Imported Translations (off-the-shelf scriptures)

Three translation pipelines, three pages.

• i18n localisation — taking a Wheel of Heaven English page (wiki entry, Article, Dispatch, etc.) and rendering it into the nine other site languages.
• WoH Translation Program — producing a fresh WoH-lens English translation of a religious or mythological source-text (Hebrew, Greek, Akkadian, etc.) through the Translator → Editor → Reviewer pipeline.
• This page: imported translations — selecting which off-the-shelf public-domain translation to ship for each site language, for the religious texts where the WoH Translation Program output does not (yet) exist. The user sees the WoH translation when available, otherwise the imported translation, otherwise the source text.

Scope of this page#

We carry two kinds of translation under data-library/{book}/:

1. WoH translations (slug suffix -woh): produced by the WoH Translation Program. Render Elohim as Elohim (plural), preserve layered-witness sources, carry per-verse commentary keyed to a versioned glossary. These are the project’s flagship output.
2. Imported translations (no suffix): off-the-shelf English versions of the same books, used as a comparison baseline + as fallback for any book we haven’t translated yet under the WoH program. These are currently English-only — the rest of this page is the plan for extending them to the other eight site languages.

The Hebrew Bible is the focus because (a) it’s the canonical-Raëlian source corpus, (b) it has the highest WoH-translation coverage already, so the imports here primarily serve as comparison reading. The same methodology can extend to the NT and to the Qur’an later.

What we’re matching against#

The English baseline for the imported Hebrew Bible is the American Standard Version (ASV) 1901. We chose ASV because of its unusually literal philosophy — it consciously stays close to the Hebrew morphology, preserves the verb pattern, and resists doctrinal smoothing. For each non-English site language, the question is: which public-domain translation has the strongest analogous philosophy?

Two constraints fence the answer:

1. Public domain. The project ships under CC0-1.0. Anything still under copyright is out unless the rights-holder publishes under a CC0-compatible licence (rare for Bible translations).
2. Text-close, not paraphrase. Dynamic-equivalence translations (NIV-style, Good News, “Common Language” series in most languages) miss the point. We want formal-equivalence in every target language.

Per-language choices#

Per-language reasoning#

German — Elberfelder 1905#

The “alte Elberfelder” (literally Elberfeld old, named for the Wuppertal-Elberfeld translators) is Germany’s closest analogue to ASV. Originally produced by Brethren-movement scholars (1855 NT, 1871 OT) who wanted a literal alternative to Luther, the 1905 revision is the last public-domain edition before the 20th-c. revisions (1985 and 2003 revisions are still under copyright). Theological footprint is minimal, the verb-pattern is preserved, and the bara/asah/yatsar distinction in Genesis 1 is honoured.

Available digital text: Internet Archive multiple scans; community ePub on Project Wittenberg-style hosting.

Considered and rejected:

• Luther 1545 / Luther 1912: theologically committed, not text-close in the ASV sense. Heavy interpretive moves.
• Buber–Rosenzweig (1925–1961): brilliantly Hebraicizing but still copyrighted (Schocken/Lambert Schneider).
• Schlachter 2000: copyrighted (Genfer Bibelgesellschaft).

French — Crampon 1923 (baseline) + PD Dhorme (Job, Samuel)#

Canonical-Raëlian context. Raël has identified Édouard Dhorme’s Bible (Pléiade edition, Gallimard 1956–1959 for the Old Testament, 1971 for the NT) as the specific French text consulted during the December 1973 encounter. The natural project choice would therefore be Dhorme.

The copyright wall. Dhorme died in 1966. Under French copyright (70 years post-mortem of last surviving author), the Pléiade Bible’s collaborator deaths push expiry to the 2030s for Dhorme’s sole contributions and ≈ 2071 for the joint work (Guillaumont d. 2000 is the longest-surviving collaborator we’ve identified). The Pléiade is firmly under Gallimard copyright today and cannot be shipped in a CC0-1.0 repository.

The workaround. Dhorme published earlier individual commentaries with translations in Gabalda’s Études Bibliques series, before the Pléiade project. Two of these are firmly public-domain and digitally available:

• Les Livres de Samuel (1910) — Internet Archive, full PDF.
• Le Livre de Job (1929 second edition; the 1926 first edition is also PD) — Internet Archive, marked PDM 1.0, with full commentary apparatus.

These ship for Samuel + Job. For every other Hebrew Bible book, Crampon 1923 is the baseline. Crampon (Augustin Crampon, 1826–1894) translated from the original Hebrew + Greek; his work was completed posthumously by collaborators (full Bible 1904; revised 1923, the canonical PD edition). Catholic literal register, fully digitized on French Wikisource book-by-book, with the original scans + transcribed parallel.

The bibliographic acknowledgement. Every book page surfaces a footer note: “Raël’s December 1973 account references the Édouard Dhorme Bible (Pléiade edition, Gallimard 1956–1959). Dhorme’s Genesis–Deuteronomy + other books in the Pléiade remain under Gallimard copyright until ≈ 2071. This site ships the Crampon 1923 (public domain) as the canonical French baseline, with Dhorme’s pre-Pléiade PD individual works (Samuel 1910, Job 1929) used where they exist. Readers wishing to consult Dhorme’s Genesis can obtain the Pléiade volumes from Gallimard.”

Considered and rejected:

• Darby French 1885-88 (Pau-Vevey): even more ASV-like in literal philosophy; rejected only because the Raël-Dhorme canonical link takes priority for French. Worth keeping in mind if Crampon’s Catholic register proves a register-mismatch in practice.
• Louis Segond 1880/1910: more readable than Crampon, less literal.
• Bible de Jérusalem 1956: copyrighted (Cerf).

Spanish — Reina-Valera 1909#

The Reina-Valera (Casiodoro de Reina 1569, Cipriano de Valera 1602 revision) is the Spanish Protestant canon. The 1909 revision is the last unambiguously public-domain edition — the 1960 revision is copyrighted (United Bible Societies / Sociedad Bíblica), and the 1995/2009 revisions even more so. Literal philosophy in the classical-Spanish-Reformed mould.

Available: BibleGateway, eBible.org, multiple PD distributions.

Considered and rejected:

• Reina-Valera 1865 (“Valera 1865”): even more literal but extremely archaic register. Available but probably too hard for modern readers.
• Biblia de Jerusalén 1976: copyrighted (Desclée de Brouwer).

Russian — Synodal 1876#

The Russian Synodal Translation, completed in 1876, is the only serious public-domain option in Russian. Produced under the Holy Synod of the Russian Orthodox Church, it carries Orthodox theological vocabulary that flavors otherwise-literal renderings (Господь, Дух Святой, Предание, etc.). Mid-19th-century register is dated but readable.

The 2011 Russian Bible Society modern translation is significantly more accessible but copyrighted.

Chinese — Chinese Union Version (和合本) 1919#

The CUV (和合本, Hé hé běn “Union Version”) is the canonical Chinese Bible, produced by a multi-denominational committee 1890–1919. Two streams exist — the “Union Mandarin” (literal) and “Easy Wenli” (simpler classical) versions. The Union Mandarin is the one in universal use today and is the literal-philosophy choice for our purposes.

Same text serves both zh (Simplified Chinese) and zh-Hant (Traditional Chinese) — the underlying characters are the same; the two script variants differ only in glyph form, automatically convertible.

PD status: corporate work published in 1919 under collective committee authorship; expired in the early 1990s under the old 50-year rule and remains PD under current Chinese copyright law.

Japanese — 口語訳 1955 (Kougo-yaku)#

Why not 文語訳 1887? The Meiji Bungo-yaku (1887 OT + 1917 Taishō NT) is firmly public-domain but in Classical Japanese (文語). Modern Japanese readers find it as opaque as English readers find Tyndale. The literal-philosophy benefit is offset by the unreadability.

口語訳 1955 verdict. Japan Bible Society published the Colloquial Japanese Bible (口語訳聖書, Kougo-yaku Seisho) in 1954 (NT) + 1955 (OT). This was the first modern-language Japanese Bible, designed specifically to be readable in post-war colloquial Japanese while staying close to the Hebrew and Greek source-texts.

JBS’s own copyright statement declares that “its copyright protection period has ended”:

• Corporate/institutional-work rule (法人著作物): 50 years from publication
• 1955 + 50 = end of 2005, public-domain from 1 January 2006
• Japan’s 2018 retroactive copyright extension (50 → 70 years) did NOT apply to works whose protection had already expired before 30 December 2018

Two caveats that affect how we ship it:

1. Moral rights persist in Japan in perpetuity for the original authors. We can display the text verbatim, quote it, redistribute it — but we cannot modify it. No typo-fixes, no register-modernization, no editorial edits. Verbatim or nothing.
2. Use the original 1955 first-printing text, not the corrected editions. JBS publishes a Bible Phrase Correction List (聖書語句訂正一覧). The corrections themselves are still copyrighted, and any edition of 口語訳 that incorporates them is a JBS-licensed derivative. Source from a clean 1955-first-printing digitization (Japanese Wikisource is the canonical community-curated source).

Korean — 개역성경 1938#

Why not 개역한글 1961 or 개역개정 1998? Both are copyrighted by the Korean Bible Society and remain so. The 1961 revision is what most Korean readers know today, but it builds directly on 1938 and the 1938 text is independently available in the public domain.

개역성경 1938. The Joseon Bible Society’s 1938 Old + New Testament revision was the first major institutional revision of the 1900/1911 missionary Korean Bible. Copyright expired in 2012 per find.bible and other public-domain catalogues. Reasonably literal philosophy, early-20th-c. literary register comparable to early-20th-c. English Bibles. The most-recent public-domain Korean Bible available.

Considered and rejected:

• Old Revised 1911: even older PD option, but the 1938 revision is improved philologically and is the direct ancestor of all modern Korean Bibles.
• 공동번역 1977 / 새번역 / 우리말성경: all copyrighted.

Hebrew — Westminster Leningrad#

Hebrew (he) is special: it’s both a site language AND the source language of the Hebrew Bible. The translation-target collapses to the source text itself.

We use the Westminster Leningrad Codex digital edition of the Masoretic Text — already in data-library/{book}-woh/chapter-N.json as the per-paragraph text field. The library-book template displays this verbatim when detected_lang == "he" and applies RTL rendering.

No separate “Hebrew imported translation” exists; the source text is the rendering.

Schema implications#

Catalog metadata#

Existing English-only imports look like this in catalog.json:

{
  "slug": "genesis",
  "tradition": "hebrew-bible",
  "originalLang": "he",
  "primaryLang": "en",
  "availableLangs": ["en"],
  "translationNote": "American Standard Version 1901 (public domain)"
}

Per-language imports extend availableLangs and add a per-language provenance block:

{
  "slug": "genesis",
  "tradition": "hebrew-bible",
  "originalLang": "he",
  "primaryLang": "en",
  "availableLangs": ["en", "de", "es", "fr", "ja", "ko", "ru", "zh", "zh-Hant"],
  "translations": {
    "en": { "name": "American Standard Version", "year": 1901, "license": "Public Domain" },
    "de": { "name": "Elberfelder 1905", "year": 1905, "license": "Public Domain" },
    "es": { "name": "Reina-Valera 1909", "year": 1909, "license": "Public Domain" },
    "fr": { "name": "Crampon", "year": 1923, "license": "Public Domain" },
    "ja": { "name": "口語訳 (Kougo-yaku)", "year": 1955, "license": "Public Domain (moral rights apply)" },
    "ko": { "name": "개역성경", "year": 1938, "license": "Public Domain" },
    "ru": { "name": "Synodal", "year": 1876, "license": "Public Domain" },
    "zh": { "name": "和合本 (Chinese Union Version)", "year": 1919, "license": "Public Domain" },
    "zh-Hant": { "name": "和合本 (Chinese Union Version)", "year": 1919, "license": "Public Domain" }
  }
}

Exception: for French Samuel + Job, the fr entry overrides to Dhorme:

{
  "slug": "1-samuel",
  "translations": {
    "fr": { "name": "Dhorme, Les Livres de Samuel", "year": 1910, "license": "Public Domain", "publisher": "Lecoffre (Études Bibliques)" }
  }
}

Per-paragraph i18n#

Same schema as the WoH translations: each paragraph carries an i18n.{lang} field populated with the imported translation’s rendering of that verse. No commentary field — imported translations are bare text. Glossary refs are not applied (the imports use their own theological vocabulary, which the WoH glossary intentionally disagrees with — applying glossary refs would create the wrong association).

Source-acquisition methodology#

For each language + translation, we source from the most text-authoritative public host:

Every imported text records the source URL + accessed-on date in the chapter JSON’s translation.sourceAccessedVia field.

Acknowledgement of register mismatch#

All eight non-English PD choices are at least 60 years old, most over 100. Readers will encounter archaic German / Spanish / Russian / Chinese / Korean register. This is the cost of the public-domain constraint, accepted because:

1. The Hebrew Bible itself reads at archaic-register in any tradition that takes the text seriously. Modern dynamic-equivalence translations sand off the strangeness; the WoH project wants readers to notice the strangeness.
2. Each imported translation is paired with the WoH-lens English translation as the modern reference point. The imported text is the “what the standard text says” baseline, not the recommended reading.
3. As post-1955 translations enter the public domain over the next decades, this page will be revised. The Dhorme Pléiade enters PD around 2037-2071 (per collaborator deaths); the modern copyrighted translations follow on their own schedules.

See also#

• WoH Translation Program — the fresh-translation pipeline these imports complement.
• i18n Localisation — fanning out WoH-produced English content into the nine other site languages.
• Library Book — the reader-facing format every translation (WoH or imported) ships inside.