Words that remain.

A collection, not an oracle.

The archive currently preserves 523,390 distinct passages and proverbial phrases, across 68297 named author sections. The collection is heavily weighted toward English-language literature and older English translations. It is a beginning, not a representative survey of every culture.

Historical anthologies contain uncertain attributions, dated language, and disputed dates. A source citation verifies where a passage was printed; it does not independently verify authorship. Entries are labelled “attributed” or “traditional”, and never presented as newly verified scholarship.

The sources.

John Bartlett, Familiar Quotations, ninth edition (1905 printing): 8,872 distinct passages. Attribution and work references follow the author sections of the anthology. Footnotes and indexes are excluded to prevent accidental misattribution.

Thomas Preston, A Dictionary of English Proverbs and Proverbial Phrases: 1,821 distinct proverbial phrases. These are traditional sayings compiled by Preston; they are not attributed to him as their author.

Louis Klopsch, Many Thoughts of Many Minds (1896): 1,686 additional passages after deduplication. Complete prose and historical attribution are retained. Unresolved personal identities are explicitly labelled; unsupported labels and potential overlapping excerpts are held for review. Klopsch is the compiler, not the quoted author.

James Wood, Dictionary of Quotations (1893): 12,427 additional English passages, with printed citations. Selected source-labelled proverbs include editorial meanings; other proverbs, translations and unsupported citations remain held for review. Unresolved personal identities remain labelled; Wood is the compiler. Bible- and Hitopadesa-labelled extracts retain his printed work labels; their underlying editions and translations are not independently identified.

W. A. Clouston, Book of Wise Sayings (1893): 485 additional selected extracts. Printed labels may identify a person or a work; unresolved attribution and unspecified translations remain explicit. The anthology’s scope does not establish each saying’s cultural origin.

J. Purver Richardson, Life and Literature (copyright 1910): 398 additional numbered prose and verse extracts after deduplication. Printed personal citations are retained; unresolved identities remain explicit. Richardson is the compiler. The edition’s copyright year does not date the quotations.

J. T. Hackett, My Commonplace Book (third English edition, 1921): 580 additional prose and verse extracts with printed citations. Editorial notes are excluded from quotation text; unresolved identities remain labelled. The edition year does not date the quotations.

H. G. Adams, A Cyclopædia of Sacred Poetical Quotations: 1,972 additional cited verse extracts. The catalogue dates this new edition to 1888; the printed title page is undated. Preface, indexes and scripture heading epigraphs are excluded. Printed subjects are source labels; unresolved identities remain explicit. Adams is the compiler, and the edition year does not date the quotations.

Maturin M. Ballou, Pearls of Thought (1881): 2,244 additional cited extracts. The printed edition is dated 1881, with copyright 1880. Preface, dedication and boilerplate are excluded. Printed subject headings remain source labels; unresolved identities are explicit. Ballou is the compiler, and the edition year does not date the words.

Hoyt’s New Cyclopedia of Practical Quotations (1922), revised by Kate Louise Roberts: 929 additional passages from a reviewed range of scan pages. Credit: Wikisource contributors; each entry links its source revision and contributor history. Historical text is public domain in the United States. Transcription contributions and our normalized transcription extracts are shared under CC BY-SA 4.0. Whitespace and formatting are normalized; annotations are excluded. Unfinished pages, unsupported markup and paired translations remain held. Anthology attribution is not independent authorship verification.

George Whitefield Powers, Handy Dictionary of Poetical Quotations (1901): 756 additional numbered excerpts from the New York, Thomas Y. Crowell edition. Digitization credit: Audrey Longhurst and the PG Online Distributed Proofreading Team. Preface, indexes, uncertain boundaries and overlapping excerpts are excluded. Powers is the compiler; printed bylines remain anthology attributions, with unresolved identities labelled. The edition year does not date the quotations.

Henry G. Bohn, A Polyglot of Foreign Proverbs (1857): 346 selected proverbs with editorial meanings, printed French wording and English translations. Language-section labels do not establish origins. Bohn is the compiler; individual translators and composition dates remain unspecified. Digitization credit: Giovanni Fini, Chris Curnow, Lindy Walsh and the Online Distributed Proofreading Team. Unreviewed or broken pairs remain held.

Goethe, Maxims and Reflections (1906): 577 complete numbered quotations in T. Bailey Saunders’s English translation, from the New York Macmillan edition. Digitization credit: Christine Bell and Marc D’Hooghe. Preface, index, unnumbered essay, defective anchors and overlapping excerpts are excluded. Timeline placement uses the author’s documented lifetime midpoint, not a quotation composition date.

Ambrose Bierce, Collected Works, Volume 08, Epigrams (1911): 182 complete prose epigrams. These historical satirical passages include irony and dated viewpoints. Verse and uncertain dialogue boundaries remain held. Credit: Paul Hollander, Govert Schipper and the Online Distributed Proofreading Team. Timeline placement uses the recorded birth year, not a quotation date; the catalogue’s uncertain death year is unspecified.

Egyptian Instructions, Gunn translation (1906): 30 complete numbered passages. Ptahhotep is a traditional literary attribution; the instruction for Kagemni has an anonymous author. Credit: Al Haines. Doubtful translations, gaps and oversized passages remain held. Timeline placement uses the Papyrus Prisse manuscript, circa 1800 BCE, as dated by the BnF, rather than a composition date or author lifetime.

NASA, Artemis II mission communications (2026): 7 complete speaker turns by Victor Glover, Christina Koch and Reid Wiseman, reproduced in the April 8 transcript. The recorded speech year places these in the timeline. Source transcription is preserved, including repetitions. Only on-duty US government astronaut speech is included; third-party music, narration and other speakers are excluded. NASA is credited as the source; no endorsement is implied.

Friedrich Nietzsche, Human, All-Too-Human: Part I (1909) and Part II (1913): 1,218 complete numbered quotations. Translations: Helen Zimmern and Paul V. Cohn, respectively. Part I digitization: Marc D’Hooghe, with images supplied by Hathi Trust. Notes, prefaces, verse, unnumbered dialogue, oversized passages and overlapping extracts remain excluded. These historical philosophical arguments include dated viewpoints; read them in context. Timeline placement uses the author lifetime midpoint, not a quotation composition date.

La Rochefoucauld, Reflections; or Sentences and Moral Maxims (1871): 456 complete numbered author passages. J. W. Willis Bund and J. Hain Friswell translation; digitization by An Anonymous Volunteer and David Widger. Translator commentary, supplements and corrupt numbering remain outside the admitted passages.

The Analects of Confucius: 206 complete Master-speech chapters; Tao Teh King: 81 complete chapters. Both use James Legge’s historical English translations. Traditional attribution is retained; other Analects speakers and dialogues are held. Printed publication years are not supplied in these digitized texts. Tao digitization credits Gregory Walker and David Widger. Laozi’s identity and dates remain uncertain; no lifetime is invented.

Leonardo da Vinci, Thoughts on Art and Life (1906): 294 complete notebook selections. Maurice Baring translation, Lewis Einstein editor; digitization by Al Haines. This is a posthumous selection, with letter, ledger and damaged image-initial units held.

Nietzsche’s Dawn of Day (1911): 476 complete aphorisms, John McFarland Kennedy translation; Joyful Wisdom (1910): 282 complete aphorisms. Joyful Wisdom credits Thomas Common for prose and Paul V. Cohn and Maude D. Petre for poetry; digitization by Marc D’Hooghe, with Hathi Trust source images. Notes, oversize units and corrupt numbering remain held. Lifetime coordinates do not date individual passages.

Baltasar Gracián, The Art of Worldly Wisdom (1892 English edition): 282 complete maxims in Joseph Jacobs’s translation. Digitization by John Bruno Hare, June 2005, at the Internet Sacred Text Archive. The title page declares the historical book public domain in the United States; modern website introductions, images and editorial translations are excluded. Malformed numbering, annotations and overlapping excerpts remain held.

Thoughts: Selected from the Writings of Favorite Authors (1901): 325 complete signed prose and verse selections. Compiled by Jessie K. Freeman and Sarah S. B. Yule for the Fabiola Hospital Association; digitization by Juliet Sutherland, Krista Zaleski and the Online Distributed Proofreading Team. Unsigned selections, mixed boundaries and uncertain bylines remain excluded. Anthology attribution is preserved without independent authorship verification.

The Optimist’s Good Morning (1911 printing; copyright 1907): 315 complete signed literary selections. Compiled by Florence Hobart Perin; digitization by Larry B. Harrison, Julia Neufeld and the Online Distributed Proofreading Team, from Google Books Library Project page images. The full book pairs readings with prayers; those surrounding prayers remain outside the imported quotations. Edition years do not date the individual selections.

Rights and acknowledgements.

The historical texts are catalogued as public domain in the United States. Rights can vary by jurisdiction and translation. The normalized archive contains historical text without ebook branding or boilerplate. Project Gutenberg is acknowledged here as the digitization source, under its licence and trademark policy. Original files are available through the source links above.

Thanks to Melissa Er-Raqabi, Aldarondo, the Online Distributed Proofreading Team, the booksmiths at eBookForge, Delphine Lettau, and Lisa Reigel, Marilynda Fraser-Cunliffe, Diane Monico, Chuck Greif, Bill Tozier and Barbara Tozier and Julia Miller, Richard Hulse, Karin Spence, Juliet Sutherland for the source transcriptions.

Reading the coordinates.

The timeline uses a recorded lifespan midpoint, source circa date, or recorded birth year as an approximate historical coordinate. It is not the date a quotation was composed. Undated texts and proverbs remain undated. The “5000 BC — now” horizon describes the long-term editorial ambition of the archive.

Topics combine controlled, transparent word matching with separately labelled, agent-reviewed editorial assignments for explained entries. A mention of “love” or “time” is a discovery aid, not a scholarly claim about a passage’s meaning or philosophical school.

Contributor-provided material

Supplied PDF and CSV collections are reused with permission confirmed by their contributor. This is a permission basis, not a public-domain declaration. Book, page and CSV row references are retained internally. The CSV’s upstream distribution URL and specific licence instrument were not supplied; its reuse basis is the contributor’s confirmation. Printed author and country attributions remain unverified collection claims; a country label does not establish the earliest origin of a proverb. Collection explanations are labelled separately from reviewed Edsverb interpretations.

Keep wisdom travelling.

Contributions should start with a legitimate source, a clear redistribution licence, and reliable attribution. The repository documents a reproducible import pipeline and validation rules. Additional language collections, source corrections, and carefully documented context will help this library grow.

Your data stays yours.

Saved words are stored locally in your browser. There is no login, paid API, or subscription. Existing Vercel Analytics runs in production; the archive adds no other analytics service.