Surveillance, But To Prevent: What The Cameras Cannot See - O.P. Singh

Earlier this week, in the national capital, the office that superintends the city's police gathered the senior officers and handed down the priorities of the season. There would be more cameras and more drones, more artificial intelligence and facial recognition, more watching of what the young say to one another online. There would be policemen, visibly, at the worst junctions, and a campaign against the small lawless theatre of the Indian road — the wrong-side rider, the family of four balanced on one scooter, the bare head where a helmet should be. The policing would be proactive, humane, citizen-centred. The safety of women, children and the old, the room was told, is non-negotiable. It was the kind of list to which no one could object and which almost everyone will misread.

The misreading begins with the cameras, because the cameras lead the list and will lead the coverage, and because there is a seduction in the all-seeing eye that every person who has worn the uniform has felt. I wore it for thirty-three years, the last of them in command of a state's police, and I never once wished for fewer instruments. The lens promises the one thing the work withholds, which is certainty — the grainy frame that turns a rumour into a name. What it does not promise, and what the people who buy it seldom pause to ask, is safety, which is a different thing entirely, and which the world has spent four decades learning to measure.

The learning is not flattering to the eye. The most patient study of public closed-circuit television we possess pooled eighty separate evaluations across forty years and found that the cameras reduce crime by something like thirteen per cent — a real number, and a modest one, and on inspection almost the whole of it earned in car parks, watching over parked vehicles, where it cut theft by half. Against the violence the order invokes — the assault in the unlit lane, the cruelty inside the home, where no municipal lens will ever reach — the effect thinned to nothing. Facial recognition, the eye's cleverer cousin, tells the same story from the far end. A British force's trials, audited by a national laboratory, show the technology has grown genuinely accurate; it will pick a wanted face from a moving crowd and rarely err. But accuracy is not prevention. The machine is superb at finding a man the state is already hunting, which is to say it works after the harm, on the question of who, and is mute on the only question a frightened city actually asks, which is whether anyone in uniform will come.

No camera has ever bought a gram of legitimacy, and no city has ever been safe without it.

Here the list, read against the evidence, quietly divides, and the unfashionable half turns out to be the gold. Take the line about visible policemen, which sounds like the oldest cliché in the trade and is in fact among the best-proven moves in modern policing — on one unforgiving condition. The constable who merely strolls through, or drifts past on his way to a call, changes nothing; the research on random patrol is blunt about its futility. What works is presence concentrated, with something close to surgical narrowness, on the few corners where harm collects. A review of seventy-eight such experiments found crime falling in the targeted places in sixty-two of them; a celebrated American trial put officers on foot in the worst violent spots and watched serious crime drop by roughly a quarter. Any Indian officer who has run a city knows the truth inside this finding in his bones. Our crime is not spread evenly across the map like butter; it pools in particular galis, particular hours, particular markets after dark. The thanedar who knows his ground has always known this. The evidence simply tells him to trust it, and to stay.

Then there is the line that reads as the least serious of all, the campaign against the bare-headed rider, which may save more lives than every algorithm on the list combined. The global tally, assembled by the Cochrane reviewers and underwritten by the World Health Organization, is that a helmet cuts a rider's chance of dying in a crash by about forty-two per cent and of serious head injury by about sixty-nine — and that the law delivers this only where a policeman actually enforces it. On the Indian road, where the two-wheeler is the family car and the helmet is too often a token swinging from the handlebar, this is not a traffic nuisance. It is, in plain arithmetic, the cheapest mass-casualty prevention the state has, and it requires no intelligence, artificial or otherwise — only a constable at a junction who does his job.

Last, and easiest to dismiss as decoration, is the instruction to police humanely, to keep the citizen at the centre. It is, on the evidence, the multiplier the whole list is searching for. When researchers pooled thirty studies for the Campbell Collaboration, they found that people who feel fairly treated by the police grow measurably more willing to trust them, to help them, to obey the law. This is the hardest currency in policing, and it has a name — legitimacy, the settled belief on a street that the force is fair and is theirs. It is what makes the woman dial for help before the blow rather than after, the shopkeeper name the boys who have been circling, the witness still a witness on the day the case is called and the easy thing is to forget. No camera has ever bought a gram of it, and no city has ever been safe without it.

None of this is a brief against the machines; the machines are necessary, and a chief who spurns them fails his people as surely as one who worships them. It is a brief about sequence, and about an order of operations the glossy half of every such list inverts. Surveil, by all means — but to prevent, not merely to watch; let detection follow prevention rather than stand in for it; and do the one thing almost no force in this country troubles to do, which is to measure what the citizen actually feels at the gate of the thana — heard, or hustled out — and then to mend it, district by district, in public, with a chief's standing tied to the result. A capital can make itself the most watched city in the country by this time next year and be no safer for a woman walking home, because safety was never a property of how much the state could see. It was always a property of whether, when she needed them, she believed the police were hers.

The above article has been written by O. P. Singh, who is former Director General of Police, Haryana and the author of his forthcoming books The Audacity of Order and The Fear Tax.

Disclaimer: The views and opinions expressed in this article are those of the author(s) and do not necessarily reflect the official policy or position of the publisher. While every effort has been made to ensure the accuracy of the information, the publisher is not responsible for any errors or omissions, or for the results obtained from the use of this information.