Explorer · candidates

Candidate echoes

All 35,794 short verbatim matches the strict catalogue filtered out. Most are coincidence. Some are real.

Read with judgment. These are 4- and 5-word phrases that appear in both Shakespeare and a Founder document and include at least one content word that’s distinctive to Shakespeare. The strict catalogue requires 7+ words. At this corpus size, most short matches between any two large bodies of English are coincidental — common phrases, stock idioms, biblical resonances. Some of what’s below is real Shakespearean echo; some is noise. The view is here so you can see what’s in the middle tier and decide for yourself.

The list is ranked by a simple quality score: longer matches first, ties broken by how many distinctive content words the match contains. The full 35,794 are loaded into a Postgres backend with full-text search; filters and queries hit it live, so you can browse the whole set rather than a curated slice.

Each candidate carries a tier label that ranks the quality of the match, not a verdict on whether it is a real echo: Strong candidate (5-word matches with three or more unique distinctive Shakespeare words), Medium candidate (5-word matches with two unique distinctive words, or 4-word matches with three or more), and Low candidate for the rest. Backend rows still carry the underlying tier keys (HIGH / MEDIUM / LOW) for filter URLs and stable joins. A four- or five-word match is short enough that even a top-tier hit is at best likely, not confirmed; read the KWIC and judge.

Some genuinely Shakespearean phrases recovered here that don’t appear in the main catalogue: full of sound and fury (Macbeth, Adams 1758 and 1813), a pound of flesh (Merchant, Jefferson 1790), farewell the neighing steed (Othello, Adams 1758).

Filter by Founder, by play, or search the text directly. Each card links to the source document on Founders Online and to the relevant Shakespeare play at the Folger.

A note on the data source. When the live Supabase backend is available, this page queries all 35,794 candidate echoes directly. When the backend is unreachable (or when the site is being viewed from a static export with the env vars unset), the page falls back to a bundled JSON of the top 5,000 echoes by quality score. The header line above the filter chips reflects whichever source is currently live.

Searching all 35,794 candidate echoes via the live Supabase backend.

Founder

Confidence

Play (top 12)

Showing 0 of 0 matching candidate echoes.

No candidate echoes match these filters.

For the stricter HIGH/MEDIUM-confidence catalogue, see the Reference Catalogue. For thematic invocations of Shakespearean characters that aren’t direct quotes, see Thematic Allusions.

About this data

Full corpus analysis. Candidate echoes are 4- or 5-word verbatim matches between the Founders' corpus and Shakespeare's complete works, requiring at least one distinctive Shakespeare content word. All 35,794 are served live from a Postgres backend (Supabase) with full-text search, so filters, pagination, and chip counts narrow to the active query rather than reading from a static slice. Most short matches between any two large English corpora are coincidental; this view treats the results as candidates rather than confirmed findings.

Every word of both corpora is included in the count behind this view.

Underlying file: Supabase candidate_echoes (35,794 rows) + the precomputed scripts/candidate_echoes.py output

For the full methodology see the Methods essay or the About this project page.