01 Executive Summary — Ephesians (English → Mandarin)

Curriculum: Ephesians 1–6 | Core passage: Ephesians 2:1–10 | Destination: Mandarin Chinese (Simplified, CUV 神版 register)

Why it matters

Ephesians is being added to a Mandarin translation pipeline whose baseline (Romans) already established that this language pair carries a structural risk, not merely a lexical one: Mandarin’s dominant religious vocabulary — Confucian, Buddhist, Daoist, and folk-religious — is saturated with merit-accounting, fatalism, and appeasement concepts that run directly opposite to core Pauline claims. Ephesians concentrates this risk more densely than Romans per chapter: its own thesis passage (2:8–10) is the single sharpest grace-versus-merit statement in the whole two-book corpus, its household-code section (5:21–6:9) collides with a still-influential Confucian hierarchy, and its spiritual-warfare section (6:10–20) sits inside the most crowded competing-cosmology semantic field in the entire curriculum. Getting Phase 2 wrong here does not produce awkward prose — it produces a translation a Mandarin-speaking reader’s existing worldview will silently “correct” back toward folk merit, fatalism, or appeasement.

Key findings

• 26 curriculum doctrines identified and risk-tiered (10 Critical, 10 High, 4 Medium, 2 Low), anchored in Ephesians 2:1–10 and spanning all six chapters with no chapter contributing zero load-bearing material.
• 117 enforced terms now in translation_memory.json for this book: 30 inherited exactly from the baseline Romans package, 87 newly introduced by Ephesians.
• Of the 117 terms, 22 are Critical risk (10 inherited + 12 new) and 43 are High risk (10 inherited + 33 new) — 65 terms total (56% of the glossary) require mandatory human theologian review on every occurrence, a materially higher proportion than the Romans baseline.
• Three doctrinal clusters carry disproportionate risk density: (1) the grace/works causal-direction cluster at the core passage (2:8–10), (2) the household-authority cluster (5:21–6:9), and (3) the spiritual-powers/armor cluster (1:21; 2:2; 3:10; 6:10–20).
• Mandarin has no adequate single-word rendering for predestination, redemption, the Jew/Gentile “new humanity,” the sealed/guarantee pair, or mutual submission — all require compound/paraphrase solutions already adopted in 08_core_glossary.md and reconfirmed here.

Risks

• Grace-works reversal (Critical): the Chinese proverb-level default “积善之家必有余庆” (accumulated good deeds produce blessing) inverts Ephesians 2:8–10’s actual causal claim; this will occur to readers by default, not only through mistranslation.
• Household-code collapse into 三纲 (Critical): rendering 顺服/元首 without the mutual-submission (5:21) and self-sacrificial-love (5:25) frame reproduces the unilateral, non-reciprocal Confucian Three Cardinal Guides structure the passage does not intend.
• Spiritual-powers appeasement drift (Critical): Chinese folk cosmology’s legitimate, appeasable celestial bureaucracy (城隍, 土地公, 玉皇大帝) makes an appeasement-register mistranslation of “rulers and authorities” not just theologically wrong but liturgically dangerous — it could produce genuine syncretistic devotional practice.
• Idolatry-naming pastoral risk (Critical): 拜偶像的 (5:5) names practices likely present in some readers’ own families; mishandling this is a pastoral-relational risk, not only a lexical one.
• Silent term drift across two curricula: with Romans and Ephesians now sharing one Language Package, inconsistent rendering of shared terms (grace, election, resurrection, powers) across the two books would fragment the learner’s vocabulary base.

Opportunities

• Positive cultural resonance to leverage, not merely defend against: 孝敬父母 (honor parents, 6:1–2) overlaps substantially and positively with 孝, giving this curriculum a rare opportunity to affirm cultural values while sharpening the Christ-centered motivation — the opposite posture from most of the terms above.
• Established CUV infrastructure reduces invention burden: over a century of CUV usage means most needed compounds (救赎, 印记, 全副军装) already exist in Chinese Christian usage; the task is disciplined defense and consistent glossing, not new coinage.
• Armor-of-God imagery is linguistically low-risk at the object level: concrete nouns (带子, 护心镜, 藤牌, 头盔, 宝剑) carry little collision risk themselves, letting reviewers concentrate effort entirely on the paired abstract doctrinal nouns (righteousness, faith, salvation) inside each armor piece.

Recommended actions

1. Route Ephesians 2:8–10, 5:21–33, and 6:10–20 to a single, consistent human theologian review queue (not distributed across parallel workers) to guarantee identical rendering everywhere each is quoted, per the Step 10 requirements document.
2. Enforce the mandatory causal-direction gloss on every occurrence of 善行/恩典/所赐的 in the core passage before Phase 2 sign-off.
3. Enforce the “in Christ” anchor as a non-droppable clause on every election/predestination term (预定/拣选/定意/预备/呼召) per the linguistic gap analysis.
4. Require pastoral-sensitivity notes, not merely doctrinal-accuracy notes, on every occurrence of 拜偶像的 (5:5).
5. Confirm before Phase 2 launch that all 30 inherited baseline terms in translation_memory.json remain byte-for-byte identical to the Romans package — any drift here would break cross-curriculum consistency for learners moving between books.

Full detail: 06_linguistic_gap_analysis.md (term-level risk reasoning), 08_core_glossary.md (term inventory), 11_doctrine_analysis.md (doctrine matrix), 04_comparative_theology.md (cross-tradition survey), assets/doctrine_risk_registry.json and assets/translation_memory.json (machine-readable enforcement databases), 12_ai_translation_requirements.md (Phase 2 instruction set).