About
The complete Latin Vulgate, with every verse also given in Latino sine flexione and in English.
The editions
The Latin is the Clementine Vulgate of 1592: 73 books, 35,809 verses.
The English is the Douay‑Rheims (Challoner, 1899), translated out of the Vulgate itself rather than the Hebrew and Greek, so it follows the Latin verse for verse. It covers all but nine of them, where the two editions divide a verse differently. Both texts are public domain.
Latino sine flexione
In 1903 Giuseppe Peano proposed that Latin becomes readable to any educated European if you throw away the inflections and give their work to prepositions and word order. He took the invariable forms from the language itself:
- Nouns and adjectives become the ablative singular — rex, regis → rege
- Verbs become the infinitive minus -re — legere → lege
- Genitive takes de, dative takes ad; the accusative is marked by position alone
- Plurals take -s
A program applies those rules to the Latin, and running it again gives the same bytes. Nothing here was translated by hand or by a language model.
Where it fails
Read the LSF as an aid to the Latin above it, not as a translation that stands on its own.
- Tense is unmarked, by Peano’s own preference. The LSF says crea where the Latin says creavit.
- The tagger is about 97% accurate here, and less so on Hebrew proper nouns. A wrong part of speech gives a wrong form.
- The program reorders a clause only where the parse is unambiguous. Leaving it in Latin order beats reordering it wrongly.
- Nobody has rendered a Bible in LSF before, so there is no gold standard beyond Peano’s own examples and the Pater Noster.
Sources
- Clementine Vulgate & Douay‑RheimsBoth texts, as used here.
- Peano, Vocabulario Commune (1915)His own vocabulary, and the attested forms the lexicon prefers.
- Manuale Practico de Interlingua (1913)Peano’s grammar of the language, written in it.
- 100 exemplo de InterlinguaPeano’s worked Latin‑to‑LSF sentences.
- XXIIVV — Latin sine flexioneWord lists (lsf_dict, lo_dict) checked against the derived forms.
- PROIEL treebankA hand-annotated Vulgate, and the tagger’s training data.
- StanzaThe tagger.
- Whitaker’s WORDSA morphological dictionary, for lemmas the corpus and Peano both miss.