Search the Latin Josephus

Search syntax

bellum

A complete word, ignoring capitalization.

bell*, *que, *bell*

Zero or more characters within a word, at any position.

"sacerdos ex Hierosolymis"

Consecutive words in that order. Punctuation, capitalization and ordinary whitespace are ignored.

"sacerd* ex Hierosolym*"

Wildcards can also appear within a quoted phrase.

A AND B · A OR B · A AND NOT B

Combine or exclude matches. Spaces between unquoted terms mean AND.

(A OR B) AND C

Parentheses control grouping. Otherwise the order is NOT, then AND, then OR. Use uppercase operators.

Text and scope

Lauro 1544 Italian De excidio: all five books and the translated Prologue use the accepted corrected-199 diplomatic display and provisional DEH correspondence. Each physical Italian span counts once in source order. Italian folios and canonical correspondences have distinct CSV metadata. Long ſ matches s; only explicit ¬ continuations join. Accents, ß, u/v and i/j remain literal. Apostrophes and genuine hyphens separate words. The accepted appreſtoſ[ſi] can match apprestossi; brackets and the editorial-supply qualification remain in results and exports. Supplies are not certified as printed letters. Headings, running headers, marginalia, notes, signatures and other non-running matter remain separate. Exact phrases stop at Prologue and book boundaries; they never concatenate separate canonical targets or witnesses.

Millet 1556 French De excidio: all five books and the translated Prologue are searchable from the accepted semi-final r1 transcription and provisional alignment. French passages follow physical source order and appear once even with several DEH correspondences. Source folios and canonical correspondences are recorded separately in CSV exports. Long ſ matches s; only explicit ¬ continuations are rejoined. Apostrophes (straight or curly) and genuine hyphens separate words: l'histoire is the phrase "l histoire". Accents remain significant; no modernization, u/v or i/j folding applies to French. Headings, running headers, catchwords, signatures, marginalia, dedications and other printed paratext are excluded from this main-text layer. Exact phrases follow physical French order and stop at the end of the translated Prologue; canonical omissions alone do not create French gaps. Within-passage Boolean scope uses each contiguous accepted French span; separated correspondences are not concatenated into artificial prose.

Currently indexed: Cardwell’s Latin Bellum Judaicum, all seven books including the proem of Book I; Ussani’s Latin De excidio, all five books and the Prologue; Boysen’s Latin Contra Apionem, both books; and the Bamberg Msc. Class. 78 Latin Antiquities, Books I–XX with the separate Proem included in Book I. Whiston’s English Antiquities covers the Proem and Books I–XX, a total of 7,376 Niese sections. Whiston’s English Bellum Judaicum covers all seven books, including Book I’s embedded preface, a total of 4,001 Niese sections. Whiston’s English Contra Apionem covers both books, a total of 616 Niese sections. Pollard’s English De excidio, version 1.1, covers the Prologue and all five books, a total of 558 chapter-and-paragraph citation units. Explicitly marked titles, citation labels and annotation notes are excluded. Source additions and untagged letters within the accepted Latin passages remain searchable, including inline Roman chapter letters.

Every word in the indexed main text remains searchable, including common words such as the and and. Spelling is preserved in the excerpts; the optional u/v and i/j equivalence setting affects Latin matching only. English keeps its source spelling apart from capitalization. No grammatical stemming or automatic spelling correction is applied.

Within each citation unit: every Boolean condition and exact phrase is evaluated in that unit. Bellum, Contra Apionem and Antiquities use Niese sections; De excidio uses the reader’s book, chapter and paragraph citations, with Prol. for the Prologue.

Whole book: Boolean conditions may be satisfied anywhere in the same book; exact phrases must still be consecutive and may cross citation or chapter boundaries. Each book, work and edition is evaluated separately. A phrase never continues into another book, from the De excidio Prologue into its Book I narrative, or from the separate Antiquities Proem into its Book I narrative.

Passages: results follow the order of the text. A citation unit with several matches appears once in citation-unit scope. In whole-book scope, a phrase spanning units can have its own excerpt. Click a citation to open the corresponding passage in the reader.

Concordance: each positive word or phrase match has its own row, with up to ten words of context on each side. Original spelling is retained. A phrase spanning citation units appears in one row.

Ussani Latin De excidio editorial text: [brackets] mark supplied letters, struck text marks deletion, and † indicates textual uncertainty. Supplied and deleted letters remain searchable, including letters inside a word. A dagger is treated as punctuation, so phrase searches may match across it. The CSV retains exact text and offsets and adds editorial columns using [supplied] and ⟦deleted⟧.

Boysen Latin Contra Apionem markings: the source’s asterisks, brackets, angle brackets and Greek quotations are retained. Exact phrases stop at asterisk-marked interruptions, including those at section boundaries; Boolean conditions can relate words on either side. Bold green text marks an addition encoded in the source, shown as ⟨added text⟩ in the CSV’s editorial columns. The search makes no new textual repairs.

Bamberg Latin Antiquities markings and coverage: added, deleted and uncertain letters remain searchable, displayed as bold green text, struck text and dotted underlining. Original brackets, ellipses and erased-dot interruptions are preserved. Exact phrases stop at marked gaps, empty source intervals, the documented Book IX lacuna and the delimited interruptions between XVI.188 and 200 and between XX.26 and 37; Boolean conditions can relate words on either side. III.240 and VII.69–99 retain their citations and encoded omissions. X.108 has a citation label but no Latin before the next Niese marker. IX.51–109 are absent from the accepted source and have no generated records. Other identities marked unavailable in the accepted maps also have no generated records; a skipped number alone does not imply a physical gap. Split portions in Books XI and XVI share Boolean scope, while exact phrases follow physical source order. The two Bellum insertions in XI retain separate citations. The CSV preserves these distinctions, source markings and coverage metadata. Proem results link directly to the separate Proem reader.

Whiston English Antiquities: the Proem and Books I–XX retain their original English spelling, punctuation and Niese section divisions. Book IX includes sections 51–109 from the accepted English source. Headings, printed citation labels and footnotes are excluded. Exact phrases follow physical source order, including the displaced sequence in Book XI. In whole-book scope they may cross physically adjacent Niese sections, but stay within one book and never cross from the Proem into the narrative. English and Latin editions are searched separately, even when both are selected; Boolean conditions and phrases cannot combine evidence from different editions. Proem results link to the separate Proem reader with Whiston selected.

Whiston English Bellum Judaicum: all seven books retain their original English spelling, punctuation and Niese sections. Book I’s preface, sections 1–30, remains part of the continuous book; whole-book phrases may cross into section 31. The accepted supplied passage spanning VII.208–209 remains searchable and is shown in [brackets]. The CSV preserves the literal text, supplied-text annotations and source information. Exact phrases may cross adjacent sections in whole-book scope, but remain within one book and edition. Citations open the reader with Whiston selected.

Whiston English Contra Apionem: both books retain their original English spelling, punctuation, literal brackets and Niese section divisions. Whiston’s own division markers, including those inside II.109 and II.112, do not add searchable words or interrupt exact phrases. Whole-book phrases may cross adjacent Niese sections, but remain within one book and edition. Citations open the corresponding English passage with Whiston selected.

Pollard English De excidio, version 1.1: the Prologue and all five books retain their accepted English wording, spelling and punctuation. Citations follow the reader’s book, chapter and paragraph divisions, with Prol. for the Prologue. Headings, printed citation labels, release notes and translation-comparison annotations are excluded. Whole-book phrases may cross adjacent paragraphs and chapters, but never pass from the Prologue into the Book I narrative or between books. Citations open the corresponding English passage with Pollard v1.1 selected. English and Latin are searched as separate editions.

Counts: a single word, wildcard or quoted phrase has an occurrence count as well as a count of matching citation units or books. Boolean queries count matching units or books; concordance rows show their positive matches and are not word frequencies. A citation unit or book can satisfy a negative condition without a word to highlight.

Download all results: the CSV includes every concordance row and the citations of any matching sections or books without positive matches, regardless of which view or page is displayed. It records the completed query, its settings, the edition and links to the reader. Changes typed into the search box take effect after you press Search.