01 Executive Summary — Hebrews Language Package (English → Vietnamese)

Why it matters

Hebrews pushes the Vietnamese Language Package into territory the Romans baseline never had to cover: a physical sacrificial system (tabernacle, mercy seat, blood, altar) described in loving typological detail, set against a Vietnamese religious landscape where sacrificial and altar imagery is not abstract — it is the household ancestor altar most Vietnamese readers grew up beside. The core passage, Hebrews 9:11–28, sits at the exact center of this collision: it is simultaneously the curriculum’s theological anchor (Christ’s once-for-all entry into the heavenly sanctuary by his own blood) and the single densest concentration of Critical-risk translation decisions in the entire book. Getting this package right determines whether Vietnamese readers understand Christ’s sacrifice as a unique, finished, divine-initiated act — or quietly reinterpret it through merit accumulation (công đức), ancestor appeasement (cúng tạ lỗi), or karmic retribution (nghiệp báo).

Key findings

• Full 13-chapter coverage confirmed. Every chapter of Hebrews was analyzed; no chapter’s doctrinal content was silently skipped, including chapter 13’s lighter closing exhortations, which still carry the book’s single highest lexical-collision risk (bàn thờ / household ancestor altar, 13:10).
• 21 doctrines identified, extending the Romans baseline framework: 7 Critical, 7 High, 6 Medium, 1 Low. 14 doctrines require mandatory human theologian review; 6 require native speaker review; 1 is automated-review-sufficient.
• ~108 terms now governed by the extended translation memory: ~30 terms inherited exactly from the Romans baseline (10 of which carry an escalated risk tier for Hebrews-specific reasons — e.g., “Lord,” “law,” “covenant” — because Hebrews’ argument structure raises their doctrinal stakes beyond their Romans-context usage) plus ~78 new Hebrews-specific terms. 17 terms are Critical-risk and 42 are High-risk — 59 terms in total requiring theologian-level scrutiny, concentrated overwhelmingly in and around the core passage (chs. 7, 9, 10).
• Three genuinely new translation problems with no Romans precedent: (1) a newly coined term for “mercy seat” (nắp thi ân) that also retroactively fills a gap the Romans baseline left open at Romans 3:25; (2) a single Greek word (διαθήκη) that must switch from “giao ước” to “di chúc” for exactly two verses (9:16–17) and switch back; (3) three lexically distinct, escalating Vietnamese apostasy terms (lìa bỏ / sa ngã / cố ý phạm tội) that must never collapse into the everyday, flattening “bỏ đạo.”

Risks

• Highest risk: the Once-for-All Atoning Sacrifice doctrine (Critical), centered in the core passage. Vietnamese merit-based folk Buddhism (công đức, phước đức) and ancestor-appeasement practice (cúng tạ lỗi) both offer a ready but doctrinally destructive misreading of Christ’s sacrifice as an unusually large “merit deposit” or a spirit-placating ritual gift, rather than a unique, non-repeatable, substitutionary legal-relational act.
• Highest lexical-collision risk: “bàn thờ” (altar, 13:10) and “thầy tế lễ thượng phẩm” (high priest, chs. 2–10). Both terms sit inside extremely high-frequency Vietnamese religious vocabulary (the household ancestor altar; the everyday address for a Buddhist monk) that has nothing to do with their biblical referents.
• Structural risk: flattening distinctions the source text deliberately preserves — the three-tier apostasy vocabulary, the dual-sense ὑπόστασις (Christ’s deity in 1:3 vs. faith’s assurance in 11:1), and the covenant/testament wordplay in 9:16–17. Any of these collapsing into a single Vietnamese term would erase a specific argument Hebrews is making.
• Escalation-tracking risk: 10 terms shared with the Romans baseline carry a higher risk tier in Hebrews than in Romans (lord, faith, glory, law, sin, holy, sanctification, covenant, called, calling). Phase 2 workers relying only on the Romans-tier routing would under-review these terms.

Opportunities

• A near-complete century-old vocabulary base already exists (sự cứu rỗi, sự sống lại, Đức Thánh Linh, ân điển) that Hebrews inherits with zero renegotiation required — most of the translation risk is genuinely new-to-Hebrews vocabulary, not a re-fight of settled Romans battles.
• The core passage’s density is also an asset: because Hebrews 9:11–28 concentrates the highest-risk terms into 18 verses, theologian review effort can be front-loaded onto this single passage for maximum risk reduction per review-hour.
• The “nắp thi ân” coinage closes a known gap in the existing Language Package (the unresolved Romans 3:25 ἱλαστήριον question), improving consistency across both curricula going forward.

Recommended actions

1. Route all Critical-risk core-passage terms (mercy seat, once/once-for-all, the covenant/testament wordplay, propitiation) through mandatory human theologian review before any Phase 2 batch touching Hebrews 9 is finalized.
2. Enforce the escalated Hebrews-specific risk tiers (Section 1 of assets/translation_memory.json) as authoritative for all Hebrews content, overriding the lower Romans-baseline tier wherever the two differ.
3. Require a standing contextual footnote at every occurrence of “bàn thờ,” “thầy tế lễ thượng phẩm,” and “sự cầu thay” to pre-empt the three highest-frequency folk-religious misreadings (ancestor altar, Buddhist monastic master, ancestor petition).
4. Lock the three apostasy terms and the ὑπόστασις dual-sense split as non-negotiable, zero-variance renderings in Phase 2 tooling, with automated validation checks (see 12_ai_translation_requirements.md) rather than relying on reviewer memory alone.

---

Cross-reference assets/doctrine_risk_registry.json, assets/translation_memory.json, and analysis/11_doctrine_analysis.md for full supporting detail.