Linguistic Gap Analysis — Gospel of Mark → Ndebele

This document executes Phase 1 Step 8 for the Mark curriculum. It builds directly on analysis/07_semantic_analysis.md, analysis/08_core_glossary.md, and assets/bible_term_registry.json, and never contradicts the baseline Romans translation_memory.json / bible_term_registry.json / doctrine_risk_registry.json. Established renderings are reused exactly; no baseline term is re-litigated here except to note how Mark’s narrative context adds new pressure to it.

Coverage is full-book: every chapter of Mark, 1–16, is addressed either through load-bearing vocabulary analysis or explicit “reviewed, no new gap” notation. The core passage (Mark 10:35–45) anchors the analysis but is not its boundary.

---

1. Method

For each of the eight curriculum doctrines, this analysis inventories: (a) which Ndebele terms are available (already attested in Nguni Bible tradition or the baseline package), (b) the specific weaknesses each available term carries when it meets Ndebele traditional religious vocabulary or social structures, and (c) a recommended strategy — reuse as-is, reuse with mandatory translator note, coin a compound phrase, or transliterate. This mirrors the structure of the baseline package’s own notes fields, extended for Mark’s distinctive vocabulary (Son of Man, ransom, servanthood-reversal, unclean spirits, the Messianic Secret’s veiling language).

---

2. Doctrine Vocabulary Matrix

2.1 Jesus as the Suffering Servant and Son of God

Available Ndebele terms	Weaknesses	Recommended strategy
iNdodana kaNkulunkulu (Son of God, baseline)	None inherent; risk only arises if merged carelessly with Son of Man or read as one honored ancestor-spirit among others.	Reuse exactly. Reinforce distinctiveness from iNdodana yoMuntu at every joint occurrence (esp. Mark 1:1, 1:11, 9:7, 15:39).
iNdodana yoMuntu (Son of Man, new coinage in core glossary)	Ndebele has no existing fixed Bible term for this Danielic title; a literal “child of a person” reading collapses the exalted-authority sense entirely into ordinary humanity.	Coin and fix as iNdodana yoMuntu (already fixed in Table 2 of Core Glossary); require identical rendering at all 14 Markan occurrences (2:10, 2:28, 8:31, 8:38, 9:9, 9:12, 9:31, 10:33, 10:45, 13:26, 14:21 [x2], 14:41, 14:62) to preserve the veiled-then-revealed arc.
udumo (glory, baseline)	Ndebele royal-court usage of “udumo” (renown of a king/chief) risks reading Jesus’ glory as court-honor rather than divine radiance disclosed only through suffering.	Reuse, but pair with explicit teaching note at 8:38, 10:37, 13:26 that glory is reached only through the cup/cross, not political ascendance.
No available term for “suffering servant” as a compound doctrinal label (it is a theological category, not a single lexical item in Mark’s Greek)	Risk of importing izinyanga/traditional-healer “suffering intermediary” associations if paraphrased loosely (e.g. “umuntu ohluphekelayo abanye,” “one who suffers for others,” could echo ritual substitutionary suffering performed by a healer or medium on a client’s behalf).	Do not coin a single fixed term. Render contextually using existing fixed vocabulary already carrying the needed weight: isikhonzi (servant), inhlawulo (ransom), isiphambano (cross), ukuhlupheka (to suffer, generic verb, Low risk) — combined through teaching material rather than a single new noun, to avoid manufacturing a new syncretism-prone term.

2.2 The Kingdom of God Breaking In

Available Ndebele terms	Weaknesses	Recommended strategy
umbuso kaNkulunkulu (baseline)	Strong historical weight from the 19th-century Ndebele umbuso (Mzilikazi/Lobengula); risk of the kingdom being read as ethnic/political restoration rather than God’s present-and-future reign.	Reuse exactly; every occurrence in Mark (1:15, 4:11,26,30, 9:1,47, 10:14,15,23-25, 12:34, 14:25, 15:43) should carry contextual framing (already/not-yet) rather than a new qualifier term.
imfihlo yombuso kaNkulunkulu (mystery of the kingdom, new)	“Imfihlo” (secret/hidden thing) is also the ordinary word for occult or diviner-guarded secret knowledge accessed through izangoma.	Reuse the coined compound, but a mandatory translator note at 4:11 must clarify this “imfihlo” is revealed teaching disclosed by Jesus to disciples, never knowledge obtained through ritual consultation.
umzekeliso (parable, new)	Risk of flattening to simple moral fable (a genre already familiar in Ndebele oral tradition, izinganekwane) losing the revealing/concealing dual function central to Mark 4.	Reuse the coined term but require a teaching note distinguishing it from a folk-tale (inganekwane), which resolves cleanly and teaches a moral, versus a parable, which can deliberately obscure to outsiders.
indlu yomkhuleko (house of prayer, new)	Low weakness; ties naturally to baseline “prayer” doctrine already established.	Reuse as coined; reinforces the direct-access-to-God theme already central to the baseline package.

2.3 Faith and Discipleship amid Fear

Available Ndebele terms	Weaknesses	Recommended strategy
ukholo (faith, baseline)	None inherent to the word itself, but Mark repeatedly juxtaposes ukholo directly against ukwesaba (fear) as competing responses (4:40-41; 5:36) — a structural feature the baseline glossary did not need to handle.	Reuse exactly; ensure translator notes at paired faith/fear verses flag the deliberate contrast so it is not smoothed over in translation.
ukwesaba (fear, new)	Single Ndebele word spans terror (rebuked, 4:40) and reverential awe (commended, 4:41; 16:8’s ambiguous ending). Collapsing both senses into one undifferentiated term risks Mark’s central “wrong fear vs. right awe” tension disappearing.	Reuse ukwesaba as the base term (no viable alternative exists that cleanly splits the two senses in standard Ndebele), but require a translator note at every occurrence disambiguating which sense is intended, with intensifiers/qualifiers (e.g., “ukwesaba okukhulu kakhulu,” “great fear/awe”) used contextually to signal the awe sense where appropriate.
ukholo lwakho lukusindisile (“your faith has saved/healed you,” new)	σώζω’s dual healing/salvation sense has no single Ndebele verb spanning both; usindisa (baseline “usindiso” root) risks either over-claiming full salvation for a physical healing, or (if a lesser healing verb like “ukusilisa” were substituted) under-claiming and losing the salvation echo entirely.	Retain -sindisa root as coined (“lukusindisile”) to preserve the salvation echo, but require a translator note each occurrence (5:34; 10:52) clarifying that the physical healing is a sign pointing toward, not identical with, full usindiso.

2.4 The Necessity of the Cross

Available Ndebele terms	Weaknesses	Recommended strategy
isiphambano (cross, new)	No prior fixed Bible term in the baseline package (Romans discusses the cross’s theology but the object itself was not glossed). Risk of reduction to generic “hardship” metaphor, or assimilation to Ndebele motifs of ancestral suffering/affliction requiring ritual response.	Fix as isiphambano (the standard Nguni Bible-tradition term for the execution instrument); reuse identically at 8:34, 10:21, 10:38 (a “cup” not “cross” — see below), 15:21-32.
kufanele / kumele (necessity, δεῖ, new)	Both verbs can carry a fatalistic “it is destined” reading resonant with ancestral-fate thinking (a concern the baseline flagged for “election” and “providence”).	Reuse coined choice but pair each occurrence (8:31, 9:31 implicit, 13:7,10) with framing language anchoring necessity in God’s purposive redemptive plan, not impersonal destiny.
indebe (cup, new)	Literal vessel term only; the OT wrath/judgment metaphor (Ps 75:8; Isa 51:17) has no independent life in Ndebele outside of biblical teaching.	Reuse literal term but require sustained catechetical support (not a lexical fix) — this is a metaphor-transfer problem, not a mistranslation risk; flag for teaching-material coordination in Phase 2.
ukubhabhathizwa (baptism-as-suffering, new)	Very high default risk: readers will assume the sacramental water-rite sense of ukubhabhathizwa learned from other church contexts, missing Jesus’ metaphorical “immersion in suffering” at 10:38-39.	Reuse the standard baptism term (no viable alternative exists without inventing an obscure new word, which would itself be a fencing problem) but mandate a translator note at every metaphorical occurrence distinguishing it explicitly from the literal water rite.
ukuzidela (deny self, new)	Moderate risk of reading as generalized self-sacrifice/asceticism disconnected from following Christ specifically.	Reuse coined term; anchor to “and follow me” clause in the same verse (8:34) so self-denial is never detached from its Christ-directed object.

2.5 Servanthood versus Worldly Greatness

This is the doctrine cluster of the core passage (Mark 10:35–45) and carries the highest concentration of new, high-stakes vocabulary in the whole book.

Available Ndebele terms	Weaknesses	Recommended strategy
isikhonzi (servant/διάκονος, new)	The root -khonza is the ordinary Ndebele verb for religious devotion/service — including, in traditional register, devoted service to amadlozi or ritual service to a chief.	Reuse coined term; every occurrence (9:35, 10:43) requires anchoring language: service modeled on Christ and directed toward people, flowing from right relationship with uNkulunkulu, never appeasement service directed to an intermediary spirit.
isigqili (slave/δοῦλος, new)	Correctly carries full rights-less weight, but risk of being read as endorsing literal servile social status, or evoking Zimbabwe’s own history of forced/indentured labor rather than Jesus’ voluntary theological paradox.	Reuse coined term with mandatory teaching note: the greatest becomes isigqili voluntarily, mirroring Christ’s own self-giving (v. 45), not a description of anyone’s actual social condition.
ukubusa ngamandla (lord it over/κατακυριεύω, new)	Built on the same root as the Critical baseline term iNkosi; this is Mark’s own deliberate wordplay (the Gentile rulers’ κατακυριεύουσιν vs. Christ’s servant-kyrios-ship), and it lands directly on the pre-flagged Ndebele royal-title collision (Mzilikazi/Lobengula’s iNkosi).	This is not a gap to fence but a feature to exploit: require a translator note making the wordplay explicit — Jesus rejects the “ukubusa ngamandla” style of rule that a historical iNkosi could exercise, redefining true lordship as servanthood. This must be coordinated with the “lord it over” and “iNkosi” registry entries as a single teaching unit, not translated in isolation.
inhlawulo (ransom/λύτρον, new) — CRITICAL	The single highest collision risk in the entire Mark package. Inhlawulo is an active, currently-practiced Ndebele customary-law institution: a family-to-family compensation payment (e.g., for pregnancy out of wedlock, injury, or death), typically paid by the guilty party’s kin to the offended party’s kin, in cattle or cash. Mark 10:45’s λύτρον runs in the opposite structural direction: the offended party’s own Son voluntarily pays with his own life for the guilty, at his own initiative, once for all.	Retain inhlawulo — no better lexical option exists, and avoiding the term entirely would sacrifice all cultural resonance the passage’s own economic metaphor needs. But mandate a translator note at every occurrence (10:45 and any cross-reference in teaching material) that explicitly inverts the customary direction: not the guilty paying the offended, but the offended party’s own Son paying for the guilty.
esikhundleni sabanengi (“in place of many”/ἀντὶ πολλῶν, new) — CRITICAL	Risk of softening to a general-benefit phrase (“ngenxa yabanengi,” for the sake of many), which would preserve gospel benefit language while quietly dropping substitution — the doctrinal core of the ransom saying.	Fix as esikhundleni sabanengi (lit. “in the place/stead of many”) and treat any drift toward “ngenxa ya-” phrasing as a Critical validation failure in Phase 2 QA.
owokuqala / owokucina (first/last, new)	Low inherent risk; straightforward paired terms.	Reuse; theological weight carried by context (status reversal), not lexical choice.
ababusi (rulers of the Gentiles, new)	Risk of being read as veiled commentary on the historical Ndebele royal court’s own ababusi/indunas, or on contemporary Zimbabwean political rulers, rather than the biblical Gentile referent.	Reuse coined term but keep surrounding phrasing tightly anchored to “abezizwe” (Gentiles, baseline) so the referent is unambiguous.

2.6 The Messianic Secret

Available Ndebele terms	Weaknesses	Recommended strategy
uMesiya (baseline)	None inherent; the doctrine’s difficulty is narrative-structural (why Jesus repeatedly silences messianic acclamation), not lexical.	Reuse exactly; ensure silencing-command verbs (e.g., at 1:25,34,44; 3:12; 5:43; 7:36; 8:30; 9:9) are translated with consistent commanding force so the pattern is visible to readers across the book.
iNdodana yoMuntu (see 2.1)	Its very function as a veiled self-title depends on Ndebele readers not over-clarifying it into “Messiah” or “Son of God” prematurely in translation.	Reuse the fixed coinage without doctrinal glossing inside the text itself; let the narrative’s own progressive unveiling (culminating at 14:62) do the theological work, with commentary-layer notes (not in-text expansion) carrying the explanation.
Nginguye (“I am”/ἐγώ εἰμι, new) — CRITICAL	Risk of flattening to a merely reassuring “it’s just me” (6:50’s narrative surface sense), losing the Exodus 3:14 divine self-naming echo that resolves at the trial (14:62).	Fix as Nginguye at both occurrences; require a translator note at 14:62 explicitly linking back to 6:50 so the resolution arc is visible to careful readers, and at 6:50 flagging the divine-echo reading beneath the surface reassurance.
abaphrofethi bamanga / oKristu bamanga (false prophets/false messiahs, new)	Sharpens rather than weakens the baseline’s prophet/diviner distinction; risk is under-emphasizing that these are deliberate deceivers, not merely mistaken teachers.	Reuse coined terms; teaching note at 13:22 should stress intentional deception, distinguishing from honest misunderstanding.

2.7 Jesus’ Authority over Sickness, Sin, and Nature

Available Ndebele terms	Weaknesses	Recommended strategy
igunya (authority/ἐξουσία, new)	Must be held apart both from amandla kaNkulunkulu (baseline, raw miracle-working power/capability) and from spiritual power traditionally claimed by izinyanga (healers) and izangoma (diviners), who are widely understood in Ndebele communities to possess real, if lesser, spiritual capability.	Reuse coined term consistently (1:22,27; 2:10; 3:15; 11:28-29,33) with a standing teaching note: igunya is Jesus’ inherent right to act, not power gained through training, ritual, inheritance, or mediumship.
umoya ongcolileyo / idimoni (unclean spirit/demon, new) — CRITICAL	The single most structurally dangerous vocabulary gap in Mark. Ndebele traditional cosmology’s spirits (amadlozi) are predominantly ancestral, not uniformly hostile, and are approached — not simply opposed — for guidance and mediation. A reader’s default cosmological map has no clean category for a spirit that is personal, evil, and to be simply expelled with no possibility of ongoing relationship or appeasement, which is exactly Mark’s category.	Reuse coined terms umoya ongcolileyo and idimoni (transliterated loanword, avoiding any native term with ancestral-spirit associations); require a standing teaching note at every occurrence (1:23-27,32-34; 3:11,22,30; 5:1-20; 6:7,13; 7:25-30; 9:17-29) reiterating the categorical difference from amadlozi: hostile to God, not ancestral, never to be consulted or appeased, only ever expelled by Jesus’ authority.
ukuthethelelwa kwezono (forgiveness of sins, new)	Risk of reading as social reconciliation or ritual appeasement of an offended party (human or ancestral) rather than God’s own authoritative act.	Reuse coined term; anchor tightly to Jesus’ direct exercise of the prerogative without priestly or ancestral intermediary (2:5-10), reinforcing the baseline’s Christ-as-sole-mediator doctrine.
isihogo (Gehenna, new)	Risk of conflation with a traditional ancestral realm of the dead or a temporary post-death state, rather than final, singular divine judgment.	Reuse coined term with teaching note distinguishing final judgment from any traditional afterlife category.
ukuvusa / ukuvuswa (temporary raising, new) vs. ukuvuka kwabafileyo (baseline resurrection)	If these are allowed to blur, the doctrinally vital distinction between Jairus’s daughter’s this-worldly restoration (ch. 5) and Christ’s unique eschatological resurrection (ch. 16) collapses — and the blur would reinforce, not resist, a cyclical amadlozi-style return-to-life reading.	Maintain strict lexical separation: ukuvusa/ukuvuswa only for this-worldly, mortal, repeatable “raisings” (5:41-42; also applicable to Lazarus-type events elsewhere in the Gospels, for consistency across future curricula); ukuvuka kwabafileyo reserved exclusively for Christ’s own resurrection. Flag any drafting deviation as a Critical validation failure.

2.8 The Ransom for Many

(Overlaps substantially with 2.5 above; treated separately here per curriculum’s doctrine list, since Mark 10:45 is the core passage’s climactic verse.)

Available Ndebele terms	Weaknesses	Recommended strategy
inhlawulo	See 2.5 above — CRITICAL, customary-law collision.	See 2.5.
esikhundleni sabanengi	See 2.5 above — CRITICAL, substitution-vs-benefit collision.	See 2.5.
igazi lesivumelwano (blood of the covenant, new)	Collides with active Nguni blood-oath/alliance-sealing customs between human parties (e.g., blood covenants between families or age-mates); Christ’s covenant-blood is a once-for-all divine-initiative sacrifice, not a mutual pact between two human parties.	Reuse coined term; mandatory teaching note at 14:24 distinguishing a mutual human blood-pact from God’s own unilateral, once-for-all redemptive act.
ukunikelwa / ukukhaphela (hand over/betray, παραδίδωμι, new)	A single Greek verb bears both Judas’s culpable betrayal and God’s sovereign redemptive purpose; collapsing to only one Ndebele verb risks resolving a tension Mark deliberately leaves open.	Use ukukhaphela where culpable betrayal is clearly in view (Judas, 14:10-11,18,21,42-44) and ukunikelwa (passive, “to be handed over”) where the text leaves the agency ambiguous or divine purpose foregrounded (9:31; 10:33); flag both choices in translator notes so the tension is visible rather than silently resolved either direction.

---

3. Chapter-by-Chapter Coverage Confirmation

Every chapter of Mark has been reviewed for load-bearing vocabulary. Chapters not listed with new terms in Section 2 are confirmed reviewed below with no additional gap-analysis items beyond those already logged in the Core Glossary.

Ch.	Coverage status
1	Covered — gospel, God, Holy Spirit, holy/unclean contrast, authority (igunya), unclean spirit/demon, Sabbath background.
2	Covered — sin, forgiveness of sins, law, Sabbath, Son of Man (2:10, 2:28 “lord of the Sabbath” — early anticipation of 10:42-45 wordplay).
3	Covered — Beelzebul, blasphemy against the Holy Spirit, unclean spirit/demon, election root (the Twelve chosen).
4	Covered — parable, mystery of the kingdom, kingdom of God, faith/fear pairing, peace (“Peace, be still”).
5	Covered — faith/healing overlap, unclean spirit/demon, Legion, raising (temporary) vs. resurrection distinction, peace (“go in peace”).
6	Covered — “I am” self-declaration (6:50), prophet (background, rejection at Nazareth), authority (mission of the Twelve), unclean spirit.
7	Covered — heart, Corban, law/tradition contrast, holy/unclean (defilement) reframed around the heart, unclean spirit (Syrophoenician woman’s daughter).
8	Covered — Messiah (Peter’s confession), Son of Man, deny self, cross, necessity (δεῖ), life/soul (ψυχή), glory.
9	Covered — Transfiguration, Son of Man, Father/Abba background (Gethsemane anticipation), Gehenna, salt, eternal life background, first/last, unclean spirit (epileptic boy), election (“elect” background).
10	Core passage chapter. Covered exhaustively — ransom, substitution, cup, baptism-as-suffering, servant, slave, lord-it-over, authority, eternal life, first/last, faith/healing overlap, Son of David, rulers of the Gentiles, kingdom of God, faith.
11	Covered — Hosanna, house of prayer, Son of David background, authority (Temple confrontation), faith (mountain-moving saying).
12	Covered — cornerstone, love/Greatest Commandment, resurrection (Sadducees’ question — distinguished from ch.5’s raisings), David, Messiah (son of David question), glory (background), giving/widow’s mite (low-risk, not separately tabled).
13	Covered — false prophets/false messiahs, election (“the elect”), Son of Man, kingdom of God (future coming), gospel (proclaimed to all nations), prophet background.
14	Covered — Abba, Father, blood of the covenant, hand over/betray, Son of Man, thanksgiving, cup (Gethsemane), “I am” (trial), covenant, Passover background (low-risk, folded into covenant/blood entries).
15	Covered — cross, King of the Jews, cry of dereliction, temple veil torn, Son of God (centurion’s confession, climactic), Lord/iNkosi (mocking-then-true title).
16	Covered — resurrection (ukuvuka kwabafileyo, distinct from ch.5’s raisings), gospel (proclamation to all creation), fear (ambiguous ending, 16:8).

Conclusion: full 16-chapter coverage confirmed; no chapter is silently omitted.

---

4. Missing Vocabulary vs. Crowded Semantic Neighborhoods

4.1 Genuinely Missing Vocabulary (no prior fixed Ndebele Bible term)

These required new coinage in the Core Glossary because no established Nguni Bible-tradition term or baseline entry covers them:

• Son of Man (iNdodana yoMuntu) — Mark’s dominant Christological title has no prior fixed form in the baseline Romans package, which did not need it.
• Ransom / substitution (inhlawulo; esikhundleni sabanengi) — entirely new economic-theological vocabulary, with inhlawulo pre-existing only as a secular customary-law term repurposed for doctrinal use.
• Baptism-as-suffering-metaphor — no distinct term exists apart from the ordinary sacramental word; the metaphor is a gap in concept-mapping, not vocabulary.
• Cross (isiphambano) — the physical object/event was not glossed in the Romans package’s more abstract treatment of the crucifixion’s theology.
• Unclean spirit / demon terminology as a sustained narrative category (rather than an isolated doctrinal aside) — the baseline touched intercession/amadlozi risk but never needed to sustain a repeated exorcism narrative across many chapters.
• Transfiguration (ukuguqulwa isimo) — a one-time narrative event with no lexical precedent.

4.2 Crowded Semantic Neighborhoods Requiring Fencing

These are cases where Ndebele already has a word, but that word is heavily populated by adjacent, doctrinally incompatible senses that must be actively fenced off in translator notes and teaching material:

1. umoya (“spirit,” in all its compounds) — the single most crowded neighborhood in the whole package. It must do service for: uMoya oNgcwele (Holy Spirit, divine person), umoya ongcolileyo (unclean/demonic spirit, hostile created being), and — nowhere used in translation but ever-present in the reader’s own cosmology — amadlozi (ancestral spirits, generally benevolent or ambivalent, approached for mediation). Fencing strategy: never let “umoya” stand unqualified in any doctrinally load-bearing sentence; always pair with -Ngcwele, ongcolileyo, or an explicit human/animal-breath sense as required, and use idimoni as a loanword safety-valve alongside umoya ongcolileyo specifically to keep the demonic category lexically distanced from amadlozi vocabulary.
2. iNkosi (“Lord”) — crowded by its unbroken historical use as the title of the actual Ndebele kings (Mzilikazi, Lobengula) and living chiefs today. Mark intensifies this collision directly through the κατακυριεύω wordplay (10:42) and the “King of the Jews” mockery (ch. 15). Fencing strategy: treat 10:42-45 and ch.15’s royal-title material as a single coordinated teaching unit (see Section 2.5) rather than translating each occurrence in isolation; never let “iNkosi” stand for Jesus without contextual reinforcement that his lordship exceeds and redefines the royal title, especially near occurrences of ukubusa ngamandla.
3. -khonza (root of isikhonzi, “servant”) — crowded by its ordinary religious-devotion sense (service/devotion to amadlozi, or ritual service to a chief). Fencing strategy: always anchor isikhonzi to a stated object of service (people, in imitation of Christ) rather than leaving the verb’s implied object ambiguous.
4. imfihlo (“mystery/secret”) — crowded by its occult/divinatory connotation (secret knowledge mediated by izangoma). Fencing strategy: pair with yombuso kaNkulunkulu every time; never let “imfihlo” stand alone as a gloss for kingdom teaching.
5. ukwesaba (“fear”) — crowded internally (not by traditional religion but by its own semantic breadth spanning terror and awe). Fencing strategy: context-sensitive intensifiers and translator notes at every occurrence, as detailed in 2.3.
6. isivumelwano (“covenant,” baseline) meeting igazi (“blood”) in the compound igazi lesivumelwano — crowded by active Nguni blood-oath/alliance-sealing customs between two human parties. Fencing strategy: every occurrence requires the “unilateral, divine-initiative, once-for-all” teaching note already established in the registry.
7. inhlawulo — crowded by its everyday, currently-practiced legal sense (see 2.5/2.8). Fencing strategy: the direction-of-payment inversion note is mandatory at every occurrence; this is the single highest-priority fencing requirement in the entire Mark package.

---

5. Transliteration vs. Paraphrase Decisions

Term	Decision	Rationale
uJesu, uMesiya, uDavida, uIsrayeli, uKristu	Transliterate (established)	Baseline convention; proper names and titles with fixed Ndebele Bible-tradition forms.
Abha (Abba)	Transliterate, untranslated	Baseline convention; preserves Aramaic filial intimacy no Ndebele gloss fully captures.
Hosana (Hosanna)	Transliterate, untranslated	Same convention family as Abba; Hebrew acclamation with no natural Ndebele equivalent.
Eloyi, Eloyi, lama sabakhithani (cry of dereliction)	Transliterate + mandatory Ndebele gloss alongside	Must preserve raw Aramaic anguish (per Abba precedent) while still making Psalm 22:1’s meaning accessible; gloss is supplementary, never a silent replacement.
iKhobhani (Corban)	Transliterate + explanatory gloss	One-off technical vow-term with no natural equivalent; low doctrinal risk beyond requiring a footnote gloss.
uBelizebhulu (Beelzebul)	Transliterate	Proper name of a demonic figure; must not be replaced by any name from Ndebele traditional cosmology.
iLejiyoni (Legion)	Transliterate	Proper name from the narrative (Roman military unit); reinforces rather than substitutes for the “unclean spirit” category.
idimoni/amadimoni (demon)	Transliterate (loanword) alongside paraphrase umoya ongcolileyo	Deliberate dual-track strategy: the loanword idimoni is used specifically to keep “demon” lexically distanced from native umoya vocabulary and its ancestral-spirit associations, while umoya ongcolileyo preserves intelligibility for readers unfamiliar with the loanword.
isiphambano (cross)	Paraphrase/compound (standard Nguni Bible-tradition term)	Retains specific historical execution-instrument weight; no viable transliteration option exists or is needed.
inhlawulo (ransom)	Paraphrase (repurposed existing Ndebele legal term)	Deliberate choice to retain cultural resonance of a real institution rather than coin an artificial neologism, provided the direction-of-payment inversion is taught at every occurrence.
ukubhabhathizwa (baptism-suffering metaphor)	Paraphrase (reuse of the standard sacramental term)	No viable alternative without inventing an obscure word that would itself need explanation; the better fix is a mandatory translator note, not a different lexical choice.
iNdodana yoMuntu (Son of Man)	Paraphrase/compound (new fixed coinage)	A literal, transparent compound (“child of the person/human”) that must be fixed by convention rather than transliterated, since it is a translated title (from Hebrew/Aramaic idiom via Greek), not a proper name.
Nginguye (“I am”)	Paraphrase (ordinary Ndebele copula construction)	No transliteration option exists for a self-referential grammatical formula; consistency of rendering across both occurrences carries the doctrinal weight instead.
igazi lesivumelwano (blood of the covenant)	Paraphrase/compound	Standard Nguni Bible-tradition phrase; transliteration is not applicable to a descriptive phrase.

General principle applied: transliteration is reserved for (a) proper names, (b) established Hebrew/Aramaic liturgical expressions already conventionally preserved untranslated in Nguni Bible tradition (Abba, Hosanna), and (c) cases where a loanword usefully avoids a native term’s syncretism risk (idimoni). Paraphrase/compounding is used everywhere a transparent, teachable Ndebele phrase can carry the concept, even where — as with inhlawulo — the paraphrase reuses a semantically “crowded” existing word, provided that crowding is actively fenced in teaching material.

---

6. Ranked List of Highest-Risk Ambiguities

Ranked by combined severity (doctrinal centrality) and likelihood (how easily a translator or reader could default to the wrong sense without active intervention):

1. inhlawulo (ransom, Mark 10:45). Highest risk in the entire package: an active, currently-practiced customary-law institution running in the opposite structural direction from the biblical concept. Every occurrence requires the direction-of-payment inversion note. Core-passage-critical.
2. umoya ongcolileyo / idimoni vs. amadlozi. Sustained across seven chapters; the reader’s entire cosmological map of “spirit” defaults toward ancestral categories the biblical text explicitly excludes. Requires the most repeated teaching reinforcement of any item on this list.
3. iNkosi / ukubusa ngamandla wordplay (Mark 10:42-45; ch. 15). Directly engages the historical Ndebele royal title; mistranslating either half of the wordplay independently (rather than as a coordinated unit) collapses the passage’s central servanthood-vs-greatness argument. Core-passage-critical.
4. esikhundleni sabanengi vs. a general-benefit phrase. A single preposition’s mistranslation (substitution collapsed into benefit) would quietly remove the doctrine of penal/substitutionary atonement from the ransom saying without any obvious surface-level error. Core-passage-critical.
5. ukuvusa/ukuvuswa vs. ukuvuka kwabafileyo. If these two are ever merged in drafting, Mark’s careful distinction between temporary this-worldly raisings and Christ’s unique eschatological resurrection collapses into an amadlozi-compatible cyclical reading.
6. ukubhabhathizwa as suffering-metaphor (10:38-39) vs. its sacramental default. High likelihood of silent misreading because the word is otherwise entirely familiar and unremarkable to readers from other church contexts.
7. iNdodana yoMuntu vs. iNdodana kaNkulunkulu. A newly coined title running in parallel with an already-fixed Critical title across the same narrative; inconsistent rendering at any of the 14 occurrences would blur the veiled/unveiled Messianic Secret arc.
8. igazi lesivumelwano (blood of the covenant). Collision with active Nguni blood-oath customs between human parties; risk of the Last Supper being read as a mutual pact rather than a unilateral divine act.
9. Nginguye (“I am”) consistency across 6:50 and 14:62. Lower narrative frequency than other items but doctrinally load-bearing for the whole Messianic Secret resolution; easy to under-translate as a merely reassuring phrase at 6:50.
10. ukwesaba (fear) sense-disambiguation across paired faith/fear verses. Frequent (occurs at five distinct doctrinally significant points) but individually lower-severity than items 1–9; risk is cumulative rather than singular.
11. ukukhaphela vs. ukunikelwa (betray/hand over). Risk of resolving Mark’s deliberate human-culpability/divine-purpose tension in translation by defaulting to only one verb throughout.
12. imfihlo yombuso kaNkulunkulu vs. divinatory “secret knowledge.” Lower frequency (concentrated at 4:11) but high severity if unfenced, given the term’s occult overtone.

---

7. Handoff Notes for Later Phase 1 Steps

• This analysis assumes and requires the Core Glossary (08) and assets/bible_term_registry.json entries as already fixed; it does not propose alternative renderings, only risk-tiering, fencing strategy, and transliteration/paraphrase rationale for those already-fixed choices.
• Items ranked 1–5 above should receive the most explicit forbidden-substitution language in the eventual Mark-specific AI Translation Requirements document (Step 12), following the pattern already established in the baseline’s 12_ai_translation_requirements.md.
• The dual-track idimoni/umoya ongcolileyo strategy (Section 4.2, item 1) should be tested with native speaker reviewers before being finalized as a hard rule, since loanword acceptance can vary by region within Matabeleland.
• No chapter of Mark requires further gap analysis beyond what is recorded here and in the Core Glossary; Phase 1 Step 8 is complete for full-book coverage.