ON 5 OCTOBER 2026, the Supreme Court declined to suspend the Chief Election Commissioner on the strength of a newspaper report. The same day, it said it had not approved any modified Form 6. Each of these facts is being squeezed into a story the record itself does not tell.
The story is that the Special Intensive Revision (SIR) is a purge, the Court has looked away, and the Election Commission is a captured body run by one man. It is a thunderous indictment. It can also be checked against the record it condemns.
So let us do something unfashionable and read the judgment before we declare it dead.
START WITH WHAT the Court actually held. In Association for Democratic Reforms v. Election Commission of India (2026 INSC 564), decided on 27 May, the Bench of Chief Justice Surya Kant and Justices Joymalya Bagchi and Vipul M. Pancholi answered the four questions before it, on power, proportionality, documents and citizenship. It answered all four for the Commission. The power flows from Article 324 read with Section 21(3) of the Representation of the People Act, 1950. The safeguards of notice, hearing, speaking orders and appeal were held sufficient.
The Court also drew a line, and drew it clearly. Deletion from the roll is not a declaration of non-citizenship. The Commission’s inquiry is prima facie and limited to electoral purposes, and doubtful cases go to the competent authority under the Citizenship Act. That is the sentence that rarely travels.
Instead, on the day of the verdict, critics said the Court had let the ruling side decide who votes and had stamped a discriminatory practice. A limited eligibility check became, in Prashant Bhushan’s reported description, a blank cheque. If the Court wrote a blank cheque, someone should show where.
THEN THE FORM 6 uproar. The charge is that the Commission has rewritten the statutory form. On 5 October, senior advocate Menaka Guruswamy told the Court that “Form 6 has two new conditions added”. The Chief Justice replied that any modified Form 6 had not been approved by the Court, and said it would examine the record to see who had modified what. That is a request for the record, not a finding. Critics may read it as a rebuke. It is better read as a judge declining to decide before the papers arrive.
The May judgment drew the distinction the loudest critics skip. Form 6 sits in the Rules, and the Commission cannot change it on its own. But an SIR under Section 21(3) may need different paperwork from an ordinary revision under Section 21(2), and the Court recognised a limited residual power to design it, including an extra declaration alongside the form. In Bihar, new voters filed the statutory Form 6 with a separate declaration, Annexure D, and the Court noted that arrangement without suggesting the form had been amended.
The online form has not stayed so tidy. The Indian Express has reported that the version on the Commission’s ECINet portal carried an additional declaration linking applicants, and their parents or grandparents, to the last SIR, and that it was live in July, after the judgment. The Commission’s press note of 26 September says only that the SIR declaration was upheld and that the forms prescribed in the Rules apply outside the SIR period. Officials quoted by Aaj Tak went further and said the original form was untouched. Neither is an affidavit. The real question, then, is not whether anything was added, but whether what was added falls within the residual power the Court recognised.
That power has a boundary. Any extra requirement must have a clear link to cleaning the roll. If the ECINet declaration goes beyond it, the Court has the tools to say so. That is what a constitutional court is for.
The same discipline governed the second decision that day. The petitions seeking Gyanesh Kumar’s suspension alleged that SIR decisions were taken without the concurrence of the other two Commissioners, and they leaned mainly on an Indian Express investigation. The Bench issued notices to the Commission, the CEC and the other Commissioners. But it refused interim relief, observing that a news report can be wholly correct, partly correct or partly incorrect. It also asked who would conduct elections if the CEC were suspended without the Commission being heard, and said any decision later found to lack the required unanimity or majority could still be injuncted.
That is not a shrug. It is the ordinary discipline of the law: serious charges need authentic records. The Commission’s answer, in a statement issued on 26 September after all three Commissioners met, is that every SIR order had their unanimous consent. Notes of dissent, on that reading, are evidence of deliberation, not breakdown. If the records say otherwise, the Court will find out.
THE REST OF the case against the SIR rests on three moves. The first is the two-classes-of-citizens argument, built on Justice Ujjal Bhuyan’s address at NALSAR on 4 October. He did not name the SIR, but his remark that the Constitution “does not provide for two classes of citizens” has been widely read as aimed at it. The argument fails on its merits. The Constitution does distinguish between citizenship and the franchise. The Supreme Court has largely treated the right to vote as a constitutional and statutory right rather than a fundamental one, in PUCL (2003) and Kuldip Nayar (2006) among other cases. Article 326 confers it on citizens who are not disqualified, and Sections 16 and 19 of the 1950 Act set out who may be registered. Not every citizen is an elector. Article 14 permits reasonable classification, and checking who falls within the qualified class is what a revision does. Removing the dead, the duplicate and the shifted takes nothing from any genuine voter. It protects the weight of her vote.
The second move is scale. PTI’s count of nearly 6 crore names deleted in the SIR’s first year is real, but it includes the dead, the duplicated and the shifted. The 10 and 13 crore figures offered by P. Chidambaram and Shashi Tharoor are projections, not counts. All three are presented as if size were itself proof of arbitrariness. Compare them with Business Standard’s state-wise analysis: a net reduction of about 6% in Bihar and about 11.6% in Bengal after adjudication. Proportions, rather than rounded outrage.
Bihar’s claims window offers a further lesson. A Times of India record check found that party agents filed very few objections relative to the number they had appointed. Parties that now allege mass exclusion had the machinery to object at the time. If the deletions were so plainly wrong, where were the objections?
The third move turns one state’s litigation into the nation’s. In March the Court itself observed that, except in West Bengal, the revision went smoothly and generated little litigation, including in states with comparable deletions. Yet Bengal’s appeals, more than 37 lakh of them pending according to the Commission’s own affidavit of 16 September, are presented as the all-India method. Nor are they all from voters who were struck off: some 16 lakh of the 38 lakh filed challenge names that were included, not deleted.
Editorial vocabulary has done the rest. The Hindu’s editorial “Validating flaws”, P.D.T. Achary’s column in the same paper, Shashi Tharoor’s call to abandon the exercise and Mukul Kesavan’s “Electoral Inquisition” in The Telegraph have supplied a ready-made phrasebook. The opposing reading, that the exercise is a protection rather than a purge, gets far less airtime.
CRITICS ARE NOT wrong about everything. Nobody who is serious pretends the SIR has been executed flawlessly. Former Chief Election Commissioner S.Y. Quraishi wrote in The Indian Express that it is “a judgment that gets the constitutional law largely right and the ground reality almost entirely wrong”. That is a serious critique, and it reaches beyond clumsy execution to the exercise’s design and the documents it demands. The Court declined to advance the Bengal appeals ahead of the Nandigram and Rejinagar bypolls, calling them “only by-elections”, and delay in an appeal is a real hardship for a real voter.
Tribunal figures obtained under RTI and reported by The Wire in August put the success rate in decided Bengal appeals at around nine in ten, though only about 2% of appeals had then been decided. That figure cannot be waved away. If it holds, many of the deletions tested so far were wrong. It also means the appeals are catching them. These are empirical disputes, and only the appellate record can settle them.
But a bungled exercise is a reason to mend it, and a slogan of “scrap it” is not an electoral-roll policy. The disputes point to deciding the appeals faster, and the Commission has itself proposed more tribunals to do so. They do not point to tearing the roll up. If the Opposition thinks the safeguards are too slow, why is its answer to discredit the process that contains them?
THE CASE AGAINST the SIR is, in the end, a method rather than a finding. It begins with a conclusion, then treats every grievance as proof of it. A pending appeal becomes a purge. A dissent note becomes a conspiracy. A tribunal’s backlog becomes proof the judgment was wrong. Yet the rolls were not clean before this exercise. The dead, the duplicated and the shifted were on them long before the Commission went looking.
The remedy is accountability, not demolition. A confident Commission would publish its numbers: deletions by category, appeals decided and restored, citizenship cases referred, and the record of each collective decision. These are questions of governance, not loyalty.
Accurate rolls are the precondition of a free and fair election. If the objection is to the method, the Court has already prescribed the remedy. If the objection is to cleaning the roll at all, the public is entitled to ask why.
