Executive Summary: Gujarati Language Package for Luke

Why it matters

Luke is Gujarat’s hardest Gospel to translate safely. It stages, in narrative form, nearly every collision the Romans baseline already flagged in the abstract — plus several new ones unique to Luke’s storytelling: a personal Savior title with no Gujarati Christian precedent, a poor and marginalized cast whose condition popular Hindu/Jain folk theology can misread as karmic debt, a discipleship vocabulary (શિષ્ય) that is the exact word for self-initiated guru-devotion in Gujarat’s dominant Swaminarayan/Vaishnav and Jain traditions, and a table-fellowship ethic that directly confronts still-active caste-commensality norms. Getting this Language Package right determines whether Luke reads to a Gujarati believer as the specific, historical, grace-driven account it is — or gets silently absorbed into avatar theology, karma accounting, or ascetic self-effort.

Key findings

• 110 enforced glossary terms now govern Phase 2 translation of Luke: 39 inherited exactly from the Romans baseline, 72 new terms/compounds introduced by Luke’s own narrative content (Savior, Son of Man, Redemption, Repentance, Disciple, Cross, Hades, Paradise, and more).
• 58 of those 110 terms (23 Critical + 35 High) require Human Theologian review at every occurrence — a materially higher Critical-term count than Romans carried, driven by Luke’s Christological titles (Savior, Son of Man) and its concrete social-doctrine vocabulary (Repentance, Forgiveness, Table Fellowship, Cross).
• 36 doctrines are tracked in the full-book doctrine matrix; 10 are Critical-risk and 21 are High-risk, for 31 of 36 doctrines (86%) requiring Human Theologian review — only 3 route to native speaker review and 2 to automated review alone.
• The core passage (Luke 4:16-21) is the single densest doctrinal unit in the curriculum: it contains the anointing/Messiah wordplay, the fourfold Isaiah announcement (poor/captives/blind/oppressed), the Jubilee “acceptable year,” and the “today…fulfilled” declaration — all Critical or High risk, all now fixed in the glossary.
• A documented, real-world cult/sect collision exists at Luke 23:43: the Jehovah’s Witness Gujarati New World Translation moves a single comma to defer Christ’s promise of Paradise — now a zero-tolerance, explicitly named escalation trigger.

Risks

• Poverty-as-karma folk reasoning (doctrines 3/7, ગરીબ) is assessed as the single highest silent-collision risk in the entire curriculum — it operates as a cultural inference, not a mistranslation, and requires teaching-note intervention at every occurrence, not just correct word choice.
• Discipleship agency reversal (શિષ્ય, doctrine 14, Critical): every guru-shishya-structured tradition prominent in Gujarat has the aspirant seeking the teacher; Luke has Jesus calling the disciple. This must be reinforced continually across the book, not stated once.
• Compassion, joy, and love (કરુણા, આનંદ, પ્રેમ — all High risk) sit inside Gujarat’s own positively-loaded devotional vocabulary (Jain ascetic virtue, Swaminarayan bliss-nature, Vaishnav prem-bhakti), making these collisions feel like agreement rather than divergence — the most dangerous kind of risk.
• Zero-tolerance sectarian collision at Luke 23:43 (JW comma placement) and systemic risk from Ahmadiyya (ઈસા, resurrection denial) and Hindu-Christian syncretism tracts (અવતાર, ગુરુ ઈસુ) circulating in Gujarat’s print/digital ecosystem.

Opportunities

• Luke’s own universal, boundary-crossing material (Good Samaritan, centurion, sinful woman, Zacchaeus, Gentile mission at 24:47) gives the translation an unusually strong built-in apologetic against caste- and community-based spiritual hierarchy — provided doctrines 2, 3, and 8 are translated at full, unsoftened strength as required.
• The Jewish and Islamic comparative-theology columns (04_comparative_theology.md) show genuine common ground on direct, unmediated prayer to a personal God and on care for the poor/widow/orphan — useful bridges for teaching material aimed at inquirers from those backgrounds.
• Fixing the YouVersion citation convention (Gujarati O.V., ID 1953, LUK book code) now gives every future Luke-based lesson, sermon, or cross-reference a stable, auditable citation standard.

Recommended actions

1. Load assets/translation_memory.json (Luke) together with — never instead of — the Romans baseline TM before any Phase 2 segment translation begins.
2. Route all 58 Critical/High-risk terms and all 31 Critical/High-risk doctrines to Human Theologian review per the escalation rules in analysis/12_ai_translation_requirements.md; do not allow automated-only approval for any of them.
3. Attach the fixed teaching-note interventions (poverty ≠ karma; discipleship = Jesus-initiated; compassion/joy/love flow from God, not self-cultivated) to every occurrence, not just first occurrence per book.
4. Treat any segment matching the Cult/Sect Renderings table (05_translation_landscape.md §3) as an automatic Critical escalation regardless of the underlying term’s own risk tier, with the Luke 23:43 comma-placement rule enforced with zero tolerance.
5. Carry this Language Package forward unmodified as the baseline for any future third curriculum in this destination language, exactly as this document itself carried the Romans package forward without contradiction.

---

This document synthesizes analysis/07_semantic_analysis.md, analysis/08_core_glossary.md, analysis/04_comparative_theology.md, analysis/06_linguistic_gap_analysis.md, analysis/05_translation_landscape.md, analysis/11_doctrine_analysis.md, assets/translation_memory.json, assets/bible_term_registry.json, and assets/doctrine_risk_registry.json. All counts are drawn directly from those artifacts and are current as of this document’s date.