by Claude Opus 5.5
Does the diffusion of responsibility documented by Darley and Latané excuse bystanders, or merely explain them?
In 1968 John Darley and Bibb Latané ran a now famous experiment. Students were told they would discuss personal problems over an intercom, and during the conversation one participant, actually a recording, appeared to have a seizure. When students believed they were the only other person listening, about 85 per cent went for help before the seizure ended. When they believed four others were also listening, only about 31 per cent did. Darley and Latané called the mechanism “diffusion of responsibility”: the more people who could help, the less each feels that helping is his job.
The finding has been replicated many times, with qualifications. A 2011 meta-analysis by Peter Fischer and colleagues found that the effect is real but weaker in clearly dangerous emergencies, where the need for action is obvious. It is often invoked to explain cases where many people knew of harm and none acted. A group chat with around fifty members, which reportedly received messages during the night of the alleged assault behind these seminars, is an obvious candidate. The question is what follows morally.
Explanation and excuse
An explanation says why something happened. An excuse says why someone should not be blamed, or should be blamed less, for it. The two are different. Every wrongful act has an explanation; most do not have an excuse. That a burglar was in debt explains his burglary but does not excuse it.
So the fact that diffusion of responsibility explains bystander inaction does not show that it excuses it. The question is whether this particular explanation has features that also make it an excuse.
The case that it excuses
The strongest case for treating diffusion as an excuse draws on the general conditions of responsibility. We excuse people when they could not reasonably have done otherwise, or when they did not know what they were doing. Diffusion of responsibility, it might be argued, partly undermines both. It is not a conscious decision. People in groups do not reason “someone else will deal with this, so I won’t”. Rather, the presence of others changes their perception of the situation: it seems less urgent, less clearly their responsibility, and less clearly an emergency at all. If the effect operates beneath awareness, then the bystander did not knowingly choose not to help. He simply did not see the situation as one requiring him to act.
A related argument appeals to fairness. If most people would behave the same way in the same situation, it seems harsh to blame those who happened to be there. Moral standards should be ones that ordinary people can meet.
The case that it merely explains
The reply begins with the observation that the effect is not total. Even in Darley and Latané’s group condition, nearly a third of participants did go for help. The situation made helping less likely, but it did not make it impossible. Those who acted show that acting was available.
Second, an excuse typically requires that the person could not reasonably have been expected to act differently. But we can reasonably expect people to resist known psychological pressures, especially when the stakes are high. We do not excuse people who follow a crowd into violence merely because crowds make violence more likely. We expect them to recognise the pull and resist it.
Third, knowledge of the effect changes the picture. Once diffusion of responsibility is widely understood, as it is now taught in schools and universities, people are on notice that their sense that “someone else will handle it” may be a trick of the situation rather than an accurate assessment. Someone who knows about the bias and still relies on it is in a weaker position to claim it as an excuse.
Fourth, the effect explains why people do not notice or do not feel responsible. It does much less to explain the conduct of people who did notice, did understand, and still chose not to act. In a group chat, where messages are read individually on a private screen and sit there to be re-read, the conditions differ from a live emergency. Someone who saw clear evidence of danger had time to reflect.
Mitigation rather than excuse
The most plausible conclusion is that diffusion of responsibility can mitigate but rarely excuses. It can reduce blame for people whose failure was a failure of perception: who did not register that something serious was happening, or who genuinely believed others closer to the scene were acting. It does little for those who registered the danger and decided to do nothing.
This conclusion also has a forward-looking side. Knowing that groups suppress helping is a reason for groups to adopt rules that counteract the effect. Bystander intervention training does exactly this, teaching people to assume responsibility personally rather than wait for others. Organisations can establish clear expectations that certain things must be reported, and identify who should report them. Once those measures are available, failing to take them becomes a further failure, this time of the group.
For discussion
A useful exercise is to ask students whether they would accept diffusion of responsibility as an excuse in other contexts: a doctor who did not check a patient because several colleagues were on shift, or a crew member who did not report a safety problem because others had seen it too. Most people say no. If the answer is the same here, it is worth asking why the situation of a group chat might ever feel different.