A lynching in Ghaziabad, and the arithmetic of impunity

The Ghaziabad lynching did not begin with a viral video. It began with the recognition of a sound, or a face, or a phone, by a stranger in a moving crowd. By the time the clip finished its circuit through WhatsApp groups and newsroom inboxes, the man on the ground had been reduced to a single function: a moral test for everyone who watched. The test, as these tests always do, ran in two directions. There was the question of what had been done to him. And there was the question of what would now be done, by courts, by police, by political leaders, by editors, in his name.

The arithmetic of impunity is the part that compounds. A death in public, filmed on a hundred phones, produces an arrest count, a chargesheet, a trial calendar, and eventually a verdict. The verdict, when it comes, measures something narrow: whether the state, in this instance, treated the killing as a crime. What it does not measure is the residue. The residue is the next mob, and the one after that, drawing the same conclusion: that recognition is enough, and that the interval between recognition and consequence has, on average, grown longer than anyone wants to admit.

What Ghaziabad tells us about the stack

India's criminal-justice system is not a single machine. It is a stack: a first information registerd at a local police station; a magistrate who decides whether the case is worth sending forward; a prosecution service whose caseload is measured not in years but in decades; a trial court that schedules hearings around the calendars of judges who are themselves handling a thousand other files; an appellate bench; a parole board; a remission process. Every layer adds latency. Every layer, by design, adds discretion.

The Ghaziabad case arrived at the bottom of this stack the way most lynchings do: as a rumour that hardened, somewhere between a scooter and a culvert, into a verdict. By the time the police arrived, the verdict had already been executed, and the only question left for the system was whether to call the execution a crime. The wire, predictably, framed it as a local incident. A scuffle, a death, a few arrests, a magisterial inquiry. The story was Ghaziabad, and the date was the date.

The arithmetic of impunity is not a Ghaziabad story. It is a stack story. The same stack, with local variations, processed the 2017 Dadri killing, the 2018 Dhule killings, the 2023 Comparator mob in Maharashtra, the Bulandshahr case that ended with a charge sheet naming a dead man. Read the charge sheets in sequence and you are reading the latency budget of a system. Sometimes the budget is short. More often it is not.

The recognition problem

The variable that mobile phones changed is not the killing. Mob violence in India predates the smartphone by decades; the 1984 anti-Sikh riots, the 1992-93 Bombay violence, the long history of so-called honour killings all show that public killing in India has a deep archive. What changed is the recognition step. In a crowd of strangers, a phone's camera now performs a function that used to require neighbourhood knowledge: it identifies the target.

That is the part that deserves scrutiny. A mob no longer needs to know who the person is. It needs only to believe that a person belongs to a category. The category does the work. Whether the category is religion, caste, language, or a rumour about the smell of food, the mechanism is identical: a signal is received, a category is assigned, a perimeter is closed. The phone does not invent the category. It removes the friction that used to slow the assignment.

Once the category is assigned, the stack starts running in reverse. The arrested are read through the same categories. Bail applications become arguments about identity. Trial calendars become arguments about who the victim was. The killing is converted, slowly, into a question about belonging.

The political economy of the verdict

What makes the arithmetic of impunity worth naming is the part that runs above the stack. A chargesheet is filed. A trial begins. Witnesses are named. Then, often, the political weather shifts. A government wants the case off the front page. An opposition wants it back on. A witness changes testimony. A judge is transferred. The verdict, when it arrives, is less a finding about what happened than a finding about what was, by the time of the verdict, politically affordable.

This is not cynicism about courts. Most trial judges in India are working under conditions that would not be tolerated in a system with a fifth of the population: cramped rooms, missing stenographers, expired witness summonses, and case lists that re-set every working day. The cynicism is about the inputs. When the inputs to a verdict are politically loaded, the verdict will be loaded too, regardless of the diligence of any individual judge.

What the repetition teaches

The point of returning to Ghaziabad, then, is not to relitigate the killing. The wire did that, and did it with the restraint appropriate to a local crime. The point is to ask what a country learns when the same shape repeats. A man dies in public. Phones film it. Arrests follow. A chargesheet follows. A trial follows. Then, usually, the news cycle moves on, and the arithmetic continues.

The question worth keeping is not whether the state can punish the killing. It usually can, given enough time. The question is whether the latency is being used, somewhere, as a feature. A long delay is a discount on the next offence. The longer the discount, the lower the cost of recognition-by-category, and the higher the probability that the next crowd, somewhere in India, will reach the same conclusion at the same speed.

India does not lack the tools to fix this. Fast-track courts exist; the 2018 Criminal Law (Amendment) Act exists; witness-protection frameworks exist on paper. What the country lacks, and what Ghaziabad quietly illustrates, is the political appetite to run the fast track when the category of the victim is politically inconvenient.

What to watch next

Three things will tell us whether Ghaziabad joins the long ledger of cases that became a precedent or the shorter ledger of cases that became a verdict. First, whether the chargesheet is filed within the timeframe the state has publicly committed to. Second, whether any accused are granted bail before trial, and on what grounds. Third, whether the political weather around the case changes before the trial concludes.

If the answers follow the pattern, the arithmetic will continue. If they break from it, we will know that something, somewhere in the stack, has begun to move.

Desk note: The wire framed this as a local crime story. Monexus read it as a stress test of India's criminal-justice stack, and asked what the repetition is telling us.