For more than half a century, introductory psychology courses have opened their unit on helping behaviour with the same story: in Queens, New York, in 1964, thirty-eight people watched a woman get stabbed to death and none of them did anything. The story is real in the sense that a woman was murdered and that a newspaper printed those numbers. Almost everything else about how it’s usually told has turned out to be wrong.
What the original story actually claimed
Catherine “Kitty” Genovese, a 28-year-old bar manager, was attacked and stabbed by Winston Moseley outside her apartment building in Kew Gardens in the early hours of 13 March 1964, in two separate assaults about ten minutes apart, in an episode that spanned roughly half an hour from the first attack to her death; she was raped during the second, fatal attack. Two weeks later, New York Times reporter Martin Gansberg published a story under the headline “37 Who Saw Murder Didn’t Call the Police,” though the article’s own opening line gave a different number: “38 respectable, law-abiding citizens in Queens watched a killer stalk and stab a woman” — for more than half an hour, the article said, in three separate attacks.
That inconsistency, 37 in the headline, 38 in the text, has sat unremarked in the historical record for decades; it’s the 38 figure that stuck, becoming the number cited in psychology textbooks and by the researchers who built a field around it.
Social psychologists Bibb Latané and John Darley read the Gansberg account and used it as the motivating case for what became the bystander effect: the finding, developed through their own separate experiments beginning in 1968, that people are less likely to help in an emergency when they believe other bystanders are present, because responsibility for acting feels diffused across the group. Their experimental work on diffusion of responsibility is real, replicated, and remains a legitimate part of social psychology. The Genovese case was the anecdote that opened the door to it, not itself a piece of that experimental evidence.
What later investigation actually found
The most thorough published reassessment is Rachel Manning, Mark Levine and Alan Collins’s 2007 paper in American Psychologist, “The Kitty Genovese murder and the social psychology of helping: The parable of the 38 witnesses.” Working from trial records, police files and contemporary reporting, they concluded there is no evidence that 38 people watched the murder, no evidence that those who were aware of it saw the attack itself, and no evidence that witness inaction caused her death. They frame the story explicitly as a parable, something taught for its moral clarity rather than for its accuracy, and note it had been repeated uncritically in psychology curricula for decades.
Journalist Kevin Cook went further in his 2014 book on the case, working from police files, trial transcripts and interviews with people involved. His account puts the actual police tally at 49 people who reported seeing or hearing something that night, of whom perhaps a dozen or so could reasonably be called eyewitnesses to any part of the events, a figure that varies somewhat depending on which secondary account is consulted, but is consistently far below “38 witnesses to a murder.” The “38” figure itself appears to have originated with a police tally passed to Commissioner Michael Murphy, who relayed it to Times editor A.M. Rosenthal.
The detail that undoes the story most completely, though, is architectural rather than statistical. The fatal second attack took place in an interior vestibule, out of view from the street and from most surrounding apartments. Investigators found no one who witnessed the entire attack from start to finish, because the geometry of the building made that impossible regardless of how many people were awake or watching from a window during the earlier, more visible portion of the assault.
What actually happened, according to the record
People did act, contrary to the story’s central image of collective paralysis. A neighbour, Robert Mozer, shouted from his window during the first attack and briefly drove Moseley off. A teenager who heard the commotion alerted his father, who called the police. Karl Ross called police later, though not in time. Sophia Farrar, a friend of Genovese’s who lived in the building, went down to her and stayed with her, holding her, before help arrived. The story that became the field’s founding case study left all of this out.
The bystander effect itself doesn’t need Kitty Genovese’s murder to be true; the diffusion-of-responsibility experiments Latané and Darley designed afterward stand on their own evidence, replicated many times since in controlled settings that have nothing to do with the original anecdote. The Genovese story mattered to that research programme mainly as a prompt, the question it made Latané and Darley want to test, not as data itself.
The story had consequences beyond psychology textbooks, too, some of them genuinely constructive. It’s frequently credited, alongside other factors, with helping build public and political support for a unified emergency telephone number in the United States, the system that became 911, on the logic that easier reporting might counter exactly the kind of hesitation the Genovese story seemed to describe. It’s a strange legacy for an account this inaccurate to have: real infrastructure and real research built partly on a version of events that didn’t happen the way it was told, even while the underlying human tendency the story was used to illustrate turned out to be genuine and well documented on its own separate terms.
What the case actually demonstrates, once the record is checked against the story, is a different and arguably more useful lesson about how a specific, memorable number in a newspaper headline can travel for sixty years largely unexamined, cited by people, including researchers, who had every professional reason to check it and mostly didn’t. A. M. Rosenthal, the Times editor who ran the original story, went on to write his own book the same year, Thirty-Eight Witnesses: The Kitty Genovese Case, which helped cement the 38-witness framing in the public imagination even further, at a moment when the more careful reconstruction that Manning, Levine and Collins would eventually publish was still four decades away.