Korva is a scroll of the internet’s best ideas.
Unreliable sources need a second source
When a convicted fraudster tells you the sky is blue, you do not conclude the sky is blue. You look for witnesses and documents that confirm it, and you use what survives. Investigators followed that method with Joe Sandberg, the Aspiration co-founder who was sentenced to 14 years in federal prison.
They discounted everything Sandberg told them unless other witnesses or records backed it up. Then the backing arrived: what he described about the Clippers' business president, Gillian Zucker, matched what the other companies' deals showed. She had followed the same steps with four companies, and the chance of a liar inventing a pattern that repeats exactly three more times is slim.
Journalists run the same test: a claim goes ahead once a second source supports it, and a claim without one waits.
Why does corroboration matter for testimony in court?
Many legal systems treat testimony from someone who took part in a wrongdoing with extra care, because that person may gain from shifting blame or earning leniency. Some jurisdictions require outside evidence to back such testimony before a conviction. Federal courts in the United States allow a conviction on it, but judges usually instruct juries to weigh it carefully. Documents, records, and independent witnesses are the usual forms of corroboration.
Where does the journalist two-source rule come from?
The practice is most closely linked to the Watergate reporting at The Washington Post in the early 1970s. Editors there pushed reporters Bob Woodward and Carl Bernstein to confirm key claims with more than one source before publishing. Many newsrooms adopted similar standards afterward, though the exact rule varies by outlet. A claim from a single source is generally held back or labeled accordingly.
Why does a repeating pattern build confidence?
Each independent confirmation makes it less likely that the original claim was invented. Statisticians describe this updating of belief with Bayes' theorem, named for the eighteenth-century mathematician Thomas Bayes. The key condition is independence: records and witnesses that do not share a source count for far more than ones that repeat each other. A pattern matched by separate documents is hard to produce by accident.
More from Pablo Torre Finds Out
