Search the record

One index over everything published here: the claims and the evidence under them, every census row with its accession number and watermark, the open questions, the provenance events and the gaps between them, the glossary, the teaching modules, the field notes, the bibliography, the policies and every logged correction.

124 indexed recordsRuns entirely in your browserNo query logging

Search the record

124 records are indexed and searched in the browser — nothing you type is sent anywhere. An exact phrase ranks above a page that merely contains all your words, which ranks above one that contains some of them.

What is searchable

Everything on the record, including the parts that are unresolved. The index is built from the same data files the pages are built from, so a claim that exists is findable and a claim that does not exist cannot be found.

Claims on the record18: Every assertion the record makes, with its field, its value and its note.
Impression census27: Institution, accession number, catalogued state, watermark and paper notes.
Open research questions12: What is unresolved, and why it matters.
Provenance events6: Documented events and the gaps between them.
Glossary28: Terms of art, defined once and linked from everywhere.
Teaching modules6: The six modules, indexed on their full text.
Field notes2: Field notes on findings.
Bibliography13: Works cited, verified and unverified.
Corrections and changes5: What changed, and the reason logged against it.
Policies7: How the record is governed.

Try one of these

Search is deliberately literal: no fuzzy matching, no stemming, no synonym expansion and no ranking signal beyond where your words appear. A record that does not contain your words does not appear, which makes an empty result a fact about the record rather than about the search. If you would rather query the data directly, the whole corpus is available through the read-only public API and as downloadable JSON and CSV.