146 Deaths Behind a Locked Door, and the Laws Bought With Them

The Triangle Shirtwaist fire began late on the afternoon of Saturday, 25 March 1911, on the eighth floor of the Asch Building near Washington Square in New York, and it was over in roughly eighteen minutes. One hundred and forty-six people died. Most of them were young immigrant women, many of them Jewish and Italian, some of them teenagers. I keep coming back to this fire not because it was the deadliest workplace disaster in history — it was not — but because of what happened in the eight years after it, and because of how thoroughly we have forgotten that the safety rules we now call red tape were bought at that price.

They had already asked

The part of the story that gets left out of the anniversary posts is that the workers had raised the alarm before there was a fire. In late 1909, shirtwaist makers across New York walked out in what became known as the Uprising of the 20,000 — a strike led largely by young immigrant women, most of them in their teens and twenties, in an industry nobody expected to organise. Triangle was one of the shops where it started. The demands were the ordinary ones: shorter hours, better pay, an end to petty fines and charges. They also included, by several accounts, doors that were not locked and fire escapes that actually worked.

The strike ended in the winter of 1910 with hundreds of manufacturers signing agreements. Triangle was among the holdouts. Its workers went back without a union contract and without the safety concessions they had asked for. The owners had faced the demand, weighed it, and declined it. That is the detail that turns the fire from an accident into something closer to a decision with a delay built into it.

It is worth sitting with how the strike was treated at the time: as an impertinence, a disruption of a busy season, the work of agitators. The women picketing outside those buildings were described in language we would recognise instantly today — troublemakers, outsiders, people who did not understand the economics of the business. Fourteen months later, the economics of the business were on the pavement of Greene Street.

Eighteen minutes

The fire started among the scrap fabric and cuttings that accumulated under the tables — the ordinary debris of a garment shop, in a building full of cloth, tissue paper and oil-fed sewing machines. It moved through the eighth floor and up to the ninth and tenth. Word reached the tenth floor, where the owners and the office staff worked, and most of the people there got out over the roof. The ninth floor, packed with machine operators, did not get the same warning.

What they found when they tried to leave is the heart of every account of that day. One of the two stairway exits from the ninth floor — the Washington Place door — was, by the widely reported testimony of survivors, locked. The company’s practice of locking doors to control pilferage and stop unauthorised breaks was described by workers before and after the fire; the owners disputed that the door was locked at that moment, and the question was never settled to everyone’s satisfaction in court. The other stairway filled with fire. The single exterior fire escape, flimsy and inadequate for the number of people on it, gave way. The elevators made a few trips and then stopped running. The fire department’s ladders reached only the sixth or seventh floor, well below where people were standing at the windows.

Dozens of them jumped. I am not going to describe it further. It was witnessed by a crowd in the street, and the people who saw it never really stopped talking about it — one of them was a young social reformer named Frances Perkins, who would spend the rest of her working life on the consequences. The last of the dead were identified only in recent years; some were buried unnamed for a century. Whatever else is arguable about that day, the arithmetic is not: a room full of people who could not get out of a building because the ways out had been narrowed, locked, under-built or allowed to rot.

The workers had asked for unlocked doors and working fire escapes fourteen months earlier, in a strike their employers rode out. The fire did not reveal an unknown risk. It collected on a refused one.

The trial, and what it settled

Isaac Harris and Max Blanck, the owners, were indicted on manslaughter charges. The prosecution’s case turned on a narrow question: not whether the door was locked, but whether the owners knew it was locked at that moment. In December 1911, after deliberating for under two hours, the jury acquitted them. A later civil settlement is widely reported to have produced a payment of about seventy-five dollars per life lost, against insurance recoveries that substantially exceeded the losses — a comparison that has been made so often it has become part of the fire’s folklore, and which I would treat as directionally true rather than precise.

I think the acquittal matters more than it is usually given credit for, and not because it proves the men were guilty. It matters because it demonstrated, in public, that the existing law had no grip on what had happened. The criminal law asks about the state of one person’s mind on one afternoon. It is a poor instrument for the question actually at issue: who decided, over years, that a floor with hundreds of workers on it would have this many exits, this fire escape, no sprinklers, no drills, no inspection with teeth. That question has no defendant. It can only be answered by a rule.

So the response that mattered was not the verdict. It was the machinery that was built around the verdict’s failure.

What was extracted

In June 1911, New York State created the Factory Investigating Commission. Robert F. Wagner chaired it; Alfred E. Smith was vice-chair; Frances Perkins was among its investigators. This was not a body that read submissions in a hearing room. Its people climbed the stairs of actual factories and canneries at actual working hours, counted exits, tested doors, looked at who was working and for how long.

Out of that work came a wave of legislation across the next few years — the count is usually given as roughly thirty to thirty-six new laws, with a larger number of bills introduced. The specifics are worth naming, because in the abstract “worker safety regulation” sounds like a filing cabinet, and in the particular it sounds like a list of the ways 146 people died:

• Doors kept unlocked during working hours, and required to swing outward.
• Automatic sprinklers in higher-risk buildings above a certain height.
• Mandatory fire drills, so that people knew the way out before smoke was in it.
• Requirements on the number, width and construction of exits and stairways, and limits on how many people a floor could hold.
• A real inspection regime, with inspectors and the power to act.
• Limits on hours — a fifty-four-hour week for women and minors — and tighter restrictions on child labour.

Read that list again with the fire in mind. Every line is a headstone. The rules were not designed by someone who disliked business. They were reverse-engineered from a specific catastrophe by people who had counted the bodies.

The other half of the aftermath was organisational. The International Ladies’ Garment Workers’ Union grew in membership and in political weight; its garment locals became a force that could make the new rules stick between inspections. Wagner and Smith carried the lesson upward — Smith to the governorship, Wagner to the US Senate and the labour law that bears his name, Perkins to the Department of Labor as the first woman in a US cabinet, where she worked on the New Deal’s wage, hour and social insurance laws. The fire produced laws; the organising produced the political durability that kept the laws from being quietly unmade.

A regulation is a fossil. It records the shape of something that once killed people, preserved in the only material that outlasts memory: the statute book.

Why every generation is invited to forget

Here is the pattern that interests me. A rule is written after a disaster. It works. Because it works, the disaster stops happening. Because the disaster stops happening, the rule starts to look like an unexplained cost — a form to file, an inspection to schedule, a door that cannot be locked even though locking it would reduce shrinkage. The rule’s own success is what makes it look unnecessary. And so, reliably, one generation later, someone proposes to remove it, and the argument sounds modern and sensible: we are being held back by rules written for a different era.

Sometimes that is true. Rules do calcify, and some genuinely protect incumbents rather than people. But the burden of proof belongs on the side asking to remove one, and the question to ask is simple: what happened that caused this to be written, and has that thing stopped being possible? For locked exits in a garment factory, the honest answer is that nothing about human nature or cost pressure has changed. Only the rule has.

This is the same dynamic I have written about in how technology gets captured: the gains from a new way of producing things flow to whoever controls the means, and the costs settle on whoever has the least ability to refuse. A century earlier, English workers who broke the machines that were destroying their trades discovered how seriously the state took property over people — when breaking a machine was a hanging offence, the asymmetry was written directly into the criminal code. The Triangle owners were acquitted; the strikers of 1909 were arrested on the picket line. That gap between what the law punishes and what actually kills people is the constant.

The same bargain, further away

The bargain struck on Washington Place was speed and margin against the bodies of the least powerful people in the building. That bargain has not been abolished. It has been relocated and lengthened until it is hard to see from the buying end.

In April 2013, the Rana Plaza building outside Dhaka collapsed with garment workers inside, killing more than 1,100 people. Cracks had reportedly been noticed in the structure the day before; workers were sent in anyway. Read the accounts and the resemblance is uncomfortable: an unsafe structure, a production deadline, workers who raised concerns and were overruled, and a supply chain in which the brand whose label was on the shirt could say, truthfully, that it did not own the building. The main difference between 1911 and 2013 is distance. Distance is what lets a purchasing decision in one country stop feeling like a safety decision in another.

The people who close that distance are usually the ones with the least protection for doing it. That is why I pay attention to history’s whistleblowers — the inspectors, clerks and insiders who reported what they saw and paid for it — and why I think the contemporary version of the 1909 strike, from warehouse walkouts to gig workers unionising over pay and safety, deserves to be read as the same activity rather than as a novelty. Someone inside the system says the doors are locked. Whether anyone acts before the fire is the only variable that has ever mattered.

So when I hear safety rules described as bureaucratic excess, I do not hear an argument about efficiency. I hear a generation that has been kept safe enough by those rules to have lost the memory of what wrote them. The rules are not the cost of doing business. They are the receipt. One hundred and forty-six people paid for the ones in that list, and they only became law because the survivors, the unions and a handful of officials organised hard enough to convert grief into statute before public attention moved on. That conversion is not automatic. It has to be done deliberately, every single time — which is the actual lesson of 25 March 1911, and the reason the door in front of you opens outward.