Executive Summary

Why it matters

Romans is the theological backbone of the New Testament, and Assamese carries a distinctive syncretism risk unlike any other North Indian language in this pipeline: the Ekasarana Dharma tradition founded by the 15th-16th century reformer Srimanta Sankardev is itself a monotheistic, anti-caste, anti-idol bhakti movement centered on single-minded devotion (naam) to one God — which means the terms that sound most naturally “monotheistic” and “devotional” in Assamese are often the ones most saturated with a specific, well-developed Vaishnavite theology of avatar-descent, guru-succession, and devotional merit. Nine of the registry’s central doctrines (incarnation, deity of Christ, sonship of Christ, messianic promise, apostleship, sainthood, church, salvation, and grace) each have a ready natural-sounding Ekasarana or wider Assamese Vaishnavite equivalent that must be deliberately rejected or carefully distinguished.

Key findings

• The registry tracks 40 doctrines across Romans 1-16; 30 require mandatory human theologian review before any translated segment ships (7 Critical, 23 High).
• Incarnation is the single highest-risk term in this Language Package specifically because Ekasarana Dharma’s own theology holds Krishna as Vishnu’s supreme avatar descending to restore dharma — a genuinely monotheistic system that still runs on repeatable avatar-descent, which is a harder case than the generic “many gods” Hinduism a translator might expect.
• Assamese has an existing, continuously used Christian Bible translation tradition dating to the 1819 Serampore Mission New Testament and the 1836 American Baptist Mission at Sadiya, which gives this Language Package settled precedent for its highest-risk proper names and core terms (ঈশ্বৰ, যীচু, প্ৰভু, খ্ৰীষ্ট).
• Ekasarana Dharma’s own institutions (নামঘৰ prayer-houses, Satra monastic centers, guru-lineage succession) supply plausible-sounding but doctrinally wrong renderings for “church,” “apostle,” and “saints” that this Language Package explicitly rejects.

Risks

• Avatar-incarnation collapse: অৱতাৰ is the most dangerous single rejected term in this registry — unlike in generic Hindu contexts, an Assamese reader from an Ekasarana background has a fully worked-out theology of why a deity would take repeated bodily form, and দেহধাৰণ must be actively distinguished from it, not just substituted for it.
• Guru/apostle collapse: রendering “apostle” or “church leadership” with guru-lineage vocabulary (গুৰু, Satradhikar-style succession) would import an institutional authority-transmission model Romans does not describe.
• Institutional-borrowing risk: নামঘৰ (prayer-house) and সন্ত/ভক্ত (realized devotee) are attractive because they are “already religious and already monotheistic-sounding” words, which makes them more tempting mistranslations than an obviously polytheistic term would be.

Opportunities

• Because Ekasarana Dharma already rejects idol worship and caste hierarchy and insists on devotion to one God by name, this curriculum’s monotheism and its “no distinction, Jew and Gentile” universalism will land as recognizable rather than foreign — the challenge is precision within a receptive audience, not persuading a hostile one.
• The 1819-onward Assamese Bible translation tradition removes ambiguity for the highest-risk proper names and core terms, unlike languages with no settled Christian translation precedent.

Recommended actions

• Route every Critical and High risk segment (30 of 40 doctrines) through human theologian review before publication; do not allow automated-only review to touch these terms.
• Brief every reviewer specifically on Ekasarana Dharma’s avatar theology and guru-succession model before they review Incarnation, Deity of Christ, Apostleship, or Church segments — a reviewer briefed only on generic Hinduism will miss why these specific Assamese words are wrong.
• Reuse this Language Package’s translation_memory.json for every Romans lesson in Assamese rather than re-deriving terms per document, per the two-phase pipeline design.