Essay · Section A24 September 2026

An umpire is trusted not because he is always right, but because he belongs to no side.

In November 1986, Pakistan’s captain Imran Khan tried something Test cricket had not seen for decades. Tired of hearing that his side’s home wins were the work of home umpires, he invited two Indian umpires, Piloo Reporter and V.K. Ramaswamy, to stand in the Lahore Test against the West Indies. Pakistani umpires had not been proved dishonest; that was beside the point. A home umpire could give a correct decision and still not be believed, because the loser could always say he belonged to the other team. By 1994 the International Cricket Council required one neutral umpire in every Test, and from 2002 both. Umpiring did not suddenly improve; what changed was the loser’s willingness to accept it.

This week a Supreme Court judge reached for the same image. Examining the 2023 law under which the Prime Minister, a Union Cabinet Minister and the Leader of the Opposition select the Chief Election Commissioner and the Election Commissioners, Justice Dipankar Datta wrote that when the umpire is picked by one side’s captain, “even if the decisions taken by such umpire are correct, it has the patina of suspicion.” And he added: “It is not enough for the EC to be independent, it must also appear to be independent.” An arbiter’s authority rests not on being always right but on the loser’s belief that the call was made by someone with no stake in the result. This essay argues that this belief is part of independence, not decoration on it; that it rests on three visible things, how the arbiter is chosen, whether dissent can be spoken inside the institution, and whether reasons are given openly; and that the second is the most telling.

The law learnt this long ago. In 1999 the House of Lords set aside one of its own judgments in the Pinochet extradition proceedings because Lord Hoffmann, one of the judges, was a director of a charity linked to Amnesty International, which had intervened in the case. Nobody alleged that he was actually biased; it was enough that he stood too close to one side. Indian law asks the same question. In Ranjit Thakur v. Union of India (1987), the Supreme Court quashed a court martial on which sat the commanding officer the soldier had complained about, holding that the test is whether a reasonable person would apprehend bias, not whether bias is proved.

Appointment is the first thing the public sees. Article 324(5) protects the Chief Election Commissioner by making him removable only in the manner of a Supreme Court judge, but for seven decades Parliament made no law on how he should be chosen. In Anoop Baranwal v. Union of India (2023), a Constitution Bench filled the gap with a committee of the Prime Minister, the Leader of the Opposition and the Chief Justice. The 2023 Act replaced the Chief Justice with a Cabinet Minister. Parliament was entitled to legislate, but a panel on which two of three members sit in the same Cabinet answers the public’s question in the weakest possible way.

The second sign is subtler. A body that is never seen to disagree cannot prove that it deliberates. The Talmud records a striking rule: if every judge of a Sanhedrin voted to convict in a capital case, the accused went free, because a court in which nobody argued for him had not truly weighed the case. Article 145(5) of our Constitution preserves a Supreme Court judge’s right to dissent, and dissents are how a court argues with its future: Justice Subba Rao’s lone view in Kharak Singh (1962) that privacy is part of personal liberty was vindicated fifty-five years later in Puttaswamy. Under Section 45ZL of the RBI Act, the published minutes of the Monetary Policy Committee record each member’s vote and statement. A visible minority vote does not weaken the Reserve Bank; it tells markets that six people, not one, are thinking.

Hence the weight of the present dispute at the Election Commission. Two Election Commissioners are reported to have recorded objections at least fourteen times in ten months, saying decisions were taken in the Commission’s name without their knowledge. The Commission replies that “differing views and observations are a normal part of deliberation in any institution” and that its decisions were unanimous; the ruling party adds that if all three Commissioners only said “yes, yes”, the Opposition would cry dictatorship. Both are right that dissent is healthy, and both miss its purpose. Dissent reassures only when it is heard, answered and, in time, disclosed; the 2023 Act itself provides for unanimity where possible and a majority where members differ. A unanimity announced over recorded objections convinces no one. In 2019, when Election Commissioner Ashok Lavasa asked that his dissents on model code complaints be recorded in the Commission’s orders, the majority kept them in the files. The files kept the dissent; the public lost the evidence.

Dissent is not only a signal to outsiders; it is how institutions avoid mistakes. In Victims of Groupthink (1972), Irving Janis showed how President Kennedy’s advisers drifted into the Bay of Pigs fiasco of 1961 because doubters stayed silent in a room that prized agreement; in the Cuban missile crisis eighteen months later, Kennedy stayed away from some meetings so that his advisers would argue freely. In 1986, engineers who warned that the Challenger’s seals might fail in the cold were overruled, and seven astronauts died. Richard Feynman’s appendix to the inquiry report ends: “For a successful technology, reality must take precedence over public relations, for nature cannot be fooled.”

The counter-position deserves a fair hearing. Perception can be used as a weapon. Losing sides have always blamed the umpire, and a perception test that yields to the loudest partisan would let every defeated party discredit every result. Disagreement can also paralyse: the annual summary revision that adds eighteen-year-olds to the rolls, held since 1958, has reportedly not been resumed, and the appraisal reports of nearly fifty officers are stuck because the three Commissioners cannot sign together. And the test cuts both ways: the collegium, which Justice Datta defended this week, is criticised by many as “judges appointing judges”.

These objections refine the argument; they do not defeat it. The right test, as the House of Lords held in Porter v. Magill (2001), is the view of a fair-minded and informed observer, not of the angriest spectator. Appearance should follow from structure, not from public relations: selection by a body that no single side controls, secure tenure, a recorded right to dissent disclosed after a decent interval, and reasons published with every major decision. Paralysis comes from a failure to decide, not from disagreement, and the law already supplies the remedy, which is a majority vote recorded as such. The same standard of openness should apply to the collegium as to the Cabinet.

Cricket offers one last lesson. When the Decision Review System was first used in a Test in 2008, many feared that letting players challenge umpires on camera would destroy their authority. It did not. Umpires are now overruled in public, yet their decisions have become easier to accept, because everyone can see that a wrong call can be put right. A republic’s umpires are no different. Voters do not need an Election Commission that is never wrong. They need one that belongs to no side, lets its members disagree and shows its working, so that even the losing captain can walk off the field believing the game was fair.