Executive Summary
Executive Summary
English → Bengali | Zephaniah 1–3 | Phase 1 Translation Requirements
Curriculum: Zephaniah (full book, 1:1–3:20) Core passage: Zephaniah 3:14–17 Format: Smart Brevity
Why it matters: Zephaniah is the first Old Testament, Hebrew-original curriculum built on top of the Bengali Romans (NT/Greek-original) Language Package. It introduces the divine covenant name יהוה — used in nearly every verse — into a Bengali translation ecosystem that had, until now, only had to discipline one divine title (প্রভু, for κύριος). Getting the সদাপ্রভু/প্রভু/ঈশ্বর three-way distinction right is now the single highest-leverage decision in the entire Bengali Bible-translation pipeline, not just this book.
Key findings:
- Zephaniah requires 9 new or extended proper-name/theological terms at Critical risk and 21 at High risk (per
08_core_glossary.mdrisk summary: 8 Critical + reused ধার্মিকতা/ধার্মিক = 9 Critical-tier items; 21 High), against a backdrop of the baseline Romans package’s own 7 Critical and 22 High NT terms — meaning Phase 2 reviewers must now hold two overlapping but distinct Critical-term vocabularies in view simultaneously. - The doctrine layer resolves to 13 total doctrines, of which 10 require mandatory human theologian review (4 Critical + 6 High) and 3 require native speaker review (Medium), with zero doctrines resolving to automated-only review — a materially higher theologian-review burden than the Romans baseline carried proportionally, driven by Zephaniah’s OT-specific collision risks.
- The single sharpest NEW cultural-collision point this book introduces is প্রতিমাপূজা (idolatry/idol-worship): the same Bengali word names both the ancient sins Zephaniah condemns (Baal, Milcom, the host of heaven) and mainstream, currently and publicly practiced Bengali Hindu image-worship (Durga Puja, Kali Puja) — sharper than any single collision term in the Romans baseline.
- A second, purely internal risk — invisible to anyone checking only against external religious traditions — is the shared Bengali root অবশিষ্ট অংশ (“remnant”), used BOTH for the condemned “remnant of Baal” (1:4) and the graciously preserved “remnant of Israel” (2:9; 3:13), the book’s own positive doctrinal centerpiece.
- The core passage (3:14–17) carries the book’s warmest material — God himself as the grammatical subject of joy over his people — and is simultaneously its highest bhakti-collision risk, given Bengali Vaishnava devotional vocabulary’s ready supply of divine-joy and divine-love language that must be deliberately fenced off (ভালোবাসা, never প্রেম).
Risks:
- Doctrinal collapse risk (Critical): conflating সদাপ্রভু with প্রভু or ঈশ্বর at any single occurrence undermines either the OT/NT proper-name distinction or the force of the NT Lordship confession already fixed in the baseline — and this risk recurs on nearly every verse of the book, not on isolated occurrences.
- Structural-argument loss risk (Critical): inconsistent rendering of সদাপ্রভুর দিন across chapters 1 and 3 would make the book’s central claim — one Day, both wrath and restoration — invisible to a reader or teacher encountering the chapters separately.
- Social-sensitivity risk (High): unfenced প্রতিমাপূজা language could read as a live, targeted accusation against Hindu neighbors in West Bengal rather than a historically bounded ancient indictment.
- Internal-consistency risk (High): the অবশিষ্ট অংশ root-collision, if left unflagged, could cause a reader to associate the salvific Remnant-of-Israel doctrine with the condemned “remnant of Baal” simply through shared vocabulary.
- Register-drift risk (High): informal oral teaching or worship-song adaptation of 3:14–17 is more likely than the base written translation to drift toward বীর or প্রেম, undoing glossary discipline established at the text level.
Opportunities:
- Zephaniah 3:14–17’s portrait of God rejoicing over his people is exceptionally well-suited, once properly fenced, to resonate warmly with Bengali readers from a bhakti-devotional cultural background — a genuine bridge for relational, joy-centered gospel communication if ভালোবাসা/উল্লাস vocabulary is deployed with the discipline this package specifies.
- The Remnant of Israel doctrine’s insistence that humility — not ethnicity, caste, or communal birth-identity — marks those God preserves offers a pointed, pastorally significant message in both West Bengal’s caste-inflected and Bangladesh’s communal-identity-inflected social contexts.
- This package establishes the OT covenant-name convention (সদাপ্রভু) and the OT/NT distinction discipline that every future Old Testament Bengali curriculum can now inherit without re-deriving it from scratch.
Recommended actions:
- Load both
assets/translation_memory.json(Zephaniah) and the baseline Romanstranslation_memory.jsontogether for every Phase 2 segment; treat them as cumulative, never as alternatives. - Route all 10 theologian-review doctrines and every Critical/High term occurrence through mandatory human theologian review before Phase 2 approval, per
doctrine_risk_registry.jsonand12_ai_translation_requirements.mdPart B. - Attach the 8 mandatory translator notes specified in
12_ai_translation_requirements.md(সদাপ্রভু distinction, remnant root-collision ×2, idolatry historical fencing, complacency idiom loss, covenant jealousy, King of Israel referent, save/deliver root-continuity, MT/LXX variant) at their designated first occurrences before any segment is marked approved. - Brief all Phase 2 reviewers explicitly on the সদাপ্রভু/প্রভু/ঈশ্বর distinction as a standalone training item before they process a single Zephaniah segment — this is the package’s highest-leverage single control point.
- Carry the অবশিষ্ট অংশ, ব্যবস্থা, ধার্মিকতা/ধার্মিক, ভাববাদী, and পুরোহিত consistency rules forward into any future Old Testament curriculum to avoid re-litigating already-settled Bengali OT terminology decisions.