Gell-Mann Amnesia
A cognitive bias whereby a reader notices that media coverage of a subject they know well is riddled with errors, then turns the page and trusts the same publication’s reporting on subjects they don’t know well — as if the demonstrated unreliability simply didn’t apply elsewhere.
Origin
Coined by Michael Crichton in a 2002 speech (“Why Speculate?”), named after physicist Murray Gell-Mann, who Crichton said found the phenomenon amusing. Crichton’s own example was show-business reporting: he’d read articles about subjects he knew firsthand, find them wrong in every particular, then read the next article — on Palestine, say — and somehow believe it.
The Pattern
“You open the newspaper to an article on some subject you know well… You read the article and see the journalist has absolutely no understanding of either the facts or the issues. Often, the article is so wrong it actually presents the story backward… In any case, you read with exasperation or amusement the multiple errors in a story, and then turn the page to national or international affairs, and read as if the rest of the newspaper was somehow more accurate… You believe stories written by men who don’t know what they’re talking about.”
- Crichton
Why It Persists
- Asymmetric verification cost: checking expertise-domain claims is cheap (you already know); checking unfamiliar-domain claims is expensive.
- Source-level trust transfer: readers attach credibility to the masthead, not to the individual article or reporter.
- Forgetting is the default: each article is encountered fresh; the prior evidence of unreliability doesn’t get carried forward.
Implications
- Media credibility cannot be safely generalised across domains from a single competent article - nor condemned across domains from a single bad one. Both moves are the same error in opposite directions.
- The bias compounds with audience expertise: experts in any field have direct evidence of media failure in their field, but no easy way to know if other fields fare better or worse.
- A weak mitigation: track which outlets are reliable in domains you can verify, and extend trust cautiously - but editorial quality varies by desk, beat, and reporter, not by publication.