Executive Summary — Ephesians Japanese Translation Requirements

Why it matters

Ephesians is being added to a Japanese Bible curriculum that already has a doctrinally sound baseline (the Romans package, anchored to Shinkaiyaku 2017). Japanese Bible translation vocabulary is mature — but Ephesians pushes several baseline terms harder than Romans did (church, temple, mystery, obedience, resurrection-union) and introduces entirely new doctrinal vocabulary (predestination, submission/headship, armor of God, the gift/works disambiguation) that sits directly on top of Japan’s most consequential religious and social fault lines: on-giri reciprocity, Buddhist merit and enlightenment vocabulary, the ie household system, uchi-soto insider/outsider instinct, and the animistic spirit-world. Getting Ephesians 2:1-10 — the core passage and the curriculum’s thesis statement on grace, faith, and works — wrong in Japanese risks the single most consequential doctrinal miscommunication in the entire package.

Key findings

• 26 curriculum doctrines analyzed; 7 Critical + 10 High = 17 require mandatory human theologian review (per analysis/11_doctrine_analysis.md’s corrected doctrine matrix); 8 require native speaker review; 1 (Thanksgiving) is automated-only.
• Translation memory now carries roughly 43 Critical/High-risk terms across the combined Romans-baseline-inherited set and Ephesians-new set — a substantially heavier concentration than Romans alone, driven by Ephesians 2 (grace/gift/works) and Ephesians 5 (submission/headship/armor).
• Two genuinely new collision risks not present in the Romans baseline: (1) 奥義 (mystery) — Japan’s living martial-arts/esoteric-Buddhism usage means “secret technique for initiates,” the exact inverse of Paul’s “now revealed to all” point; (2) the 賜物/贈り物 split — Ephesians uses the same Greek root for “the gift of salvation” (2:8) and “gifts for ministry” (4:7-11), which Japanese cannot disambiguate by context alone and requires a hard lexical rule.
• The core passage (2:1-10) sits directly on Japan’s on-giri reciprocal-obligation logic and Buddhist merit-accumulation (功徳) framework: “not by works” risks sounding evasive rather than good news without heavy scaffolding, and v.9’s rejected 行い / v.10’s affirmed 良い行い share the same Japanese root, risking an apparent self-contradiction.
• Household codes (5:21-6:9) and the armor of God (6:10-17) are the two other Critical-risk clusters: the pre-1947 ie system makes submission/headship the single most culturally loaded passage in the book, and samurai-armor imagery, while a vivid bridge, risks romanticized-combat or militarist-era resonance.
• Several genuine positive cultural bridges exist and should be leveraged, not just risk-managed: Japan’s inkan/seal culture for the sealing of the Spirit (1:13-14), adult-heir adoption (養子縁組) and modern inheritance culture (相続) for adoption/inheritance, and 職人技 craftsmanship respect for “workmanship” (2:10).

Risks

• Doctrinal: 奥義 (mystery), 一致 (unity), 和解 (reconciliation), and 従う/服従する (submission) each carry a dilution or inversion risk rather than an outright wrong-word risk — the correct term is chosen, but its default Japanese connotation runs opposite to or short of Paul’s meaning, making the error harder for reviewers to catch than an obviously wrong term would be.
• Structural: πλήρωμα (fullness, 1:23; 3:19; 4:13) has no Japanese lexical equivalent at all; the required paraphrase is fragile under length constraints and could be silently compressed away by a Phase 2 worker.
• Consistency: the 賜物/贈り物 split and the 行い/良い行い fruit-not-basis scaffolding must be enforced identically across every batch-processed segment referencing Ephesians 2 or 4; a single inconsistent worker output reintroduces the exact doctrinal conflation the split exists to prevent.
• Cultural-historical: armor-of-God and kingdom-of-God language both carry residual pre-1945 militarist/ultranationalist resonance (神国日本) for readers old enough to remember it, requiring non-triumphalist framing throughout chapter 6.

Opportunities

• Japan’s personal-seal culture, adult-heir adoption/inheritance culture, and craftsmanship-respect culture offer three genuine, low-risk positive bridges this curriculum can lean into more than Romans could.
• Ephesians’ own internal correction mechanism — 5:21’s mutual submission and 5:25’s self-giving love — gives translators a built-in, textually mandated counterweight to the ie-system risk, rather than requiring an imported apologetic.
• The letter’s “already revealed, no longer hidden” mystery framing, taught explicitly against 奥義’s inverse connotation, can become a memorable, distinctive teaching point rather than merely a risk to manage.

Recommended actions

1. Lock the assets/translation_memory.json and assets/doctrine_risk_registry.json extensions as authoritative before any Phase 2 segment processing begins; do not allow ad hoc term substitution.
2. Route all Ephesians 2:1-10 and 5:21-6:9 segments to human theologian review as a non-negotiable Phase 2 gate, regardless of automated quality score.
3. Enforce the 賜物/贈り物 hard split and the 奥義 “now revealed” scaffolding as automated validation checks (not just style guidance) in the Phase 2 pipeline, per analysis/12_ai_translation_requirements.md.
4. Provide Phase 2 workers and reviewers with the ranked risk list from analysis/06_linguistic_gap_analysis.md §5 as a triage checklist before batch assignment.
5. Preserve and highlight the three positive cultural bridges (seal, inheritance, workmanship) in teaching-material design, not just in translation notes, to give the curriculum’s Japanese readers memorable points of felt connection alongside its risk-managed vocabulary.