Executive Summary — Titus | English → Odia Language Package

Why it matters

Titus is a compact, three-chapter letter that packs an unusually high concentration of doctrinally load-bearing vocabulary into a small space — and Odisha’s Jagannath-Vaishnav devotional environment supplies false-friend collisions for nearly every one of its distinctive terms. The core passage, Titus 2:11–3:8, alone carries 4 of the letter’s 5 Critical-risk doctrines: the deity of Christ (2:13’s “great God and Savior”), salvation by grace not works (3:4–5), regeneration by the Holy Spirit (3:5–6), and justification and heirship (3:7). Get this passage wrong in Odia and the letter’s central argument — that God saves by mercy, not merit, through a one-time, permanent, Spirit-wrought new birth — collapses into something an Odia reader could mistake for pilgrimage-and-ritual religion or the rebirth cycle it is written to contradict.

Key findings

• 31 of 57 registered Titus terms (54%) are Critical or High risk (11 Critical, 20 High), requiring mandatory human theologian review at every occurrence — a notably higher density than typical epistolary material, driven by Titus’s compressed, formula-heavy style.
• 16 of 19 named doctrines (84%) require theologian-level review (5 Critical + 11 High); only 2 Medium-risk doctrines (Faithful Saying formula; Church Discipline) route to native-speaker review, and just 1 Low-risk doctrine (closing greetings) is automated-only.
• παλιγγενεσία (“regeneration,” Titus 3:5) is the single sharpest false-friend risk identified across this entire pipeline to date — its literal Greek etymology, “again-birth,” is closer to Odia’s reincarnation vocabulary (ପୁନର୍ଜନ୍ମ) than any term the Romans baseline had to guard, yet teaches the opposite doctrine. The required coinage, ନୂତନ ଜନ୍ମ, is a neologism-by-avoidance, not a pre-existing safe word.
• Titus introduces σωτήρ (“Savior”) as a new Critical term absent from the Romans baseline — used six times of both Father and Son, with zero occurrences of κύριος (“Lord”), meaning the letter’s Christological center of gravity shifts entirely to a title the baseline never had to define.
• Three separate Odisha-specific collisions stack inside one verse (3:5): “washing” risks evoking Puri’s Snāna Pūrṇimā festival; “regeneration” risks evoking reincarnation; “renewal” risks blurring into “regeneration” if not kept distinct. No other single verse in the curriculum carries this much concentrated risk.
• “Good works” (ἔργον) appears six times (1:16; 2:7,14; 3:1,8,14) and must never use the natural Odia word କର୍ମ, which would reintroduce a karma/merit-accumulation framework the letter’s own grace-not-works argument (3:5) explicitly denies.

Risks

• Deity of Christ (2:13): A split rendering of “our great God and Savior Jesus Christ” would read as two beings, not one — and Odisha’s familiarity with the worshipped Puri triad (Jagannath, Balabhadra, Subhadra) makes this misreading feel locally natural rather than obviously wrong.
• Grace-vs-works negation (3:5): Ambiguous negation scope in Odia word order could imply self-performed righteous works are a partial, contributing cause of salvation alongside mercy — precisely what the verse denies.
• Regeneration/washing cluster (3:5-6): A single dropped qualifier (ଆତ୍ମିକ before ସ୍ନାନ, or ପୁନର୍ଜନ୍ମ substituted for ନୂତନ ଜନ୍ମ) would produce a fully syncretistic misreading of the letter’s most important verse.
• Grace-as-trainer agency (2:11-12): If sentence structure lets the believer, not grace, appear as the self-initiating agent of “training” and “self-control,” the doctrine collapses into Odia Vaishnav ascetic self-discipline (tapasya/sādhanā).
• Elder qualifications (1:5-9): Excerpted or abbreviated use of “elder” without its full character-and-teaching qualification list risks defaulting to Puri’s hereditary pandaa/sevayat/daitapati temple-functionary model.
• Myth/genealogy (1:14; 3:9): The natural Odia word for “myth” (ପୁରାଣ) would gratuitously insult the Odia Bhagavata Purana devotional tradition already protected under the Romans baseline; genealogy disputes risk reading as caste-lineage (gotra/kula) commentary.
• Submission language (2:5,9; 3:1): Risks being absorbed into Odisha’s pervasive caste-based hierarchical subordination patterns unless kept explicitly tied to its grace-motivated grounding (3:3-7).

Opportunities

• The Romans baseline’s forbidden-substitution infrastructure (ଠାକୁର, ମୋକ୍ଷ/ମୁକ୍ତି, ପୁନର୍ଜନ୍ମ, ଅବତାର, ଶକ୍ତି, ପରମାତ୍ମା) transfers directly and requires no rework — Titus extends rather than replaces it, giving Phase 2 a stable, already-validated foundation.
• Titus’s compact size and formulaic style (repeated σώφρων vocabulary, the “faithful saying” formula, the three-appearings structure) make it unusually well-suited to consistency validation rules once the core terms are fixed — a single correct decision at first occurrence (e.g., 1:8’s self-control framing) propagates cleanly through five later occurrences.
• The letter’s own grace-not-works argument (3:4-5) gives translators a built-in, textually-explicit tool for directly confronting the pipeline’s second-sharpest doctrinal risk (salvation by merit/pilgrimage) rather than needing to import that corrective from outside the text.
• Household-code and civic-submission material (2:1-10; 3:1-2), read alongside its own grace-grounding (3:3-7), offers a natural teaching bridge for explaining voluntary, gospel-motivated conduct against the backdrop of Odisha’s caste-hierarchical social structures.

Recommended actions

1. Load the Titus translation_memory.json extension alongside (never instead of) the Romans baseline before any Phase 2 segment work begins.
2. Treat Titus 2:13 and Titus 3:5 as indivisible, single-unit theologian-review passages rather than segmenting them clause-by-clause.
3. Apply the grace-as-grammatical-subject rule to every occurrence of “trains” (2:12) and the grace-fruit framing to every occurrence of “self-controlled” (1:8; 2:2,4,5,6,12) at first pass, not as a later correction.
4. Enforce the six new Critical/High forbidden substitutions (ପୁନର୍ଜନ୍ମ for regeneration, ଯୁଗ, କର୍ମ, ପୁରାଣ, unqualified ସ୍ନାନ, unqualified ଭକ୍ତି) with the same non-negotiable validation status as the baseline’s own forbidden list.
5. Route all 16 Critical/High doctrines to human theologian review per doctrine_risk_registry.json; do not allow automated or native-speaker-only approval to substitute for theologian sign-off on any of these.
6. Never excerpt the elder-qualification list (1:5-9) or the good-works terms (κ ἔργον family) in derivative lesson material without their full qualifying/framing context intact.

---

See 06_linguistic_gap_analysis.md, 07_semantic_analysis.md, 08_core_glossary.md, 11_doctrine_analysis.md, and the Titus assets/bible_term_registry.json / assets/doctrine_risk_registry.json for full supporting detail behind every finding above.