Executive Summary — 2 Peter (English → Gujarati)

Why it matters

2 Peter packs an unusually high concentration of doctrinally load-bearing claims — the deity of Christ, the origin of Scripture, the certainty of Christ’s return, and final judgment — into just three chapters, and every one of these claims has a live, culturally comfortable, and theologically incompatible counterpart already circulating in Gujarat’s dominant Vaishnav/Swaminarayan bhakti and Jain religious environments. A fluent-sounding but doctrinally careless Gujarati rendering of this letter would not merely blur meaning — in several verses (1:4, 1:20, 2:19) it would actively hand the reader a plausible non-Christian reading of the text. This Phase 1 package extends the proven Romans Language Package with the additional guardrails 2 Peter specifically requires.

Key findings

• 80 tracked theological terms now sit in assets/translation_memory.json for this curriculum — 19 inherited verbatim from the Romans baseline, 61 new to 2 Peter — of which 51 (18 Critical + 33 High) require human theologian review on every occurrence, not just at first use.
• 20 doctrines are tracked in assets/doctrine_risk_registry.json, spanning all three chapters of the letter: 8 Critical, 8 High, 3 Medium, 1 Low — a total of 16 doctrines requiring human theologian review, 3 requiring native speaker review, and 1 sufficient for automated review only.
• The single highest-risk phrase in the entire letter is “partakers of the divine nature” (1:4) — it collides simultaneously with Advaita Hindu ātman=Brahman identity and Jain self-attained siddhahood, both live, prestigious frameworks in Gujarat; a technically approved Gujarati compound can still transmit false doctrine if used without its promise-mediated causal scaffolding.
• 2 Peter 2:19 (“promising them freedom”) creates a self-defeating-error risk unique to this letter: the natural Gujarati word for religious liberation (મુક્તિ) is the same word the Romans baseline already forbids for salvation — using it here would plant the forbidden term inside the very verse warning against a false liberation.
• 2 Peter 1:20 (“no prophecy… of private interpretation”) is flagged as the letter’s single most likely structural mistranslation, independent of any Hindu/Jain collision — a literal Gujarati rendering will naturally be misread as a rule restricting readers’ own Bible study rather than Peter’s actual claim about the prophets’ divine origin.
• The core passage (1:16–21) alone touches five Critical-risk doctrines (Deity of Christ, Sonship of Christ, Lordship of Christ, Inspiration of Scripture, Certainty of Christ’s Return), making it the theological center of gravity for the whole curriculum, consistent with its designation as the anchor passage.

Risks

• Syncretistic drift on “divine nature,” “godliness,” and “knowledge” (1:3-7, 3:18): Gujarat’s jñāna-mārga and Jain guṇasthāna/kevalgyān frameworks offer self-attainment models that a Sanskrit-rooted Gujarati vocabulary can slide into without any lexical “error” being visible to an unwarned reviewer.
• Cyclical-time contamination of judgment/eschatology vocabulary (2:5, 3:5-13): પ્રલય-adjacent terms risk assimilating the flood, the Day of the Lord, and the new heavens and new earth to Hindu pralaya or Jain beginningless-endless cosmic cycles, destroying the letter’s linear, once-for-all argument.
• Avatar-theology assimilation of the Parousia (1:16, 3:4, 3:12): આગમન alone carries no built-in uniqueness marker, and Dwarka’s location in Gujarat makes the repeatable-avatar-descent misreading unusually accessible.
• Forbidden-term leakage via translator instinct: સ્વામી (for lord/master), મુક્તિ (for freedom/salvation), અવતાર (for incarnation/return), and bare પ્રલય all remain culturally “available” first-instinct words that a translator not actively restrained by this package could reach for.

Opportunities

• Genuine native strengths exist and should be leveraged deliberately: પ્રત્યક્ષદર્શી (eyewitness) has no competing sacred-text category in Gujarati religious vocabulary (neither shruti nor Agama tradition is eyewitness-based), making 1:16-18’s apologetic argument unusually persuasive once properly anchored.
• The virtue-chain (1:5-7) offers a natural teaching bridge: its list-structure superficially resembles the Jain guṇasthāna staged-progression model familiar to many readers, which — properly fenced with mandatory grace/faith anchoring — can be used pedagogically to highlight the contrast (grace-given vs. self-earned) rather than only to guard against it.
• This package’s dual-file translation memory architecture (baseline Romans + 2 Peter extension) sets a reusable pattern for future NT curricula in this Language Package, without requiring re-litigation of already-settled Critical terms (પ્રભુ, પરમેશ્વર, ઉદ્ધાર, કૃપા, ન્યાયીપણું, પવિત્ર આત્મા).

Recommended actions

1. Lock the four sentence-structure-level scaffolding rules identified in the Linguistic Gap Analysis — divine nature (1:4), private interpretation (1:20), Parousia singularity (throughout), and false-freedom containment (2:19) — into Phase 2 segment validation as mandatory checks, not optional flags.
2. Route all 16 theologian-tier doctrines and 51 Critical/High terms through mandatory human review before any Phase 2 output is marked approved; do not allow automated or native-speaker-only sign-off to substitute for theologian review on these items.
3. Brief all Phase 2 translators explicitly on the five known cult/sect and syncretistic-tract rendering patterns documented in the Translation Landscape analysis (JW restructuring of 1:1, bhakti-framed divine-nature paraphrase, tapa-framed virtue-list paraphrase) so these are recognized and rejected on sight, not only caught in review.
4. Confirm the Gujarati O.V. YouVersion numeric version ID before Phase 2 tooling setup begins, and lock it into translation_memory.json as youversion_version_id per the citation convention already fixed in this package.
5. Carry the core passage (1:16-21) as the first Phase 2 batch, since its five-doctrine density makes it the highest-value proving ground for the validation pipeline before scaling to the full three-chapter book.