Sam Wineburg and Sarah McGrew ran a study in 2017 that compared three groups evaluating the credibility of unfamiliar websites: Stanford undergraduates, historians with doctorates, and professional fact-checkers.
The undergraduates and the historians did badly. The fact-checkers did well, and — the finding that matters — they did it faster.
The difference was not knowledge or intelligence. It was direction.
Students and academics read vertically. They stayed on the page and evaluated it: the design, the domain name, the About page, the references, the tone, the presence of citations. This is exactly what a well-built deceptive site optimizes for, because every one of those signals is under the site's own control.
Fact-checkers read laterally. Within seconds of arriving, they opened new tabs and left. They searched for the organization's name, looked for what other sources said about it, checked whether the claim appeared elsewhere, and established the source's reputation before reading its content.
Mike Caulfield's SIFT method operationalizes this into four moves that can be taught in an hour.
Stop. Notice that you are about to react, and check whether you know the source. Emotional reaction is the trigger, not the exception — a piece of content that produces a strong response is the one to slow down on, because that is what it was selected for.
Investigate the source. Leave the page. Who publishes this, who funds it, what is their track record. Thirty seconds.
Find better coverage. Rather than evaluating this article about the claim, look for the best available reporting on the claim itself. You are trying to find out whether the thing is true, not whether this page is trustworthy.
Trace claims to the original. Follow the quote, the study, the statistic back to its source. A remarkable proportion of viral claims dissolve at this step: the study says something different, the quote is truncated, the number is from a different year, the original was a preprint that was never published.
The general principle is that credibility is not a property of a page. It is a property of a source's relationship to the rest of the record, and that relationship cannot be assessed from inside the page.
Which is chapter 298's counter: laundering works by severing provenance, and tracing to origin is the only thing that reattaches it.
This counters Law 28.