Essay · Section A5 October 2026

A decision is only as legitimate as the reasons its maker is willing to give.

Every examination teaches one humbling thing: a correct answer without working earns less than a correct answer with it. The examiner is not doubting that I know; he is asking to check. This small rule of the exam hall is the whole of my subject. Anyone who decides about other people, whether a clerk, a collector, a judge or a minister, holds a power others cannot see inside. The only bridge between that hidden judgment and the person who must live with it is the reason given. My claim in this essay is that a decision gets its authority not from the rank of whoever made it, nor merely from being right, but from the maker’s readiness to say why and to be told he is wrong.

Begin with why discretion cannot be abolished. No rule-book can foresee every case. Administration needs the officer who can see that a flood-hit farmer’s paperwork is late because the bridge is gone. The question is what keeps their deciding honest. Sir Edward Coke answered in 1598, in Rooke’s Case, with a definition that still reads like a lesson in public administration. Commissioners had been given authority to act “according to their discretions”. Coke held that their proceedings must still be bound by reason and law, since discretion is “a science or understanding to discern between falsity and truth, between wrong and right”, and not to act “according to their wills and private affections”. Discretion, in other words, was never freedom from reasons. It was the duty to find the right ones.

Why does the reason carry such weight? Because power seen from outside looks the same whether it is wise or wicked. Lord Acton wrote to Bishop Creighton in 1887 that “power tends to corrupt and absolute power corrupts absolutely”, adding that any presumption runs against holders of power. A reason is how that presumption is rebutted. It converts “because I say so” into “because of this, and you may test it”. Madison, or Hamilton, for the authorship of Federalist No. 51 is still debated, put the same thing in constitutional terms: in framing a government of men over men, “you must first enable the government to control the governed; and in the next place oblige it to control itself.” Obliging an officer to state his reasons is the cheapest form of that self-control.

The cost of omitting reasons is best seen in the Indian record. In 1976, in ADM Jabalpur, a majority of the Supreme Court held that during the Emergency a detained citizen could not even ask a court why he was held. Justice H.R. Khanna dissented, at the cost of the chief justiceship, and the nation has since treated his dissent as the Court’s true voice. The Court itself repaired the damage: in E.P. Royappa (1974) it held arbitrariness antithetic to equality, and in Maneka Gandhi (1978) it required that any procedure depriving a person of liberty be fair, just and reasonable, not whimsical. In 2014 the Court cancelled more than two hundred coal blocks allocated over two decades by a process that no one could explain on any principle. The larger loss was the public’s discovery that a national resource had been distributed by discretion that could not account for itself. The Right to Information Act, 2005, section 4(1)(d), now obliges every public authority to give reasons for its administrative decisions to those affected.

Reasons matter beyond the courtroom, and science shows it. In 1960 a newly joined FDA reviewer, Frances Oldham Kelsey, declined to approve thalidomide for sale in the United States. She had only a stated reason, that the manufacturer’s safety data were insufficient, and she repeated it through months of pressure. When the drug’s link to birth defects emerged in Europe, that reason had protected countless American children. What saved them was not her rank but a refusal that was written down, contestable and checkable. Peer review in science works on the same principle. A result is accepted not because a famous name announces it but because the method is open to those who would like to see it fail.

Literature has long known the opposite horror. Kafka’s Josef K. is arrested in The Trial and never told what he is charged with; the nightmare is not cruelty but opacity, a power that will not be asked. Shakespeare’s Measure for Measure shows Angelo, handed the Duke’s discretion, using it to condemn Claudio for a fault he is himself about to commit. Neither writer was against authority. Both saw that authority which will not explain itself decays from within, because the one who decides stops answering to anyone, his own conscience included.

The counter-position deserves its full weight. Some decisions are legitimate without stated reasons. A jury returns a verdict and says nothing more, yet we trust it, because its legitimacy comes from process: it is drawn from the public, it deliberates under rules, and its verdict is bounded by what was put before it. A surgeon in the middle of an operation, a commander in the field and a diplomat in a delicate negotiation cannot always stop and justify. National security may keep full reasons confidential. And there is a quieter danger: the demand for reasons can degrade into ritual, files of boilerplate justifying whatever was already decided. Finally, one who gives none can still be right; yet if correctness alone were the test, a lucky tyrant would be a legitimate ruler.

Each of these points sharpens the thesis rather than refuting it. The jury’s silence is acceptable because the system around it is open: the trial is public, the evidence is tested, the judge’s directions are on record and an appeal lies. The commander’s reasons are given later, to a court of inquiry or Parliament. Confidential reasons are acceptable only when someone independent, a judge or a committee, may read them. In every case reasons are deferred or delegated, not dispensed with. Justice Frankfurter, writing for the US Supreme Court in McNabb v. United States (1943), put the lesson in a sentence: “The history of liberty has largely been the history of observance of procedural safeguards.” Liberty lives in the unglamorous steps, the notice, the hearing, the written order, which make power answer. Against ritual reasons the remedy is not to drop them but to make them reviewable, so that boilerplate is exposed for what it is.

The shortcut is always tempting. Giving reasons is slow, invites challenge, and may reveal that the decision was weaker than it looked. The administrator who skips it saves an afternoon, but the saving is borrowed against trust, the one resource an institution cannot print. After the Ariyalur rail accident of November 1956, Lal Bahadur Shastri’s resignation as Railway Minister, offered once already after an earlier accident that year, was accepted by Nehru, who told Parliament he did so as an example of constitutional propriety, not because Shastri was personally to blame. Answerability is part of the office. A decision made by one person must in the end be held up before many, and the reason is the hand that lifts it. Whoever decides for others owes them, at least, the working.