The reason a rumour gets shorter and more certain

It begins as someone heard that a supplier might be under review, possibly about invoicing, and nobody is sure. Six retellings later it is the supplier is being investigated for fraud, and every teller has a source.

What everyone sees

Distortion is attributed to dishonesty — someone must have exaggerated deliberately. Almost nobody does. Transmission itself imposes a filter: what is retained is what is memorable and tellable, and hedges are neither, so a chain of honest tellers produces a sharpened claim without any single act of exaggeration.

What is actually happening

Allport and Postman described serial transmission producing three regular effects — levelling, in which detail is discarded, sharpening, in which a few elements are retained and emphasised, and assimilation, in which the remainder is adjusted to fit the teller’s expectations. Their experimental method has been criticised as artificial, and later work qualifies the generality of the pattern. DiFonzo and Bordia’s review of rumour in organisations grounds it better: rumour circulates as collective sense-making under uncertainty, and its content stabilises around whatever explanation resolves the ambiguity for the group, which is why plausible causal completeness outcompetes accuracy in what survives.

Why it stays hidden

The sharpening hides because each teller reports honestly and the loss happens between tellings. Nobody experiences having dropped a qualifier; they experience remembering the important part. The chain leaves no record, so the final version arrives with no way to trace the seven small omissions that produced it.