Executive Summary — Jonah Language Package (English → Portuguese)

Why it matters

Jonah is being added to a Portuguese-language Bible study pipeline whose baseline (Romans) already established that Brazil’s religious landscape is not generic “secular” or “Catholic-only” territory — it is a mainstream, textually codified, organized rival-doctrine environment, dominated in relevant part by Kardecist Spiritism (the world’s largest self-identified Spiritist population) and Afro-Brazilian Candomblé/Umbanda. Jonah intensifies every one of these risks rather than introducing new ones: its central plot — a prophet’s death-like ordeal, a fixed time period, and emergence to new life — is the single most reincarnation-suggestive narrative pattern in either curriculum, and it is also the Old Testament passage Jesus himself cites (Matthew 12:40) as the type of his own resurrection. Getting this book wrong in Portuguese doesn’t just blur a doctrine; it can accidentally hand a Spiritist reading its own proof-text.

Key findings

• 20 Jonah-specific doctrines identified, full-book (chapters 1–4), anchored on the core passage Jonas 3:1-10; 10 are Critical or High risk (4 Critical + 6 High), all routed to mandatory human theologian review; 8 Medium-risk doctrines route to native speaker review; 2 Low-risk doctrines are automated-only.
• 21 individual Critical/High-risk terms identified at the term level (10 Critical + 11 High), spanning both terms inherited exactly from the Romans baseline (Deus, Senhor, ressurreição, salvação, fé, graça, pecado, providência) and 40 newly registered Jonah-specific terms.
• The single highest-stakes new risk in this Language Package is the nacham/“God relented” — shuv/“turn, repent” pair at the doctrinal center of the core passage (Jonas 3:9-10), which must be taught as an immediate, God-enabled turning met by a covenant-faithful (not evolving, not fallible) God — directly contra Kardecism’s “reforma íntima,” a gradual self-improvement doctrine spread across reincarnations.
• The manah/“designou” verb chain (1:17; 4:6-8) requires absolute cross-chapter consistency; this is a structural translation requirement with no precedent in the Romans baseline, since no Romans doctrine depends on one verb repeating identically across four widely separated verses.
• “Voto” (vow, 1:16, 2:9) and “Seol” (Sheol, 2:2) both collide with well-entrenched, mainstream Brazilian devotional and cosmological vocabulary (vows to saints/Nossa Senhora Aparecida; Kardecist/Umbanda intermediate spirit-realms) that neither the Almeida nor Catholic Portuguese Bible traditions were built to address.

Risks

• Critical: Any softening of “ressurreição” toward “reencarnação,” or any ambiguous rendering of the fish-ordeal (1:17–2:10) that reads as cyclical death-and-rebirth rather than a once-for-all historical sign, would directly reinforce — rather than confront — the dominant rival doctrine in the target culture.
• Critical: Collapsing the four related-but-distinct Hebrew compassion roots (chesed, chanun, rachum, chus) into one flattened Portuguese word would destroy the deliberate literary echo between 4:2 and the book’s climactic, unresolved question in 4:10-11 — the doctrinal spine of “God’s Compassion for the Nations.”
• High: Inconsistent Portuguese rendering of “appointed” (manah) across its four occurrences would make God’s sovereignty over creation read as coincidence rather than deliberate personal action, opening a path to the Kardecist “lei de causa e efeito.”
• High: Unglossed use of “voto” or “Seol” would allow readers to unconsciously supply a Nossa Senhora intercessor or a Kardecist/Umbanda intermediate spirit-realm the Hebrew text does not contain.

Opportunities

• Jonah’s Old Testament setting gives this curriculum a uniquely strong platform to teach the resurrection/reincarnation contrast before the New Testament — reaching learners with lower NT literacy but high cultural fluency in reincarnation concepts.
• The book’s universal-compassion argument (a wicked Gentile empire spared) offers a sharper, more concrete pastoral illustration of “no distinction” universality than Romans’ more abstract argument, useful for catechizing audiences with little OT background.
• Because 15 of the curriculum’s 26 newly logged terms already carry Critical/High risk tags with fully drafted teaching notes, Phase 2 review routing can be automated with high confidence and minimal additional theologian triage.

Recommended actions

1. Lock assets/translation_memory.json (Jonah extension, this version) and load it alongside the unmodified Romans baseline before any Phase 2 segment translation begins.
2. Route all of Jonas 3:1-10 (the full core passage) to human theologian review as a block, regardless of individual segment term-risk, per the updated 12_ai_translation_requirements.md escalation rule.
3. Run the manah/“designou” cross-chapter consistency check as a batch validation pass after 1:17, 4:6, 4:7, and 4:8 have all been translated, not as a per-segment check.
4. Require the mandatory theologian-authored fencing note on every occurrence of “arrepender-se (aplicado a Deus)” and every occurrence of “voto,” before those segments can be marked approved.
5. Brief all Phase 2 native-speaker reviewers on the “who knows?” (quem sabe?) and “cast lots” (lançar sortes) risk profile before their first Jonah assignment, since both terms carry colloquial-register traps not present in the Romans baseline’s vocabulary.