Can somebody please explain what is going on in political India? The agitation Rahul Gandhi has joined is, in Prashant Bhushan’s words, a “fight for regime change”. Regime change is what Donald Trump wished on Iran’s clerics: The end of a political system, not a Congress victory in 2029. The foreign press, once the voice of relative sanity, has joined the chorus. The Economist (October 1) says the Election Commission “is barring tens of millions from voting”. The Financial Times (October 15) headlines an overhaul “that removed 130mn voters”, though its own chart says these are names “flagged for possible exclusion”. The protesters’ slogan is 13 crore votes deleted. And in these pages, Jean Drèze and Yogendra Yadav (DY) (‘Let us not be in denial, disenfranchisement is real’, October 20) find “well over 10 crore missing voters”, about 14 crore for all India.These are all-India numbers, including states whose final rolls are not yet out. So consider the nine states where the final rolls are known: Bihar and the eight large states of the second phase, home to three in five Indian adults.AdvertisementWhere registration is voluntary, the roll always falls short of the adult population: 4-7 per cent of eligible Canadians are missing from the lists, 11-14 per cent of Britons, and a quarter of American voting-age citizens. Nobody calls that disenfranchisement. Disenfranchisement is the gap between the final roll and the roll one should expect: Everyone who would normally register, and no one who should not be there.DY measure against all adults instead. Their table sets each state’s projected adult population from the government’s Technical Group against its final SIR roll. The nine states have 59.9 crore adults and 52.8 crore names on the final rolls: 7.1 crore, or 12 per cent, short. DY call this disenfranchisement. Their table, drawn from my Substack, leaves out a crucial column: The young who did not register to vote. In 2024, the ECI’s own rolls show, four in 10 young adults in Uttar Pradesh and Bihar were not registered. In these nine states, that is 5.2 crore young people, 8.7 per cent of all adults. The SIR cannot have deleted people who were never there. Add 1.5 per cent of adults for older people who never registered, or moved and never re-registered, and 6.1 crore adults would normally not be on the roll. Put that column back, and DY’s own numbers give the answer: About 1 crore disenfranchised, not 7.1 crore.The expected roll is 53.8 crore; the final roll is 52.8 crore. West Bengal, where the young register well, expected 7.4 crore and got 6.8: About 60 lakh short. And there the appeals failed: Of the 34 lakh Bengalis who appealed, The Economist reports, fewer than 0.05 per cent had their cases heard before election day, and the tribunals have since found 90 per cent of appellants valid all along. Bihar, where the outcry began, has 7.4 crore on its final roll against 6.8 expected: Largely migrants who work in other states but stay registered at home.AdvertisementNor were the old rolls clean. Before the SIR, these nine states had 58.6 crore names on their rolls. A clean roll, the table shows, would hold 53.8 crore. The difference, nearly 5 crore names, was the dead, the departed and the duplicated, and this on rolls that still missed 5.2 crore young people. Bloated rolls are not an excuse invented for the SIR. They are what the SIR removed. Drèze himself now concedes, in another newspaper (October 21), that the old rolls left out “up to 10 per cent perhaps” of adults, and that the “majority” of deletions are legitimate.Accepting Uttar Pradesh’s low youth registration, DY say, is “like saying if you skipped breakfast, you might as well skip lunch”. Clever, but nobody skipped lunch. Uttar Pradesh’s 2.3 crore unregistered young were counted by the ECI in 2024, before the SIR. Applying the national average to Uttar Pradesh is like measuring Rajathan’s drought with India’s rainfall.K B S Sidhu, a former civil servant, put it simply on X: The 130 million in the headlines “was never a count of the disenfranchised”. Article 326, he reminds us, sets only three conditions for the roll: “a citizen of India, eighteen years of age, ordinarily resident. A person who meets them shall be entitled to be registered. Everything else is fluff.”you may likeBy Sidhu’s test, the SIR was loaded with fluff, and the blame lies with the ECI and the government. In these pages on September 24 (‘I have voted for three decades. Why is my SIR incomplete?), I set out, from my own experience, what went wrong: The haste, the paperwork demanded of poor and migrant voters, and “logical discrepancy” notices for a misspelt name. Those criticisms stand. The government changed the law in 2023 so that it effectively chooses the Election Commissioners, and now wonders why the Commission’s impartiality is doubted; its ministers talk of “infiltrators” without producing even an approximate number for us mere mortals to assess. The Commission ran a hurried exercise, often just before state elections, let its discrepancy rules fall hardest on Muslims in Murshidabad and Malda and on Matua Hindus, added a declaration to Form 6 without changing the rules, and has still not published a national post-SIR roll, or removals by reason, state and booth. An umpire who will not explain his decisions invites the charge of bias. A crore of eligible voters wrongly removed is not nothing. But inflating 1 crore into 7 does not help the 1 crore; it lets the ECI dismiss the real crore along with the imaginary six.So start from first principles, from Article 326. The citizen is entitled to be registered; it is for the state to show that he is not. Take the passport. India issues one only to its citizens, after police verification. Yet in June an External Affairs ministry official declared the passport a travel document, “not a document of citizenship”, and the ECI had to clarify the next day that it still accepts one. A passport is the state’s own certificate of citizenship; if the state now doubts it, the state must prove its case before a tribunal, not ask the citizen to prove it again. That is a reform worth marching for. Regime change is not.The writer is chairperson of the Technical Expert Group for the first official Household Income Survey of India. Views are personal