On March 27, 1964, The New York Times ran a front-page story under the headline “37 Who Saw Murder Didn’t Call the Police.” Two weeks earlier, Kitty Genovese had been killed in Kew Gardens, Queens. “For more than half an hour thirty-eight respectable, law-abiding citizens in Queens watched a killer stalk and stab a woman,” it began. Preachers, professors and commentators, Darley and Latané later wrote, blamed everything from “moral decay” to “dehumanization produced by the urban environment.”
Two psychologists, John Darley at New York University and Bibb Latané at Columbia, suspected something else. Taking the newspaper’s account as they found it, they reasoned that each witness, seeing lights and figures in other windows, knew others were watching, and that none could tell what the others were doing.
They built a test. A student arrived at the lab, was shown into a small room off a long corridor, and given a headset. The group would discuss the personal problems of college life over an intercom, the student was told, so that nobody had to face strangers, and the experimenter would not be listening. One microphone would be live at a time.
The first speaker said he was having trouble adjusting to New York. Very hesitantly, he added that he was prone to seizures. On his next turn his speech fell apart: “I’m gonna die-er-er-I’m . . . gonna die-er-help-er-er-seizure-er-.” Then he choked and went quiet. His voice was cut off 125 seconds after his turn began.
Every voice except the student’s own was a tape. The only thing that varied, in the main design, was how many other listeners the student believed were there: none, one or four.
Students who thought they were alone with him reported the emergency 85% of the time before he was cut off. With one other listener, 62% did; with four others, 31%. The paper’s table puts the response times at 52, 93 and 166 seconds. Everyone in the first group reported the seizure in the end. In the largest groups, 62% ever did. Nobody who had stayed put for three minutes came out later.
What the other listener was like barely mattered. Women who thought the other bystander was a man, a woman, or a premedical student who worked in an emergency ward responded about equally fast.
The students who never came out did not look apathetic. When the experimenter came in, they often asked, “Is he being taken care of?” Many had trembling hands and sweating palms. Asked afterward, students in the larger groups said the presence of the others had made no difference to them.
In 2007, Rachel Manning, Mark Levine and Alan Collins checked the Genovese story against trial transcripts and other records. They found no evidence for 38 witnesses, or that witnesses saw the murder, or that they did nothing. There were two attacks, not three. A neighbor’s shout scared the attacker away after the first. The fatal second attack happened in a stairwell that none of the witnesses who testified could see.
The newspaper’s 38 had not held up. In the lab, most of the students who believed four others were listening were still in their rooms when the tape ran out.