Executive Summary — Colossians (English → Indonesian)

Why it matters

Colossians concentrates the highest density of Critical-risk Christological content of any curriculum processed against the Indonesian Language Package to date. Its core passage (1:15-20) makes four escalating claims — Christ as the visible image of the invisible God, supreme Creator and sustainer, head of the church and first in resurrection, and the one in whom the full deity of God bodily dwells and reconciles all things through his blood on a real cross — each of which lands directly on a specific, well-taught point of Sunni Islamic doctrine (tawhid, tanzih, aniconism, and the Quranic denial of the crucifixion) that dominates religious education across Indonesia’s ~87% Muslim-majority population. Getting this right is not a stylistic preference; it is the difference between faithfully transmitting the letter’s central claim and inadvertently agreeing with the very errors — Arian-adjacent created-Christ readings, idol-image readings, temporary-prophetic-inspiration readings — that the text itself was written to foreclose.

Key findings

• 32 doctrines identified in doctrine_risk_registry.json: 8 Critical, 18 High, 4 Medium, 2 Low — 26 of 32 (81%) require mandatory human theologian review on every occurrence, exceeding even the Romans baseline’s proportion of Critical/High doctrine.
• 122 terms now recorded in translation_memory.json version 2 (25 inherited unchanged from the Romans baseline + 97 new Colossians-specific terms): 16 Critical and 33 High risk (49 terms, 40%) requiring theologian-reviewed teaching notes at every occurrence; 51 Medium (native speaker review) and 22 Low (automated review).
• The single highest-risk phrase identified across both curricula processed to date is kepenuhan ke-Allahan secara jasmani (“the fullness of Deity, bodily,” Col 2:9) — a simultaneous, direct negation of both tawhid and tanzih, requiring mandatory theologian sign-off at every occurrence, with no permissible fencing or softening strategy.
• A specific, documented sectarian mistranslation technique (the Jehovah’s Witness New World Translation’s insertion of “yang lain”/“other” into Colossians 1:15-17) has been identified and is now a hard-coded automatic-flag rule in the Phase 2 instruction set, preventing an AI system from “smoothing” this passage into an Arian-adjacent reading.
• Household code material (3:18-4:1) and the false-teaching polemic (2:8-23) intersect live, present-day Indonesian religious and social practice — wifely-obedience law (taat kepada suami), circumcision (sunat/khitan) as a near-universal identity marker, and wali/ancestral-spirit veneration — more directly and more frequently than anything in the Romans baseline.

Risks

• Doctrinal collapse risk (highest): an imprecise rendering of “firstborn of all creation” (1:15) or “fullness of Deity, bodily” (2:9) could cause the translation to agree with rather than refute the Islamic view of Jesus as a created prophet — the opposite failure mode from most Critical terms in this project, where the risk is usually under-translation diluting a claim, not accidental agreement with a rival doctrine.
• Social/pastoral offense risk: an unqualified rendering of “self-made religion” (2:23, ethelothrēskia) as bare ibadah would be heard as Christianity condemning worship/devotion as a category — a serious, avoidable misrepresentation given ibadah’s centrality to Indonesian Islamic vocabulary.
• Household-code misreading risk: dropping the “in the Lord” qualifier or the husband’s counter-command to love (3:19) from “submit” (3:18, tunduk) would flatten Paul’s relativizing household ethic into the surrounding culture’s unqualified taat framework.
• Lexical collision risk: “kerajaan” used for created spiritual “dominions” (1:16) sits one word away from the baseline’s own “Kerajaan Allah,” and “tuan” (human master, 4:1) sits one letter/one capitalization away from “Tuhan” (divine Lord) — both require sustained typographic and contextual discipline across every Phase 2 segment.

Opportunities

• Colossians’ union-with-Christ material (2:11-13, 3:1-4) offers a doctrinally rich, culturally resonant teaching opportunity precisely because Indonesian has no competing framework to misdirect it — unlike most Critical terms in this project, “buried/raised with Christ” is a genuine positive import requiring sustained teaching rather than defensive fencing.
• The letter’s own internal polemic against angel-worship and self-made asceticism (2:16-23) gives the curriculum built-in, textually grounded language for addressing live syncretism concerns (grave veneration, spirit appeasement) pastorally, without requiring the curriculum to name any contemporary practice directly.
• Establishing “kepenuhan” as an identical rendering across 1:19 and 2:9 creates a reusable model for enforcing intra-book terminological echoes in future curricula.

Recommended actions

1. Route all segments touching the 8 Critical doctrines (Apostleship, Redemption/Forgiveness, Son’s Kingdom and Sonship, Deity of Christ as Image of God, Supremacy over Creation, Fullness of Deity Bodily, Reconciliation through the Cross, Union with Christ) to human theologian review with zero exceptions, per doctrine_risk_registry.json.
2. Deploy the “yang lain” automatic-flag rule and the “ibadah”-without-qualifier automatic-flag rule as hard gates in Phase 2 before any segment can be marked approved.
3. Enforce the tuan/Tuhan and kerajaan/Kerajaan Allah disambiguation checks as mandatory validation-rule line items, not discretionary reviewer judgment calls.
4. Brief theologian reviewers specifically on the reverse risk pattern unique to this book (under-claiming Christ’s deity rather than over-claiming it) so reviewers calibrate scrutiny toward accidental Arian-adjacent softening in 1:15-17, not only toward standard over-contextualization risks.
5. Carry forward the same YouVersion citation anchor (TB, version ID 306, book code COL) established here into all downstream Colossians Phase 2/3 materials for pipeline-wide consistency with the Romans baseline.

---

This summary is derived from doctrine_risk_registry.json, bible_term_registry.json, translation_memory.json v2, 08_core_glossary.md, 11_doctrine_analysis.md, 06_linguistic_gap_analysis.md, 05_translation_landscape.md, and 04_comparative_theology.md, all produced for this curriculum. It does not introduce new doctrinal or terminological decisions; all figures are traceable to those source documents.