Nearly five months after West Bengal voted, a remarkable number has entered the public record. The Election Commission has informed the Supreme Court that 22.21 lakh of the 27.16 lakh electors deleted following adjudication of Special Intensive Revision, or SIR, "logical discrepancy" cases appealed against their exclusion. That works out to about 82 per cent.

The figure does not come from a political party, an activist organisation or a social media calculation. It comes from the Election Commission's own submission before the Supreme Court. But its significance must also be understood carefully. It does not establish that all those deletions were wrongful, nor does it prove that every appellant was an eligible voter. An appeal establishes that the decision was disputed. Whether the original decision was right or wrong can only be determined through adjudication.

The real democratic issue lies elsewhere. When such a large proportion of deleted electors challenge their exclusion and their appeals cannot be resolved before polling, the issue becomes one of timing and due process. Electoral remedies operate differently from many other administrative remedies because a vote lost in an election cannot be restored retrospectively.

How the Bengal SIR reached this point

The West Bengal SIR became unusually complicated because of what the Election Commission termed "logical discrepancies" in electoral records. These included various inconsistencies involving names, ages, parentage and other information. In February 2026, the Supreme Court intervened amid what it described as a "trust deficit" between the Election Commission and the then West Bengal government.

Around 700 judicial officers from West Bengal, Jharkhand and Odisha were eventually tasked with determining the eligibility of approximately 60.06 lakh electors flagged through this process. Of these, 27.16 lakh names were ordered to be deleted, while the remaining electors were retained. The Supreme Court subsequently directed the establishment of appellate tribunals to hear challenges to these decisions.

Nineteen tribunals began functioning on 13 April 2026. West Bengal went to the polls in two phases on 23 April and 29 April. The appellate mechanism therefore became operational only ten days before the first phase and sixteen days before the second.

That narrow window is central to the controversy.

What the 38.31 lakh appeals actually mean

One source of confusion in the public debate has been the repeated use of the figure of roughly 38 lakh appeals without distinguishing between different kinds of cases.

The Election Commission has now clarified that 38.31 lakh appeals were pending before the appellate tribunals. Of these, 22.21 lakh were filed by deleted electors seeking restoration of their names. Another 16.10 lakh challenged the inclusion of people who had remained on the electoral rolls and therefore sought further deletions. The Commission did not identify who had filed those challenges, although electoral law permits electors and electoral officials to seek deletion of persons alleged to be ineligible.
The distinction matters because the entire pool of pending cases cannot accurately be described as appeals by deleted voters.

Equally important is what the restoration appeals do not establish. A voter who was properly excluded may appeal, just as a voter who was wrongly excluded may appeal. Treating every appeal as proof of wrongful deletion would therefore be misleading. But treating every original deletion as conclusive proof of ineligibility would be equally problematic, because the very existence of an appellate process recognises that initial decisions can be challenged and reversed.

The most consequential number may be 1,607

According to the Election Commission's disclosure reported by The Indian Express, only 1,607 deleted electors were restored to the rolls in time to participate in the April Assembly election.
This does not mean that all the other restoration appeals were rejected. That conclusion would be mathematically and legally incorrect because large numbers of cases remained unresolved. The problem is precisely that adjudication had not been completed.

If an appellate tribunal eventually concludes that someone should have remained on the electoral roll, restoring that person's name months after polling does not restore the opportunity to participate in the election that has already taken place. There is no retrospective ballot.

This gives electoral due process an unavoidable temporal dimension. A remedy can be legally available yet practically ineffective for a particular election if it arrives after polling.

What the Supreme Court has already decided

Any examination of the controversy must also acknowledge the Supreme Court's May 2026 judgment concerning the Election Commission's authority to conduct SIR.

On 27 May, the Court upheld the legal power of the Election Commission to undertake Special Intensive Revision and held that such revision has a direct connection with maintaining accurate electoral rolls and ensuring free and fair elections. It rejected the broader argument that the Commission lacked the constitutional and statutory authority to conduct such an exercise.

That significantly narrows the present debate. The central question is no longer simply whether the Election Commission possesses the power to undertake an intensive revision. The Court has answered that legal question.

The unresolved issue is how that power was implemented in individual cases and whether the institutional machinery provided enough time to correct erroneous decisions before voting.

Legal authority to conduct revision and the accuracy of every decision made during that revision are separate questions. An administrative process can be lawful while individual decisions produced within it remain subject to appeal.

Did the SIR determine the Bengal election?

This is where evidence must take precedence over political assumption.

The Election Commission's published result page recorded the BJP winning 207 constituencies and the All India Trinamool Congress winning 80 among the results then declared. Former Chief Minister Mamata Banerjee and other TMC leaders alleged that the electoral-roll process helped the BJP, while the BJP rejected such allegations and maintained that the result represented the voters' verdict.

The existence of mass restoration appeals cannot establish how those electors would have voted. There is no credible basis for assigning them collectively to the TMC, BJP, Congress, Left or any other party. Some might also have chosen not to vote.

Nor can a statewide vote difference by itself answer the question. Assembly elections are decided constituency by constituency. Any serious attempt to examine whether exclusions could have affected individual results would require constituency-level information on deletions, restoration appeals, final appellate decisions, turnout and victory margins.

Without such evidence, the claim that the SIR changed the winner of the 2026 West Bengal election remains an allegation or hypothesis rather than an established conclusion.

The procedural question can be examined independently of that political claim: were eligible citizens given a reasonable opportunity to establish their eligibility before polling?

The case for rigorous electoral-roll revision

There is a legitimate case for maintaining clean electoral rolls. Deceased voters should not remain registered. Duplicate enrolments should be corrected. People permanently registered at the wrong address may need to be removed from that roll, and persons who do not satisfy the legal requirements for enrolment cannot acquire voting rights simply because their names once appeared on a list.

The Election Commission therefore has a constitutional responsibility not merely to conduct elections but also to maintain credible electoral rolls.

The difficulty arises because electoral integrity involves two different kinds of error. Allowing an ineligible person to remain on the roll compromises the integrity of the system. Removing an eligible citizen compromises it as well.

The institutional objective cannot therefore be measured by how many names are deleted or how many are retained. The relevant measure is whether eligible electors are correctly included and ineligible entries are correctly excluded, with sufficient safeguards to correct mistakes before they become irreversible for an election.

An Equidynamist reading of the problem

The controversy provides a useful example of the kind of institutional tension that Equidynamism attempts to address.

The competing principles here are electoral integrity and democratic inclusion. Neither can simply be sacrificed in favour of the other.

A system that concentrates exclusively on preventing wrongful inclusion risks wrongful exclusion. A system designed only to prevent wrongful exclusion may make it harder to remove entries that should not remain on the rolls.

Dynamic equilibrium requires institutional mechanisms capable of pursuing both objectives simultaneously. Verification must be rigorous enough to protect the credibility of the roll, while the notice, hearing and appellate processes must be fast and accessible enough to protect legitimate electors.

Seen in that way, the quality of an electoral revision should be judged not by the scale of the exercise but by the accuracy and timeliness of its final outcomes.

The information that is still missing

The next stage of this debate needs data rather than competing political narratives.

Constituency-level figures showing how many electors were flagged, deleted, retained, appealed, restored or finally rejected would allow the public to understand the scale and distribution of the problem. Aggregate reason codes would also be valuable. They could show how many exclusions arose from age inconsistencies, name discrepancies, parentage records, addresses, duplication, citizenship-related questions or other grounds.

Most importantly, final appellate outcomes need to become available.

Those outcomes could substantially change how the original deletion process is assessed. If the overwhelming majority of appeals are eventually dismissed after full adjudication, that would provide evidence supporting the reliability of many initial decisions. If substantial numbers are allowed, that would provide evidence of significant error in the earlier process.

At present, neither conclusion can responsibly be assumed.

For future large-scale revisions, one institutional question also deserves consideration: whether an extraordinary revision process should proceed into polling while a very large number of eligibility disputes remain undecided, particularly when the delay is not attributable to the voter.

The tribunal backlog raises another question

The Election Commission has proposed increasing the number of appellate tribunals in West Bengal so that their number corresponds with the state's parliamentary constituencies, with the objective of accelerating disposal of pending cases.

That proposal raises an administrative-capacity question. If more tribunals are now considered necessary to process the backlog efficiently, it is reasonable to examine whether nineteen tribunals were adequate when millions of cases had to be addressed around the time of an approaching Assembly election.

This need not be framed as an accusation. It is a question of whether the available adjudicatory capacity matched the scale and urgency of the task.

Electoral integrity begins before a vote is cast

India devotes enormous institutional attention to EVM security, polling booths, ballot secrecy, counting procedures and result certification. All of these are essential.

But accurate counting protects only votes that were actually cast. Electoral integrity also depends on whether eligible citizens were able to reach that stage.

The newly disclosed appeal figures therefore deserve examination without exaggeration and without dismissal. They do not establish mass wrongful deletion, election theft or a different hypothetical winner. They establish something more limited but still significant: an exceptionally large body of deletion decisions was formally contested, while the electoral calendar moved ahead before the appellate process could resolve most of those disputes.

Ultimately, credible elections require both sides of electoral-roll integrity to function together. Ineligible entries must be removed, and eligible citizens must have a meaningful opportunity to remain on the roll before their opportunity to vote disappears.