The 38 Witnesses Who Never Were: How One Famous Crime Story Shaped What We Believe About Helping Strangers

In the early hours of 13 March 1964, a 28-year-old bar manager named Catherine "Kitty" Genovese was attacked as she walked home to her apartment in Kew Gardens, in the New York borough of Queens. She was stabbed, and she died of her injuries. Her killer, Winston Moseley, was arrested days later and confessed.

Two weeks after the murder, The New York Times published a front-page story that would turn a local tragedy into a global symbol. It claimed that 38 respectable, law-abiding neighbours had watched the attack unfold over more than half an hour, and that not one of them had called the police until it was too late.

The story horrified readers. It was retold in newspapers, sermons, plays and, above all, psychology textbooks. For generations of students, "the 38 witnesses" became shorthand for a grim truth about modern life: people will watch a stranger die rather than get involved.

There was only one problem. It wasn't true, at least not in the way it was told.

What really happened

Over the following decades, journalists, researchers and the victim's own family went back over the evidence. The picture that emerged was very different from the original account.

The attack happened in two separate stages, in different places, and much of it took place where few people could see it. Many of the neighbours who were later counted among the "witnesses" heard a noise in the night, perhaps a scream, without seeing what was happening. In a city where late-night shouting was not unusual, many assumed it was a drunken argument. At least one person did shout at the attacker from a window, which briefly drove him away. Some accounts indicate that calls were made to the police. And one neighbour, Sophia Farrar, went out to help and held Genovese in her arms as she was dying.

In 2007, the psychologists Rachel Manning, Mark Levine and Alan Collins published a paper in the journal American Psychologist examining how the case had been presented in textbooks. They concluded that the evidence did not support the story of 38 people watching and doing nothing, and argued that the "parable of the 38 witnesses" had distorted how the psychology of helping was taught. Genovese's brother, Bill, later made a documentary, The Witness, released in 2015, tracing his own search for what really happened to his sister. When Moseley died in 2016, The New York Times itself acknowledged in his obituary that its original account had been flawed.

None of this makes the murder less terrible. It does mean that one of the most famous stories about human indifference was, in large part, a myth.

The experiments the story inspired

The original story did have one lasting and valuable effect. It inspired two young social psychologists, John Darley and Bibb Latané, to investigate how people behave in emergencies. Their experiments, published in the late 1960s, became classics of the field.

In one study, participants talked to other students over an intercom, supposedly about university life. Partway through, one of the voices appeared to have a seizure. The real variable was how many other people participants believed were listening. When people thought they were the only one who could hear, most quickly went to help. When they believed several others were also listening, they were slower to act, and some never did.

In another experiment, participants filled in questionnaires in a room that slowly began to fill with smoke. Those sitting alone usually reported it quickly. Those sitting with others who calmly ignored the smoke, because they had been instructed to, often stayed put, coughing and waving smoke away, rather than raising the alarm.

Darley and Latané identified two forces at work. The first is diffusion of responsibility: when many people are present, each feels less personally obliged to act. Someone else will surely do it. The second is social influence: in an ambiguous situation, we look at others to decide whether something is really an emergency. If everyone else seems calm, we conclude it must be fine.

This became known as the bystander effect, and it is a real, well-replicated finding. But it describes something specific. In the lab, the presence of others makes any given individual less likely to help. That is not the same as saying people in groups don't help at all.

What the cameras show

For a long time, it was hard to test what happens in real emergencies. Researchers couldn't stage assaults in the street, and eyewitness accounts are unreliable. Then CCTV gave them a new source of evidence.

In 2020, a team led by Richard Philpot published a study in American Psychologist that analysed footage of 219 real public conflicts captured by security cameras in Amsterdam, Lancaster in England and Cape Town in South Africa. The incidents ranged from heated arguments to physical violence.

The result was striking. In around nine out of ten incidents, at least one bystander intervened: stepping between people, calming someone down, pulling someone away or otherwise trying to help. The rate was remarkably similar across the three cities, despite their very different levels of crime and safety.

Even more interesting was the effect of group size. The more bystanders were present, the more likely it was that someone would step in. This doesn't contradict the original laboratory findings. Each individual in a big crowd might be a little less likely to act, but with more people present, the chance that at least one of them will help goes up.

Why the myth was so persuasive

If the evidence for widespread indifference was so weak, why did the story of the 38 witnesses take hold so firmly?

Part of the answer is timing. In the 1960s, many people were anxious about city life, anonymity and the decline of close-knit communities. A story that seemed to confirm those fears felt true, even if the details were shaky. It was also simple and memorable, a perfect teaching example. And once a story is in the textbooks, it tends to be repeated rather than checked.

There is a cost to that. If people believe that strangers never help, they may be less likely to ask for help themselves, or to expect it from others. A cynical story about human nature can, in a small way, make itself come true.

How to get help when you need it

The research does offer practical lessons, rooted in the real psychology of diffusion of responsibility and ambiguity.

Make it unambiguous. If you are in trouble, say clearly that this is an emergency. Bystanders often hesitate because they are unsure what is happening.

Pick one person. Instead of shouting "Somebody call an ambulance!", point at a specific person: "You, in the blue jacket, call an ambulance." That turns a diffuse responsibility into a direct request, and most people respond.

If you are a bystander, be the first. Once one person acts, others often follow. Even a small step, such as asking "Are you okay?", can change the whole situation.

A more hopeful story

The legacy of Kitty Genovese is complicated. Her death was a real tragedy, and the way it was reported has shaped how millions of people think about each other. The famous version of the story told us that strangers will look away. The evidence, from the neighbour who held her as she died to the security cameras of three cities, tells us something different.

People hesitate. They misread situations. They wait to see what others do. But when it becomes clear that someone needs help, most of the time, somebody steps in. That is the story worth teaching.