Executive Summary — Ezekiel (Hindi) Language Package

Why it matters

Ezekiel is the highest-collision-risk Old Testament book yet processed for Hindi in this pipeline. Its central images — a throne-borne divine glory, a departing-and-returning presence, dry bones brought to life by breath/wind/Spirit, an aniconic future temple, and an individual-accountability argument that directly confronts hereditary guilt — sit on top of the deepest and most active seams in Hindu religious vocabulary: darshan, prāṇa, ātman/paramātman, karma-and-rebirth, Pātāla, and mandir-as-murti-housing. Get any one of these wrong and the curriculum doesn’t just mistranslate a word — it silently teaches reincarnation, an impersonal life-force, a merit-transferable cosmology, or a devotional-image theology the text itself was written to exclude. This package exists to make those decisions once, correctly, and enforce them everywhere.

Key findings

• 82 terms are fixed in assets/translation_memory.json: 26 inherited exactly from the Romans/Galatians baseline (e.g. परमेश्वर, प्रभु, वाचा, पाप, धार्मिकता, पवित्र आत्मा) and 56 new terms this book alone introduces.
• 14 Critical-risk terms and 31 High-risk terms (45 combined) require mandatory human theologian review at every occurrence — more than half the glossary.
• 22 doctrines are tracked in assets/doctrine_risk_registry.json: 7 Critical, 10 High, 5 Medium, 0 Low. 17 of 22 doctrines (77%) route to human theologian review; the remaining 5 route to native speaker review. No doctrine in this book is automated-only.
• The single highest-stakes decision in the entire curriculum is not using the baseline’s Critical term पुनरुत्थान (resurrection) for Ezekiel 37’s dry bones — Ezekiel 37:11 self-identifies its own referent as corporate Israel’s national restoration, not primarily individual bodily resurrection, and conflating the two would damage Christ’s unique resurrection doctrine elsewhere in the pipeline.
• The Hebrew wordplay on רוּחַ (ruach = Spirit/breath/wind) across Ezekiel 37:1-14 has no single Hindi equivalent and requires a three-way, context-governed split (आत्मा / श्वास / हवा) bound by a mandatory translator’s note — this is the core passage’s defining linguistic-engineering problem.

Risks

• Sheol → पाताल, four winds → वायु, nasi → राजा, temple → unqualified मन्दिर, and YHWH → generic ईश्वर/भगवान are five distinct forbidden-substitution traps, each of which imports an entire foreign cosmology or erases a deliberate textual choice if triggered even once.
• Ezekiel 18’s individual-responsibility argument is uniquely exposed: India’s live hereditary-dharma and karma-across-lifetimes assumptions are precisely what the chapter’s ancient proverb (“the fathers ate sour grapes”) was written to refute — this is a sharper collision than the general baseline धर्म-guardrail anticipated.
• The throne-vision vocabulary (מַרְאֶה/דְּמוּת) and the new-temple vocabulary (מִקְדָּשׁ/הֵיכָל) both sit inside Hindi’s single most crowded religious semantic neighborhood — darshan and mandir/murti — and both require mandatory recurring notes across many chapters (1, 8, 10, 40-43 and 37, 40-48 respectively), creating sustained review burden, not a one-time fix.
• The covenant-unfaithfulness/harlotry allegories (chs. 16, 23) carry real pastoral-delivery risk (explicit, sustained sexual imagery) requiring combined native-speaker-plus-theologian review, not just terminology review.
• ~217 occurrences of “Lord GOD” and ~70 occurrences of the recognition formula create a high-volume consistency-discipline burden where a single stylistic slip (varying यहोवा/परमेश्वर/प्रभु) blurs the book’s central rhetorical strategy.

Opportunities

• Ezekiel 18’s individual-accountability doctrine, properly framed, is a uniquely powerful apologetic against fatalistic and hereditary-guilt worldviews for a Hindu-background audience — this curriculum can teach it as a direct, respectful confrontation rather than an incidental ancient-history lesson.
• The new-heart/new-spirit promise (chs. 11, 18, 36) and its dry-bones dramatization (ch. 37) offer an unusually vivid, emotionally resonant on-ramp to the gospel’s grace-not-self-effort message, provided the आत्मा-as-gift (not self-attained ātman-realization) framing is taught explicitly and early.
• The aniconic new-temple vision (chs. 40-48), rightly distinguished from mandir/murti theology, gives this curriculum a rare opportunity to teach positively about worship of an unseen, unrepresented God — a distinctly countercultural but clarifying contrast for the destination audience.
• The recommended primary Scripture reference (Hindi Holy Bible / BSI Old Version, YouVersion id 819) already anchors nearly every established rendering in this glossary, minimizing friction between this curriculum’s text and what learners will encounter in their own Bibles.

Recommended actions

1. Load assets/translation_memory.json (Ezekiel) together with the baseline Romans/Galatians translation memory before any Phase 2 segment translation — never one without the other.
2. Route all 45 Critical/High-risk terms and all 17 theologian-review doctrines through mandatory human theologian sign-off before any segment is marked approved, with special priority on the ten Critical terms (yhwh_divine_name, lord_god_adonai_yhwh, recognition_formula, mareh_demut_vision, ruach_wordplay, new_heart_new_spirit, dry_bones_revival, sheol_realm_of_dead, glory_departs_returns, yhwh_shammah).
3. Apply the mandatory translator’s-note requirements verbatim wherever specified (ruach wordplay, Son of Man address, temple aniconic distinction, darshan distinction, dry-bones vs. resurrection) — these notes are load-bearing controls, not optional stylistic flourishes.
4. Schedule theologian review capacity above the Romans/Galatians baseline rate, given this book’s unusually high 77% theologian-routing proportion at the doctrine level.
5. Flag chs. 16 and 23 for combined native-speaker-plus-theologian review before any lesson drafting begins, per the honor/shame and explicit-content routing precedent already established in the baseline.

---

This summary synthesizes assets/translation_memory.json, assets/bible_term_registry.json, assets/doctrine_risk_registry.json, analysis/07_semantic_analysis.md, analysis/08_core_glossary.md, analysis/11_doctrine_analysis.md, analysis/04_comparative_theology.md, analysis/06_linguistic_gap_analysis.md, and analysis/05_translation_landscape.md. It does not introduce new terminology decisions; all figures are drawn directly from those documents.