Executive Summary — Luke 1–24 (English → Japanese)

Why it matters

Luke is the largest, most narrative-dense curriculum this Language Package has processed to date. Its 24 chapters carry roughly double the Critical-tier doctrinal risk of the Romans baseline (16 Critical doctrines vs. Romans’ 7) because Luke narrates the Incarnation, Virgin Birth, Transfiguration, Crucifixion, Resurrection, and Ascension directly, in concrete historical and biological detail, rather than assuming them as settled background the way Paul’s letter does. Two genuinely new Japan-specific Critical collision risks — Hades vs. 黄泉 (Yomi) and the Ascension vs. Shinto posthumous apotheosis — have no parallel anywhere in the Romans baseline and require dedicated forbidden-substitution rules. Getting this Language Package right determines whether Japanese learners encounter Jesus as Savior for all nations and all people, or encounter a syncretistically blurred figure absorbed into Japan’s existing kami, bodhisattva, or arahitogami categories.

Key findings

• 38 total doctrines catalogued across full-book coverage (every chapter reviewed, none silently skipped): 16 Critical, 15 High, 5 Medium, 2 Low — a combined 31 doctrines (82%) require mandatory human theologian review, versus the baseline’s 26-of-40 (65%).
• ~119 terms now governed by the combined translation memory: 40 terms inherited exactly from the Romans baseline, plus 79 newly introduced Luke-specific terms. Of the new terms, 8 are Critical and 24 are High risk — 32 of 79 new terms (41%) require mandatory theologian review, a higher proportional concentration than Romans’ new-term risk profile.
• Luke’s highest-frequency load-bearing term is 神の国 (Kingdom of God), occurring 30+ times versus a handful in Romans — the pre-1945 State Shinto slogan risk (神国日本) must be re-anchored at every chapter cluster, not resolved once.
• Two structural Greek-to-Japanese vocabulary gaps have no clean fix: σῴζω (save/heal/make whole) and ἄφεσις (release/forgiveness) each collapse into two different Japanese surface words, silently erasing Luke’s own deliberate wordplay unless every occurrence carries a translator cross-reference note — this affects the core passage itself (4:18-19) and the book’s thesis verse (19:10).
• The core passage (Luke 4:16-21) is the single most doctrinally concentrated five verses in either curriculum, stacking Messianic Promise, Anointing, Fulfillment of Prophecy, Inspiration of Scripture, and Kingdom of God doctrines simultaneously — every segment within it is flagged for mandatory theologian review.

Risks

• Newly identified, no baseline precedent: Hades/黄泉 (Yomi, the named Shinto/Kojiki underworld), Ascension/Shinto posthumous apotheosis (Yasukuni-pattern deification), Transfiguration/kitsune-tanuki shape-shifting folklore, and Paradise/極楽 (Pure Land Buddhist Paradise) plus secular travel-marketing dilution.
• Japan-specific discrimination-history risk: leprosy terminology (らい病人) is not merely doctrinally imprecise but actively harmful given Japan’s 1907–1996 Leprosy Prevention Law history — this is a living-memory harm risk, not only a theological one.
• Affective-flattening risk (not mistranslation): “the poor,” disciple-cost language, and table fellowship can be rendered denotatively correct while still landing with the wrong emotional/social weight, given Japan’s 一億総中流 middle-class self-image, its guild-apprenticeship model of 弟子, and modern dining customs lacking first-century honor/shame boundary-crossing associations respectively — these failures are harder to catch in review than outright wrong-word errors.
• Cumulative-exposure risk: 神の国 and 聖霊 occur so frequently in Luke that a single un-anchored later instance can undo correct earlier framing — the baseline’s “anchor once” instruction is insufficient at Luke’s frequency.

Opportunities

• Genuine cross-tradition teaching bridges exist and should be used deliberately (never as equivalence): Pure Land Buddhism’s tariki (other-power) salvation-by-faith structure for the Savior doctrine; Japan’s established adult-heir adoption practice for sonship/inheritance imagery; the existing Japanese idiom 放蕩息子 (already culturally embedded from this very parable) as an entry point for teaching non-transactional grace; Islamic tawhid-adjacent affirmation of the virgin birth as a shared starting point before pivoting to the (rejected) Incarnation claim.
• Luke’s narrative form is pedagogically well-suited to a biblically illiterate audience: unlike Romans’ dense argumentation, Luke’s parables, healing accounts, and table-fellowship scenes give concrete, memorable anchors (Zacchaeus, the Prodigal Son, the Good Samaritan) for doctrines that would otherwise remain abstract.
• This Language Package’s dual-file structure (Romans baseline + Luke extension) sets a reusable pattern for future Gospel or Acts curricula, since the majority of Christological and soteriological vocabulary is now permanently fixed across the whole Bible-translation program rather than re-derived per book.

Recommended actions

1. Route all 31 Critical/High doctrines and all 32 Critical/High new terms to mandatory human theologian review before any Phase 2 segment is marked approved, per the escalation rules in 12_ai_translation_requirements.md.
2. Lock the exact Japanese rendering of Luke 4:18-19 (core passage) and Luke 19:10 (thesis verse) immediately, following the same cross-document consistency protocol the baseline applies to Romans 1:16-17 and 8:28.
3. Brief all reviewers explicitly on the two newly identified Critical, no-baseline-parallel risks (Hades/Yomi; Ascension/apotheosis) before Phase 2 begins, since no prior Romans-package training material covers them.
4. Require the sōzō and ἄφεσις translator cross-reference notes as a hard validation gate (not a best-effort flag) given how easily these two structural gaps could pass review undetected.
5. Confirm the leprosy terminology rule (重い皮膚病の人, never らい病人) with a dedicated sensitivity review pass, independent of the standard doctrinal-fidelity review, given the real-world harm dimension involved.