When the Count Becomes the Problem It Was Meant to Solve

The Register Is Not the Reality

The satire in "13 crore gone, in a few months" makes a serious point precisely by refusing to be serious about it: an exercise whose success is measured by how many names it deletes has already confused the register for the reality it is supposed to reflect. That confusion is not new and it does not stay confined to elections.

What SIR's Numbers Actually Show

A voter roll that shrinks by nearly a third in one city or by more than two crore names in one state, within months, demands one of two explanations: either the machinery that maintained those rolls for years - including through a general election - was catastrophically unreliable or the current exercise is itself generating errors at a scale that should worry anyone who cares about accurate governance. Neither explanation is reassuring and the essay's satirical suggestion that India could "solve" overpopulation by simply deleting enough names from a register makes the underlying danger vivid: a bureaucracy under pressure to show results can start optimizing the measurement instead of the thing being measured.

Why This Is Not Only an Indian Story

This same drift appears wherever a state's success is judged by a number it directly controls. In 1937, the Soviet Union conducted a census that found the population significantly smaller than official propaganda had claimed - a gap partly explained by the deaths caused by forced collectivization.

Rather than confront what the shortfall revealed, the state suppressed the census results and had its senior statisticians arrested, with some subsequently executed, for the crime of an "undercount." A second census, conducted in 1939 under direct political pressure, obligingly produced a larger, more politically acceptable figure.

The register did not describe reality; reality was made to fit the register and the people who reported the discrepancy paid for it directly.

The Common Failure

Both cases share the same structural temptation: when a count becomes politically consequential - determining who can vote or what a regime's demographic success looks like - it becomes administratively easier to adjust the count than to fix what the count was measuring.

A voter deleted in error cannot correct the record on their own initiative; a statistician silenced cannot publish a revised figure. In both, the people the register was built to serve become instruments of the register's own credibility instead of its beneficiaries.

Where This Leaves the Present Exercise

The lesson is not that electoral rolls should never be cleaned, any more than it is that no country should ever take a census. Dead voters, duplicates and permanently shifted electors genuinely should be removed; that is not in dispute.

The lesson is narrower and more exacting: the measure of success can never be the size of the deletion. It must be whether every eligible citizen remains findable, correctable and heard - because a wrongly excluded voter, unlike a wrongly included one, has no register left through which to object.

A citizen does not stop existing because a Booth Level Officer did not find them at home, any more than a Soviet peasant stopped existing because a suppressed census said so.