On April 16 and 17, 2026, something rare happened in Indian parliamentary politics: a constitutional amendment failed. The Constitution (131st Amendment) Bill, which sought to redraw Lok Sabha constituencies based on 2011 Census data and implement women's 33 percent parliamentary reservation, fell 54 votes short of the two-thirds majority required. What made the defeat significant was not just the numbers—298 votes for, 230 against, a supermajority broken for the first time under Prime Minister Modi—but what it exposed about the machinery of representation itself. The government had rushed to conduct delimitation before the caste census could yield data that might reshape the political arithmetic altogether. The rush failed. The silence that followed matters more than the defeat.
Begin with a concrete scene: Swarupnagar Assembly Constituency in North 24 Parganas district, West Bengal. During the Special Intensive Revision of electoral rolls conducted in early 2026, approximately 3.25 lakh voter names were deleted—a 55 percent deletion rate, the highest in the state. Among them were members of the Matua, one of West Bengal's largest Scheduled Caste communities—a Namasudra-rooted reformist sect numbering an estimated 2.5 to 3 crore in the state, and thus an electorally consequential bloc. The Matuas voted in record numbers in 2026—the April elections saw 94 percent turnout across West Bengal, a demonstration of democratic enthusiasm. Yet approximately 2.7 million deleted voters, a disproportionate share Matua, remained pending tribunal appeals on polling day itself. They were counted in the census; they were enumerated in the electorate; and then they were erased from the rolls. This is the paradox that ought to animate what follows: Dalit voters are visible to the state's counting apparatus and vanish from its participation mechanisms with equal efficiency.
The machinery works like this. Because constituency boundaries were frozen in 1976 using 1971 Census data—an amendment inserted during the Emergency that was meant to be temporary but metastasized into permanent entrenchment—a Lok Sabha member from Bihar now represents approximately 2.7 million constituents on average, while one from Tamil Nadu represents 1.8 million. This disparity violates the foundational democratic principle of equal representation. It is also not accidental. Frozen constituency boundaries encode upper-caste political geography: the northern states that have remained over-represented under the freeze are precisely those where upper-caste landed interests consolidated their political base before and after independence. The southern states, which achieved higher literacy, lower fertility, and greater urbanization, were penalized by a mechanism dressed in the language of procedural neutrality.
The government's April 2026 rush attempted to break that freeze—but only partially, and strategically. A delimitation based on 2011 Census data would shift approximately 19 to 26 seats from southern states like Tamil Nadu, Kerala, and Andhra Pradesh to northern states, particularly Bihar, Uttar Pradesh, and Rajasthan. This would advantage the BJP in its strongholds; it would disadvantage the regional parties that dominate the south. But it would do something else as well: it would preempt the caste census. Because India is conducting its first caste enumeration since 1951, with Phase 1 of Census 2027 beginning on April 1, 2026, caste data was about to become available. That data could reshape representation calculations. Caste enumeration functions as a communal disaggregation tool: once detailed caste breakdowns become public, they become weapons in sectional politics. The government appears to have understood this clearly. The strategic analysis of the government's rush reveals that caste enumeration threatened the frozen seat distribution. Conduct delimitation first, before caste data lands, and you fix the arithmetic in place. The April defeat stopped this calculus.
Now the silence is instructive. With the constitutional amendment's failure, the Delimitation Bill and Union Territories Laws (Amendment) Bill were withdrawn. Delimitation must now await Census 2027 results, including caste data. The government has not announced a new timeline. The caste census proceeds—Phase 1 house listing continues through September 2026, population enumeration begins in February 2027, provisional results are expected by late 2027. And the state sits silent about when it will act on what caste enumeration reveals. This is not neutral procedural delay. This is conscious entrenchment by another means.
To understand what is being silenced, one must return to B.R. Ambedkar's warning in the Constituent Assembly Debates. Ambedkar, who chaired the Drafting Committee, argued that suffrage without substantive political power is ornamental. The vote matters only if it yields legislative muscle. To grant Dalits the franchise while freezing the boundaries that determine representation is to reduce them to symbolic inclusion—counted, visible, and powerless. This is what the 1976 freeze accomplished: it locked caste geography into place, ensuring that the upper-caste political formations that dominated certain regions would remain entrenched regardless of demographic change or democratization.
The parallel is worth naming. During Reconstruction in the United States, freedmen were granted the ballot but not redistricting power. Electoral geography remained shaped by slavery's legacy. It took a century of struggle to begin redrawing those lines. India faces a sharper dilemma: it is about to produce the most granular caste data in the nation's history, and the government's response is silence. Not denial. Silence. The state cannot count what it claims not to see, as one framework has it—but the inverse is also true. The state can count what it refuses to act upon. Caste enumeration without delimitation is enumeration divorced from consequence.
The counterargument deserves serious hearing. Southern states argue that delimitation without waiting for Census 2027 would constitute historic injustice. These states contribute disproportionately to national tax revenue—up to one-third of GDP—yet receive proportionally less in central transfers than less developed northern states. They have achieved demographic control through effective public health and education investments. To penalize them with seat losses would reverse the 1976 freeze's protective logic and reward fertility rather than development. This is a powerful argument about federalism and fairness.
But it contains a silence of its own. The southern states' defense of their overrepresentation does not address caste. It speaks of tax contribution, fiscal federalism, and demographic management. It does not ask: Within Tamil Nadu, Kerala, and Andhra Pradesh, who has access to the resources that enable lower fertility? Whose education enabled demographic transition? The answer involves caste—which is precisely why caste enumeration matters. Once caste data is public, the question becomes unavoidable: What do constituency boundaries look like when you disaggregate by caste? Which reserved seats are inadequate? Which nominally unreserved constituencies contain large Dalit populations without reserved status?
This is what the government's silence protects against. The projection of seat losses under 2011 Census delimitation shows that eighteen states would see changes in representation, with multiple states gaining or losing three or more seats. The data already exists. Once caste enumeration is complete, another layer arrives: precisely which Dalit populations would gain or lose representation under different delimitation scenarios? Which communities would be packed into unreserved constituencies?
The caste census, intended by its framers as an instrument of justice, becomes instead a measuring device for silence. It will count—in granular detail—what the political system continues to refuse to act upon. Dalit voters deleted from rolls. Dalit families facing land acquisition by tech capital. Dalit populations trapped in frozen constituencies that predated their political visibility. And the state will count them all, file the data away, and maintain the machinery that renders them visible only to be erased.
What is required is not neutrality. It is not procedural patience. It is deliberate action: delimitation that incorporates caste data, that redraws boundaries to reflect the actual composition and representation deficits of constituencies, that treats the caste census as what it was always meant to be—a tool for substantive political power, not ornamental enumeration. Until that happens, India has achieved the cruelest form of irony: a democracy that counts its marginalized with precision and denies them consequence with equal rigor.
Sources
The Constitution (131st Amendment) Bill, 2026 – PRS India
Lok Sabha Rejects Constitution (131st) Amendment Bill 2026 – LiveLaw