Essay · Section A7 October 2026

The mirror and the window: institutions that cannot see themselves cannot serve others.

The hardest exercise in my preparation has not been learning facts. It has been marking my own answers. When I reread what I wrote, I read what I meant, not what is on the page, and I give myself marks the examiner never would. Only when a friend reads the same answer coldly do I see the missing example and the weak conclusion. Institutions suffer from the same blindness, with far larger costs. They need two kinds of sight. A mirror lets an institution see itself: its errors, its leaks, the distance between its claims and its results. A window lets citizens look in and lets the institution look out at the people it serves. My argument is that the mirror must come first. A window without a mirror becomes a shop display, and an institution can tell the public only what it has first allowed itself to know.

History offers a rare example of a state holding up a mirror. After the conquest of Kalinga, Ashoka had his Thirteenth Rock Edict record its cost: a hundred and fifty thousand people carried away, a hundred thousand killed, many times that number dead, and the king’s own remorse. Few rulers have carved their victims into stone. Yet even this mirror had a blind spot. The edict is missing at Dhauli and Jaugada, in Kalinga itself; there, separate edicts speak instead to officials, warning them that injustice grows from anger, haste and laziness. The lesson is double. Honest self-accounting is possible at the height of power, and even the most honest institution chooses where to place its mirror.

Science is the most developed mirror humanity has built. In September 2011 the OPERA experiment reported neutrinos that appeared to travel faster than light. Instead of announcing a revolution, CERN invited physicists everywhere to scrutinise the measurement and repeat it. By February 2012 the cause was found, a loose fibre-optic cable, and the experiment’s leaders stepped down the next month. Science lost a headline and kept its authority. The Nobel Prize awarded this week for the IceCube neutrino telescope rests on that same habit: a claim earns trust only when its makers have tried hardest to break it.

Governance shows what happens when the mirror is missing. Articles 315 to 323 create Public Service Commissions so that merit is not only done but seen to be done. In Karnataka, the Enforcement Directorate now alleges that the Commission’s own controller of examinations leaked questions for money. These are allegations, but one fact stands already: the alarm came from outside agencies, not from the Commission’s own records. An institution that cannot tell who saw a merit list before it was published has no mirror. The Press Council of India has worked for nearly a year without a single working journalist, because nominations await a ministry’s notification; the press’s own mirror is held up by a pending signature. Measurement can also flatter. India met the global threshold for eliminating leprosy as a public health problem in 2005, yet the latest programme report finds transmission halted in only 148 districts against the 300 projected. A target met can hide a disease still moving.

The window matters because the people who suffer an institution’s failures often see them first. In the 1990s the Mazdoor Kisan Shakti Sangathan in Rajasthan read government records aloud at village hearings, and labourers heard of work never done and wages they had never received. That window became the Right to Information Act, 2005, and the social audit. Today the test is digital. A new survey finds that Indians who trust online payments most are likelier to have been defrauded, while banks tell victims that they “collaborated”. Trust that cannot see is not confidence but exposure. Hans Christian Andersen understood this: in his tale, an entire court admires the emperor’s invisible clothes, and only a child, standing outside, says what everyone can see. Kabir advised the same humility: “निंदक नियरे राखिए, आँगन कुटी छवाय। बिन पानी साबुन बिना, निर्मल करै सुभाय॥” Keep your critic close, give him a hut in your courtyard; without water or soap, he makes your nature clean.

The counter-argument is serious. Too much mirror can paralyse. Officers who fear that every file will be audited, investigated and reported learn to decide nothing, and policy suffers. Some deliberation needs privacy; the RTI Act itself shields Cabinet papers until a decision is taken and the matter is complete, because advisers who speak for the record stop speaking frankly. Self-scrutiny can also turn into self-exoneration, since no one should be judge in his own cause. And windows can be staged: a dashboard of green ticks shows the public exactly what the institution wishes it to see.

These objections do not argue for less sight; they argue for better-designed sight. First, part of the mirror must be held by others: the Comptroller and Auditor General, independent members, a protected complainant, a court. Second, an institution must separate honest error from concealment. OPERA’s physicists were not disgraced for finding their own cable; they would have been for hiding it. Where reporting a mistake is safe and hiding one is costly, people will look. Third, it must measure what matters, transmission rather than prevalence, fraud recovered rather than complaints closed. Fourth, a window can open after the decision: records kept honestly today and opened tomorrow give advisers room to argue while still holding them to account.

I return to my answer sheet. I improve only when I let someone else read it and when I stop defending what I meant. Institutions are not different. Ashoka’s edict, the physicists’ cable and Kabir’s critic in the courtyard teach one discipline: to look at ourselves as others will see us, before they must. The citizen can see an institution through its window only as clearly as the institution has first seen itself in the mirror.