Caste census in a fraught political moment: Selfie of the nation

Indian Express

16-06-2025

There are two obvious reasons the recently notified national Census of 2026-27 will be like no other Census in independent India. These have already attracted a lot of attention in the media: The counting of castes beyond the Scheduled Castes and Scheduled Tribes, and the new delimitation exercise that will determine the political weight of the states of the Union and the shape of our Parliament. However, the real reason the coming Census will be unique is the particular conjuncture in which the questions raised by caste and delimitation will have to be answered.
Neither delimitation nor the counting of caste is new. Caste had been counted nationally till 1931; even after Independence, caste data have been collected for the Scheduled Tribes and Scheduled Castes; and more recently, census-like surveys have been conducted in the states of Karnataka, Bihar and Telangana. Delimitation, too, has happened three times before, and the current delimitation formula (for the number of Parliament and assembly seats) is based on the 1971 Census, while the boundaries of constituencies have been redrawn based on the 2001 Census.
However, the context of delimitation today is very different from earlier times. The 84th Amendment of February 2002, enacted under the Atal Bihari Vajpayee-led NDA government, froze states' share of Lok Sabha seats to the allocations based on the 1971 Census 'until the relevant figures for the first Census taken after the year 2026 have been published' (the previous reference year had been 2000). Because seat allocation has been proportionate to population (to ensure that each parliamentarian represents roughly similar numbers of electors), this freeze was intended to forestall the apprehensions of the northern states gaining at the expense of the south.
In the mid 2020s, these apprehensions have grown stronger. Population projections suggest that (barring exceptions like Himachal Pradesh) the difference in the populations of northern and southern states will be proportionately more than it was in the 2000s. As widely noted, there is the narrative of the south being 'punished' for better performance on population control and larger contributions to national income, and the north being 'rewarded' for its backwardness on both counts. Perhaps the all-important factor here is that, electorally, the ruling BJP dominates the north but is almost absent in the south. The unusual delay in conducting the Census and the decision to push it beyond 2026 thus appears to be an unsubtle move to reap electoral advantage from the expected change in seat allocation for the next general election, due in 2029. At the very least, it will provide the ruling regime with a valuable and versatile political bargaining chip.
The circumstances surrounding caste enumeration are also very different today. The last nationwide caste count happened in the 1931 Census amidst protests from nationalist groups who saw this as a coloniser's divisive ploy. In the early 20th century, the Census did in fact help to accelerate the consolidation of caste identities because this was the moment of aggregation — a time when coalitions were being built across clusters of castes. The aggregation phase has continued after Independence (despite the absence of caste counts) and reached its peak towards the end of the century, with the Mandal upsurge of the 1990s. Today, caste enumeration promises to play an entirely different role for two reasons.
First and most important, as the first national Census after Independence that will count the castes of all Indians, the coming Census will end the statistical anonymity of the so-called 'upper' castes who have benefited disproportionately from state-sponsored development. The same castes have gathered even more disproportionate gains from the market-led development of the last two decades. The popular discourse on subjects like reservation — a discourse dominated by the 'upper' castes — has encouraged the illusory belief that visible welfare programmes are the only ways in which the state bestows benefits. On the contrary, as the history of post-liberalisation India shows, the state has been able to shower benefits on the already privileged in countless ways through the market as well. While the data it generates will be very basic, a census that counts the 'forward' castes will create the basis for a more complete understanding of the extent to which caste inequalities have been deepened by a supposedly caste-blind state — despite the presence of targeted welfare programmes for 'backward' castes.
Second, developments in caste politics over the past three or four decades have taken it out of the aggregation phase into a new phase of disaggregation. By offering data on individual castes within existing official or political caste-clusters and categories, the coming Census will help to highlight internal divisions that will challenge currently dominant political identities. Just as the consolidation of similarly situated castes was a necessary phase in history, today the questioning of existing categories and coalitions is the need of the hour. This applies equally to political identities other than those based on caste.
The challenge that irrefutable evidence of inequalities will pose to dominant ideologies of political mobilisation — whether of Hindutva, caste, religion or region — is an open-ended one. It leaves open the possibility that the challenge will be met with genuine attempts to address inequality. It also opens the door for the invention of new identities that bring together hitherto separate groups on the basis of common interests beyond existing labels. Either way, it will force currently dominant political ideologies to provide more substance to the unity they demand in the name of caste, community, or even nation.
There is another feature of the current conjuncture that may prove to be decisive. This is the global tendency of electorally secure authoritarian regimes to produce 'alternative facts' in response to inconvenient facts. Taken to its logical end, this tendency ensures that, wherever and whenever possible, inconvenient facts are aborted before they can be born. We can only hope that the census machinery will be spared such expectations.
Because, ultimately, the Census is a selfie of the nation, a self-portrait of the people that can only be painted by their government.
Deshpande is a retired professor based in Bengaluru