ritually graded social groupings with real force in marriage, commensality, and status were well documented in pre-colonial India. The colonial census only sharpened, standardised, and in places distorted an existing institution rather than inventing one (Bayly, 1999).

The 1931 census's shadow lengthened rather than shortened after independence because it was never repeated. The INC-led UPA government, responding to political pressure, commissioned the 2011 Socio-Economic and Caste Census (SECC) as a survey run in parallel to, rather within the regular census. It omitted a pre-defined list of castes; respondents were asked to name their caste, unprompted (Yadav, 2026; Pathak, 2026). Given that the SECC was not conducted under the Census Act of 1948, participation and data disclosure were technically voluntary. This weakened both its response rate and its legal authority (Meshram, 2023). 

The result was an unusable dataset of roughly 4.67 million (46 lakh) distinct caste names, compared with the 4,147 named castes the 1931 census had managed to compile using local officers' judgement and a rough occupational taxonomy; the raw data was never released (Pathak, 2026). 

Political scientist Yogendra Yadav critiques that this was a form of “bureaucratic deflection.” The open-ended question produced data too chaotic to use, insulating the state from the redistributive consequences of a workable caste count (Vithayathil, cited in Yadav, 2026). 

Bihar's 2023 state-level caste survey, by contrast, worked from a pre-compiled list and produced usable results, a template now cited by commentators such as Yadav and the sociologist Surinder S. Jodhka as the model the Census of India should have followed for the caste enumeration planned as part of Census 2027.

Instead of leaving the question completely open-ended, they suggest returning to an approach similar to the 1931 census, where people choose from a defined list of Scheduled Castes (SC), Scheduled Tribes (ST), and Other Backward Classes (OBC). To build this list, the government could use a major historical project, ‘People of India’ published by Anthropological Survey of India in 1980s and 90s, that cataloged 4,635 distinct communities (Yadav, 2026; Jodhka, 2026; Bhagat, 2006).

Table: The shifting count of India's castes

Source: Ram B. Bhagat, 2006 & Jodhka, 2026

Both Jodhka and Yadav converge on a paradox at the heart of the 1931 legacy: a colonial census whose caste categories were, on the government's own admission, provisional, locally negotiated, and internally inconsistent, but has nonetheless become the last reliable statistical baseline available to a democracy debating caste-based redistribution nearly a century later. 

What Hutton treated as a regrettable, but necessary task that the census would eventually stop.  Instead, this has become an ongoing, unfinished business of Indian statecraft. A question the state keeps almost answering, and keeps failing, by accident or design, to answer well.

(Authors: Nishitha Mandava, is an independent researcher and research consultant for Center for Legislative Education and Research, FLAME University, Pune;

Gaurav Kalyani works as Research Associate for the India State Stories Project at the Center for Legislative Education and Research, FLAME University, Pune;

Shivakumar Jolad works as Associate Professor (Public Policy), and is the Chair of Center for Legislative Education and Research and Director India State Stories, FLAME University, Pune)

(Author Contributions: Nishitha and Gaurav did research, primary writing and editing. Shivakumar contributed to conceptualization, research and editing)

Bibliography

Ambedkar, B. R. (2014 [1947]). Dr. Babasaheb Ambedkar: Writings and Speeches, Vol. 1, edited by Vasant Moon. New Delhi: Dr. Ambedkar Foundation, Ministry of Social Justice & Empowerment, Government of India. [Quotation as reproduced in Denault, 2023.]

Appadurai, Arjun (1993). “Number in the Colonial Imagination.” In Carol A. Breckenridge and Peter van der Veer (eds.), Orientalism and the Postcolonial Predicament: Perspectives on South Asia, 314–339. Philadelphia: University of Pennsylvania Press. [As discussed in Denault, 2023.]

Bayly, Susan (1999). Caste, Society and Politics in India from the Eighteenth Century to the Modern Age. The New Cambridge History of India, IV.3. Cambridge: Cambridge University Press.

Bhagat, Ram B. (2006). “Census and Caste Enumeration: British Legacy and Contemporary Practice in India.” Genus 62(2): 119–134.

Census of India, 1931 (1933). Appendix A, “Some Thoughts on the Caste System” In Volume XVIII — United Provinces of Agra and Oudh, Part I: Report, Chapter XII (“Race, Tribe or Caste”), pp. 541–543 (anonymous).

Cohn, Bernard S. (1987). “The Census, Social Structure and Objectification in South Asia.” In An Anthropologist among the Historians and Other Essays, 224–254. Delhi: Oxford University Press.

Denault, Leigh (2023). “Census, Caste Enumeration and the British Legacy.” In Surinder S. Jodhka and Jules Naudet (eds.), The Oxford Handbook of Caste. Oxford: Oxford University Press.

Deshpande, Ashwini (2023). “Reservations and Affirmative Action.” In Surinder S. Jodhka and Jules Naudet (eds.), The Oxford Handbook of Caste. Oxford: Oxford University Press.

Dirks, Nicholas B. (2001). Castes of Mind: Colonialism and the Making of Modern India. Princeton, NJ: Princeton University Press.

Fuller, C. J. (2016). “Colonial Anthropology and the Decline of the Raj: Caste, Religion and Political Change in India in the Early Twentieth Century.” Journal of the Royal Asiatic Society, Third Series, 26(3): 463–486.

Galanter, Marc (1984). Competing Equalities: Law and the Backward Classes in India. University of California Press.

Hodson, T. C., ed. (1937). India Census Ethnography 1901–1931. Delhi: Manager of Publications, Government of India.

Hutton, J. H. (1933). Census of India 1931, Volume I: India, Part I — Report, Chapter XII (“Caste, Tribe and Race”) and Appendix I (“Exterior Castes”), pp. 425–476. Delhi: Government of India Press.

Hutton, J. H., and Guha, B. S. (1935). Census of India 1931, Volume I: India, Part III — Ethnographical. Part A, “The Racial Affinities of the Peoples of India,” by B. S. Guha; Part B, “Ethnographic Notes by Various Authors,” edited by J. H. Hutton. Simla: Government of India Press.

Jodhka, Surinder S. (2026). “Cast(e)ing the Census.” The Tribune.

Marquet, Julie (2023). “Caste Disputes in Colonial India: Conflicts and the Legal Shaping of Caste.” In Surinder S. Jodhka and Jules Naudet (eds.), The Oxford Handbook of Caste. Oxford: Oxford University Press.

Menon, Dilip M. (2023). “Transformations of Caste in Colonial India.” In Surinder S. Jodhka and Jules Naudet (eds.), The Oxford Handbook of Caste. Oxford: Oxford University Press.

Meshram, Tanoj (2023). “The Importance of a Caste-Based Census.” Round Table India.

Padmanabha, P. (1978). Indian Census and Anthropological Investigations. New Delhi: Office of the Registrar General, India. [Prepared for the Xth International Congress of Anthropological and Ethnological Sciences.]

Pathak, Vikas (2026). “How India Plans to Count Caste This Time — and What Went Wrong Earlier.” The Indian Express.

Thirumali, I. (1992). Dora and Gadi: Manifestation of Landlord Domination in Telangana. Economic and Political Weekly, 27(9), 477–482. http://www.jstor.org/stable/4397664

Vithayathil, Trina. Counting Caste: Census Politics, Bureaucratic Deflection, and Brahmanical Power in India.

Yadav, Yogendra (2026). “Caste Census Is Dead. Don't Expect a Formal Announcement or Funeral.” The Indian Express.