Executive Summary — Galatians (English → Gujarati)

Why it matters

Galatians is Paul’s most doctrinally concentrated and polemically exact letter, and Gujarat is arguably the single most challenging Indian destination-language environment in which to translate it. The letter’s core argument — that righteousness comes by faith, not by law-observance, ritual practice, or self-effort — runs directly against the grain of Gujarat’s two dominant religious traditions: Vaishnav/Swaminarayan bhakti (where devotional merit, guru-blessing, and self-effacing devotion are the default religious idiom) and Jainism (where a soul’s own ascetic effort, not any grace-giver, produces right standing and eventual liberation). This Language Package extends the proven Romans baseline to Galatians’ full six chapters, with the core passage (2:15–21) as the theological anchor, ensuring every Phase 2 translation carries doctrinal precision rather than defaulting to the culturally “obvious” — and frequently forbidden — Gujarati vocabulary.

Key findings

• Full-book coverage confirmed. All six chapters of Galatians were analyzed section-by-section; no chapter or major section was silently skipped. Nineteen doctrines are tracked: the fourteen assigned curriculum doctrines plus five additional doctrines (Deity and Sonship of Christ, Resurrection of Christ, Unity of Believers in Christ, Election and Divine Calling, Deliverance from the Present Evil Age) that this book teaches across its full span.
• Doctrine-risk concentration is the highest of any curriculum built on this Language Package to date. Per assets/doctrine_risk_registry.json: 10 Critical, 8 High, 1 Medium, 0 Low doctrines — a combined 18 of 19 doctrines require mandatory human theologian review, with only one (Bearing One Another’s Burdens) downgraded to native-speaker review, and zero fully automated.
• Term-level risk is comparably concentrated. Per assets/bible_term_registry.json, approximately 33 Critical-risk and 37 High-risk terms (70 combined, out of ~116 tracked terms across inherited-and-new vocabulary) require mandatory theologian-level enforcement in every Phase 2 pass — nearly double the density of Critical/High terms relative to total vocabulary seen in the Romans baseline.
• One genuinely new risk category, absent from the Romans baseline, was identified: circumcision (સુન્નત). Its primary risk is contemporary socio-religious (association with Islamic practice in Gujarat), not purely doctrinal, requiring a review category — joint theologian and native-speaker sign-off — not previously needed.
• The single highest-priority enforcement point in the entire letter is the qualifier પવિત્ર in “પવિત્ર આત્માનું ફળ” (fruit of the Spirit, Gal 5:22-23). Dropping it converts Spirit-produced fruit into self-cultivated virtue — the exact collision point with both Hindu ātman-theology and Jain jīva-theology — and existing Gujarati devotional-song adaptations were found already making this exact error.

Risks

• ἐλευθερία (freedom) drifting to મુક્તિ/મોક્ષ. The single most intuitively “obvious” mistranslation in the letter: a translator’s natural first instinct for “freedom” is precisely the term the Romans baseline already forbids for “salvation” on Hindu-liberation-from-samsara grounds — now newly and explicitly extended to this adjacent doctrine.
• σάρξ (flesh) polysemy collapsing into one word. Without an enforced દેહ/શરીર (neutral) vs. દેહસ્વભાવ (sin-nature) split, Galatians 5–6’s Flesh-versus-Spirit argument degrades toward a Jain-tapa-style “the body is the enemy” reading Paul explicitly does not hold.
• σπέρμα (seed) flattened to વંશજ. Nearly every existing Gujarati Bible edition surveyed, including the formal-equivalence Old Version, defaults to the inherently plural વંશજ, silently erasing Paul’s argument that the promised “seed” is fulfilled specifically in the one person of Christ (Gal 3:16).
• ἀνάθεμα/κατάρα (curse) absorbed into folk curse-belief. Gujarat’s active cultural belief in mechanical/karmic curses (śāp, nazar, doshas) risks converting Paul’s personal, judicial divine pronouncements into impersonal occult-force statements.
• Incarnation and “Christ lives in me” (2:20) assimilated to avatar theology or Vedantic/Jain self-dissolution. Both risks were documented as already occurring in circulating oral discipleship material aimed at Hindu- and Jain-background hearers.
• Non-Trinitarian (Jehovah’s Witness-style) circulating renderings were identified as a live contamination risk for Christ’s-deity and Spirit’s-personhood passages (2:20, 4:4, 5:22-23) and must be actively screened against, not merely avoided by default.

Opportunities

• A single, non-negotiable core-passage rendering (Gal 2:15-21) can anchor the entire curriculum, exactly as Romans 1:16-17 anchored the prior package — five fixed phrases (justification, works of the law, faith in Christ, crucified with Christ, Christ lives in me) are now locked for verbatim reuse across every lesson document.
• The Romans package’s forbidden-substitution architecture transfers cleanly and strengthens with reuse: rather than building trust term-by-term from scratch, Galatians could extend nine already-proven Critical-risk safeguards (grace, salvation, righteousness, resurrection, incarnation, law, God, Holy Spirit, Son of God) with zero renegotiation, freeing review capacity to focus on this letter’s fifteen genuinely new Critical/High risk points.
• Gujarat’s comparative-theology landscape, while high-risk, is well-mapped: analysis/04_comparative_theology.md documents specific, well-understood collision points (bhakti prem-devotion, Jain nirjara/tapa, guru-paramparā, yuga-cycle cosmology) for all nineteen doctrines, giving Phase 2 reviewers precise, pre-identified checkpoints rather than open-ended cultural risk.
• Explanatory-note infrastructure is already specified, not left to translator improvisation, for the five terms with no ready Gujarati equivalent (seed, guardian/tutor, elemental spirits, circumcision, new creation) — reducing Phase 2 rework.

Recommended actions

1. Load assets/translation_memory.json and assets/bible_term_registry.json (Galatians editions) alongside the Romans baseline files before any Phase 2 segment processing begins; verify identical rendering for all shared terms.
2. Enforce the Galatians-specific forbidden-substitution list in analysis/12_ai_translation_requirements.md as a hard validation gate, especially the સ્વતંત્રતા-not-મુક્તિ rule for freedom and the પવિત્ર-never-dropped rule for fruit of the Spirit.
3. Route all 18 Critical/High doctrines to mandatory human theologian review at every occurrence (not first-occurrence-only), per the escalation rules; route circumcision segments to joint theologian-plus-native-speaker sign-off as a distinct review category.
4. Confirm the YouVersion version_id for the Gujarati Old Version lineage against the live catalog before Phase 2 build, per analysis/05_translation_landscape.md §5, and record it in translation memory.
5. Brief all Phase 2 human reviewers on the five documented cult/sect and folk-paraphrase rendering patterns already circulating in Gujarat (§4 of analysis/05_translation_landscape.md) so they can recognize, not merely avoid, these specific errors on sight.