Navigating Identity Politics through Census Data in India

Numbers have rarely been neutral in India. When the British colonial government began systematically counting caste in the 1870s, it was conducting what looked like a bureaucratic exercise – but the consequences rippled across centuries of social and political life. Today, as India prepares to include caste enumeration in its next national census for the first time since 1931, those same questions about counting, classification, and identity are back at the center of public debate. Understanding how the census became so entangled with identity politics requires looking at both the machinery of the state and the social forces that shaped what that machinery measured.

The census as a tool of governance 🔗

The Indian census, first conducted in 1872 under British colonial rule, was designed to classify and administer an enormously diverse population. On the surface, this was a logistical project. But the act of counting is never simply neutral – it shapes what it measures. This is precisely what the French philosopher Michel Foucault captured in his theory of governmentality: the idea that modern states govern not just through law and force, but by producing knowledge about populations that guides how people see themselves and their place in society.

In colonial India, the census was a prime instrument of governmentality. Foucault traced a historical shift from sovereignty – power exercised through commands and coercion – toward a rationality of “government” that took the management of population as its core objective. The colonial census institutionalized this logic: by enumerating Indians according to caste, religion, language, and occupation, the British created frameworks through which people came to understand and define themselves. As scholars have argued using Foucauldian analysis, the colonial census helped construct caste at the national level, turning what had been locally varied, fluid social arrangements into standardized, nationally harmonized categories.

From fluid identities to rigid categories 🔗

British policies, including the enumeration of the population into rigid categories during the decadal census – particularly in 1901 and 1911 – contributed to the hardening of caste identities. The colonial classification system ranked and cross-referenced caste groups in ways that resembled zoological taxonomies more than social description. Census Commissioner H.H. Risley’s use of physical measurements to classify castes is a stark illustration of how the state’s knowledge-making apparatus embedded pseudoscience into the social fabric of India.

What made this particularly consequential was the feedback loop it created. Once communities were officially named and numbered, they began organizing around those official identities. Colonial classifications of Indian society, though often arbitrary and inaccurate, had the effect of fixing and objectifying what had earlier been a fluid sense of identity into enumerated communities. Caste associations lobbied British officials for favorable classifications that could translate into administrative privileges. The census did not simply describe Indian society – it actively participated in reshaping it.

Religion was caught in the same dynamic. In the 1891 census, the Commissioner’s approach meant that after counting Jains, Buddhists, Sikhs, tribals, Muslims, Christians, Parsis, and Jews, whoever remained was labeled as ‘Hindu’ – a residual category that bore little relation to the complex reality of religious life on the subcontinent. The enumeration of religious communities sharpened boundaries between groups, providing future political actors with data they could use to assert demographic claims and demand representation.

Post-independence India: counting and not counting 🔗

When India gained independence in 1947, there was a deliberate decision to step back from caste enumeration. After independence, the government led by Prime Minister Nehru stopped caste enumeration in 1951 to avoid reinforcing social divisions, retaining it only for Scheduled Castes and Scheduled Tribes, whose identification was necessary for constitutional protections and affirmative action. The logic was straightforward: if the census had helped entrench caste distinctions, removing caste from the census might help India move toward a more egalitarian society.

But absence from the census did not mean absence from social life. Caste continued to organize marriage, employment, residence, and political affiliation. And crucially, caste-based political mobilization continued – and intensified – without the support of fresh national data. This is a critical point that any account of census and identity politics must hold onto: the census does not create social identities from nothing. The political mobilization of communities often precedes and drives the demand for enumeration, rather than the other way around.

The Mandal Commission and the politics of OBC enumeration 🔗

The clearest illustration of this dynamic is the story of Other Backward Classes. The Mandal Commission, formed in 1979, was tasked with identifying socially and educationally backward classes, and it relied on 1931 census data – the only available – to calculate that OBCs made up 52% of India’s population. It recommended a 27% reservation quota for OBCs in central government jobs and educational institutions.

When Prime Minister V.P. Singh’s government implemented these recommendations in 1990, the response was explosive. Widespread student protests erupted, with incidents of self-immolation by students who feared their career prospects would be severely limited. Yet the same implementation also galvanized a pan-Indian OBC political identity. The Mandal implementation permanently changed India’s politics, leading to powerful regional OBC parties and more OBCs in Parliament and state legislatures – what political scientist Christophe Jaffrelot called a “silent revolution”.

What is significant here from a sociological standpoint is the sequencing. Backward caste movements had been building for decades before the Mandal Commission reported – socialist leaders like Ram Manohar Lohia had been mobilizing OBCs in North India since the 1950s and 60s. These trends had already merged in the Janata Party, an opposition coalition that defeated Indira Gandhi’s Congress in 1977, and it was this government that appointed the Mandal Commission in 1978. In other words, the political pressure came first; the census-adjacent enumeration followed as a response. The Mandal Commission did not invent OBC politics – it formalized and amplified a movement that was already in motion.

Does enumeration solidify social divides? 🔗

This is one of the most contested questions in the sociology of the Indian census. Supporters of caste enumeration argue that census categories merely reflect existing divisions and that it is necessary to link caste identity with inequality. Opponents point to the past experience of mobilization around officially recognized identities and the consequent dangers of both distorted data and increased social tensions. Both positions contain real insight, and neither is wholly right.

On the one hand, there is strong evidence that official classification can deepen and harden identities. When a community is named in a government list and attached to specific benefits or quotas, that naming has material consequences that encourage people to identify with the category. Since the middle of the 1980s, many Indian Christians have begun to assert their Dalit identity, mounting energetic campaigns to have Dalit Christians included on the Scheduled Caste list – a direct example of official classification reshaping how communities understand and present themselves.

On the other hand, caste as a lived social reality does not require a census to reproduce itself. Caste still shapes life: marriage, surnames, where people live, and even political choices for candidates and appointments. Between 2019 and 2020, a Pew Research survey found that 98% of Indians identified as members of a caste. The system persists with or without enumeration, which complicates the argument that simply not counting caste will weaken it.

Religion, demography, and political mobilization 🔗

The enumeration of religious groups in the Indian census has followed a similar pattern. Religious demographic data has become a resource for political parties, who use population shares as the basis for electoral strategy and claims about representation. When shifts in the relative size of religious communities appear in census data – or when such shifts are anticipated – they can generate anxiety and political mobilization among communities that feel their influence is under threat.

This is consistent with what Foucault described as the productive power of state knowledge: by making populations visible in specific ways, the state creates possibilities for action – including forms of political action the state itself may not have anticipated or desired. The census renders communities “countable” and therefore “claimable.” Once a group has a number attached to it, that number becomes a political argument.

Bihar’s caste survey and the renewed demand for enumeration 🔗

In 2023, the state of Bihar completed a yearlong caste survey that illustrated exactly how pre-existing political mobilization drives enumeration, rather than simply the reverse. The Bihar government found that disadvantaged castes comprised 84% of the population, and on this basis hiked the proportion of government jobs and school and college places reserved for those castes from 50% to 65%. The survey also showed that traditionally privileged groups like Brahmins were overrepresented in government jobs relative to their population size. This was not a revelation produced by counting – it was the confirmation, through data, of inequalities that communities had been mobilizing around for decades.

Activists have argued that a caste census is critical to uncover intersectional disparities across caste, region, religion, and economic status, and to design policies and programs that are truly equitable and inclusive. At the same time, critics warn that repeated official emphasis on caste categories could entrench those identities further and deepen divisions – a tension that has no easy resolution.

In April 2025, the Union Cabinet under Prime Minister Narendra Modi formally decided to include caste enumeration in the next national census. This will be the first comprehensive caste census since 1931, and is likely to reignite debates around social justice, equality, and the future of affirmative action programmes. Politically, the decision reflects sustained pressure from opposition parties and caste-based organizations – again, a case of mobilization driving enumeration.

What sociology tells us about the census-identity relationship 🔗

Foucault’s framework of governmentality offers a useful lens, but it is not the whole story. His insight – that the state’s act of categorizing populations shapes how people understand and govern themselves – clearly applies to India’s census history. The colonial census turned locally varied caste identities into rigid national categories, and that transformation had lasting political consequences.

But a purely Foucauldian reading can overstate the power of state classification relative to the agency of the communities being classified. While official classifications are important determinants of identity, they are not the sole factor. Social movements, economic grievances, cultural memories, and local political dynamics all shape how communities form and assert identities – often independently of what the state chooses to count. The census can amplify and formalize identities, but it does not manufacture them from scratch.

The India case suggests a more dynamic picture: census data and social identity are in a relationship of mutual influence. Communities shape what gets counted by mobilizing to demand recognition; what gets counted then becomes a resource that communities use for further mobilization. The census is neither a mirror that passively reflects society nor an all-powerful architect of social division – it is a site where the state’s governing ambitions and the political ambitions of communities meet, negotiate, and sometimes clash.

What do you think? Given that pre-existing social mobilizations often shape census categories more than census categories shape mobilizations, does counting caste in India’s next national census risk deepening social divisions – or does accurate data ultimately serve the cause of greater equality? And who should have the authority to decide how communities are classified in a national census – the state, community organizations, or individuals themselves?

References

1. https://www.epw.in/journal/2018/26-27/special-articles/1872-census.html
2. https://link.springer.com/chapter/10.1057/9781137385925_4
3. https://www.researchgate.net/publication/265445481_Caste_Confusion_and_Census_Enumeration_in_Colonial_India_1871-1921
4. https://en.wikipedia.org/wiki/Caste_system_in_India
5. https://stophindudvesha.org/indias-caste-census-debate-the-legacy-of-a-british-plot-to-undermine-hindus/
6. https://en.wikipedia.org/wiki/Caste_census
7. https://en.wikipedia.org/wiki/Mandal_Commission
8. https://en.wikipedia.org/wiki/Mandal_Commission_protests_of_1990
9. https://m.thewire.in/article/caste/a-look-back-at-the-indias-caste-census-journey
10. https://journals.sagepub.com/doi/10.1177/0011392100048003008
11. https://www.researchgate.net/publication/377822070_Census_politics_and_the_construction_of_identities_in_India
12. https://www.science.org/content/article/first-time-century-india-s-states-will-count-politically-sensitive-caste-membership
13. https://www.nationalheraldindia.com/national/explainer-caste-census-history-politics-and-its-significance-today
14. https://www.tandfonline.com/doi/full/10.1080/03068374.2025.2556238