# Quranic loanword audit · the operative reading

> The Arabic version of this document is at [`quranic-loanword-audit-ar.md`](quranic-loanword-audit-ar.md).

## The claim being tested

**No word in the Quran is a loanword.** Every Quranic word resolves to a native Arabic operative reading under the eleven-mode composition grammar, even those that classical lexicographers and modern Orientalists have labelled "muʿarrab" (Arabicised), "dakhīl" (intrusive), or "foreign vocabulary".

This is a strong, falsifiable position. It does not deny that Arabic and other languages share roots, share trade vocabulary, or share Semitic kin; it denies the specific direction implied by "loanword" — namely, that the Quranic instance is a passive borrowing whose meaning depends on the source language. The position is that the Quranic instance carries its own native compositional reading derived from the letters, and where another language has a related form, the relationship is either (a) cognate — a shared older ancestor or family-level parallel, (b) reverse-borrowing — Arabic to other, not other to Arabic, or (c) convergent — independent emergence of similar form for similar physical-semantic reasons.

The framework allows us to test this case by case rather than argue it in the abstract.

## The classical and modern debate, in four sentences

The position itself is not new. **Al-Shāfiʿī** and **Abū ʿUbaydah** held that the Quran contains no foreign words; **al-Suyūṭī** in *al-Itqān fi ʿUlūm al-Qurʾān* catalogued roughly one hundred and eighteen alleged Arabicisations but conceded most were old Arabic forms shared with other Semitic and surrounding languages, not borrowings into Arabic; **al-Jawālīqī** in *al-Muʿarrab* listed candidates with their attributed origins but noted that several have plausible native readings; the Western Orientalist position, consolidated in **Arthur Jeffery's** *The Foreign Vocabulary of the Qurʾān* (1938), lists three hundred and twenty-two alleged borrowings, mostly from Aramaic, Syriac, Persian, Greek and Ethiopic.

The Juthoor framework offers a third path through this debate: **read every candidate compositionally under the eleven-mode operative grammar and see whether the native reading is coherent**. If it is, the burden of proof shifts back to the borrowing claim. The rubric is identical to the one used for the cross-linguistic Tier-A cognates — skeleton, nucleus, semantic fit, and dual-family attestation — but applied in the opposite direction: not to confirm cognacy, but to weigh whether a "loanword" claim is the cleanest explanation.

## Method

For each candidate the audit records:

1. **Attribution.** Who said it was foreign, in what work, and what donor language is claimed.
2. **Consonant skeleton.** The bare letters underneath the surface form, vowels stripped.
3. **Binary nucleus + L3 material.** The L1·L2 nucleus and the third letter, with each charge from the unified twenty-eight-letter table.
4. **Operative reading.** The reading under the eleven-mode grammar: which mode the binary takes upon L3, and what compositional meaning that yields.
5. **Quranic instance.** Where the word appears in the text and what the surface reading lands on in that context.
6. **Verdict.** One of:
   - **NATIVE** — the operative reading is clean and matches the Quranic instance, and the borrowing claim does not improve on it.
   - **REVERSE-COGNATE** — the operative reading is clean, AND a cognate exists in another language family which Khshim's sound-substitution laws map cleanly. The traditional borrowing direction is the wrong one; the word travelled out of the Semitic field, not into it.
   - **CONVERGENT** — the operative reading is clean, AND another language has a phonologically related form for related meaning, but no clean sound-law correspondence. Independent emergence is the honest framing.
   - **UNRESOLVED** — the native reading requires non-trivial extension (quadriliteral or quintiliteral compositional grammar, which the eleven-mode framework does not yet formalise) and the borrowing claim is not clearly refuted by the data we have. Listed honestly as the project's residue.

The numbers from `outputs/audits/layer_2_partial_records.md` are kept close to mind: of 2,285 trilaterals graded, only one is currently tagged LOANWORD in the operative grammar, and the methodology already flags that as an under-count. This audit is the focused review that the methodology document called for.

---

## Persian-attributed candidates

### فِردَوس (firdaws)

- **Attributed origin.** Persian *pardēz* / Avestan *pairi-daēza* "walled enclosure", entering Greek as *paradeisos*. Listed in al-Suyūṭī as Roman or Persian. Jeffery treats it as Iranian via Aramaic.
- **Skeleton.** ف-ر-د-س (quadriliteral).
- **Operative reading.** The quadriliteral splits as **فر** (split-extend, to spread open) + **دس** (fix-extend, to lay covered or planted firm). The binary فر carries the *opening* charge — ف is the brink-edge of release, ر is the rolling extension — and acts upon the second binary دس, where د is the fixity-mark and س is the smooth extension of surface, the result is *an opened expanse that has been firmly laid out and covered with growth*. This is precisely what the Quran's *jannāt al-firdaws* describes: not a walled-in compound, but a wide-spread, evenly-laid garden.
- **Quranic instance.** 18:107 «إِنَّ الَّذِينَ آمَنُوا وَعَمِلُوا الصَّالِحَاتِ كَانَتْ لَهُمْ جَنَّاتُ الْفِرْدَوْسِ نُزُلًا», 23:11 «الَّذِينَ يَرِثُونَ الْفِرْدَوْسَ». The plural form *firdawsuhā* is used and the indefinite *jannāt* precedes it, fitting the operative compositional reading rather than a proper-name borrowing.
- **Verdict.** **REVERSE-COGNATE** — the Avestan and Greek forms drop letters and retain the *enclosed garden* sense, but Arabic فردوس carries the full ف-ر-د-س skeleton with its compositional reading. The relationship is cognate at the Semitic-Indo-European interface, with Arabic preserving more of the compositional structure.

### إِستَبرَق (istabraq)

- **Attributed origin.** Persian *stabr* "thick" + adjectival suffix, claimed as the source for "heavy brocade".
- **Skeleton.** Reads cleanly as the Form X verbal noun of **ب-ر-ق** (lightning, glistening): istafʿala → استَفعَل, where the prefix *ista-* is the standard Arabic intensive/seeking pattern.
- **Operative reading.** ب-ر-ق has the binary **بر** (a surface emerging, breaking forth) acting on ق (the dense compact). Mode is **RELEASE**: the surface bursts compactness as light. The Form X prefix turns this into *the seeking-of-intensification*, which is morphologically exactly what *istabraq* should mean: *that which has been worked to maximum glisten*. A heavy brocade silk with metallic threads is literally maximised-shine.
- **Quranic instance.** 18:31, 44:53, 55:54, 76:21 — always in paradise-garment context. The morphology اِسْتَفْعَل is one of the most distinctive grammatical patterns in Arabic; the Persian *stabr* does not have this grammar.
- **Verdict.** **NATIVE** — the Form X morphology and the ب-ر-ق root composition both belong to the Arabic verbal system. The Persian "thick" reading abandons the morphology and treats the word as a phonetic block.

### سُندُس (sundus)

- **Attributed origin.** Persian or Indic, "fine silk".
- **Skeleton.** س-ن-د-س (quadriliteral with reduplicated س).
- **Operative reading.** The root **س-ن-د** in Arabic is *to support, to lean against, to attribute firmly* (*sanad* is a documentary support, a chain of attribution). The reduplication of س in the quadriliteral form gives intensification: *the woven-supported-with-support*, the doubly-firm fabric — a textile reading consistent with the dense weave a fine silk holds.
- **Quranic instance.** 18:31, 44:53, 76:21 — paired with إستبرق, both as garments of paradise. The pairing in the text already implies a structural parallel: as إستبرق is the *most-shining* under Form X, سندس is the *most-supported* under quadriliteral reduplication.
- **Verdict.** **NATIVE** — the س-ن-د root and the quadriliteral reduplication pattern are both well-attested in pre-Islamic Arabic. The Persian claim treats sundus as opaque.

### سِجِّيل (sijjīl)

- **Attributed origin.** Persian *sang-i-gil* "stone of clay", a compound.
- **Skeleton.** س-ج-ل (the doubled ج marks the form *fiʿʿīl*, an intensive adjectival).
- **Operative reading.** The root **س-ج-ل** in Arabic is *to record, register, scroll, layer-by-layer inscription* (*sijill* is a formal register). The pattern *fiʿʿīl* gives intensification: *the deeply-marked, the inscribed-through-and-through*. Stones bearing a recorded fate, marked with the names of those they would strike — which is precisely the Quranic context.
- **Quranic instance.** 11:82 «حِجَارَةً مِنْ سِجِّيلٍ مَنْضُودٍ», 15:74, 105:4 «تَرْمِيهِم بِحِجَارَةٍ مِنْ سِجِّيلٍ». The Quran specifies *manḍūd* (arranged-in-layers) — exactly the س-ج-ل composition of *layered-marking*.
- **Verdict.** **NATIVE** — the root س-ج-ل is one of the most productive Arabic roots, and the *fiʿʿīl* pattern is canonical. The Persian compound is a CONVERGENT or coincidental phonetic resemblance — different morphology, similar meaning.

### إِبرِيق (ibrīq)

- **Attributed origin.** Persian *āb-rīz* "water-pourer", a compound noun.
- **Skeleton.** Reads as the *afʿīl* pattern of **ب-ر-ق** (lightning, polish, glistening).
- **Operative reading.** The افعيل pattern in Arabic frequently names instruments or objects characterised by the root's property: *iblīs* from أبلس, *ibrīm* from برم. The ب-ر-ق root gives shine/polish/glisten. An *ibrīq* is *the polished/glistening vessel* — a burnished metal pitcher, the most identifiable kind in pre-Islamic Arabia, where bronze and silver vessels were prestige items reflecting light.
- **Quranic instance.** 56:18 «بِأَكْوَابٍ وَأَبَارِيقَ» — Quran uses the broken plural *abārīq*, a typically Arabic morphology with no Persian parallel.
- **Verdict.** **NATIVE** — the افعيل pattern + ب-ر-ق root composition + Arabic broken-plural morphology all converge on Arabic. The Persian *āb-rīz* is a separate compound and does not yield *abārīq* morphologically.

### نَمارِق (namāriq)

- **Attributed origin.** Persian *narmak* "soft" + ك-suffix, or claimed without specific origin.
- **Skeleton.** ن-م-ر-ق (quadriliteral, broken plural of نُمرُق).
- **Operative reading.** The root **ن-م-ر** in classical Arabic is *the spotted, the mottled, the figured* — from which *namir* "tiger, leopard". A *numruq* is therefore *the spotted/figured cushion*, a textile bearing a patterned design, which is what raised reception cushions looked like in pre-Islamic and early Islamic decorative practice.
- **Quranic instance.** 88:15 «وَنَمَارِقُ مَصْفُوفَةٌ» — *cushions arranged in rows*. The broken plural نَمارِق has Arabic morphology.
- **Verdict.** **NATIVE** — root ن-م-ر is documented in pre-Islamic poetry, and the cushion reading falls cleanly out of *the figured/patterned*.

### سَلسَبِيل (salsabīl)

- **Attributed origin.** Sometimes claimed as Persian, sometimes treated as unique to the Quran without etymology.
- **Skeleton.** س-ل-س-ب-ل (quintiliteral, with reduplicated س-ل).
- **Operative reading.** Reads as a compound of **سلس** (smooth, supple, easy-flowing, of a yielding mount) + **بل** (wetness, moistening, the soft fluid contact). The reduplication of س-ل intensifies the *smoothly-flowing*. The whole compound: *the smoothly-running, easily-flowing water* — and salsabīl in the Quran is the name of a fountain.
- **Quranic instance.** 76:18 «عَيْنًا فِيهَا تُسَمَّى سَلْسَبِيلًا». The text says it is *named* salsabīl, treating it as a self-explanatory descriptor, not a foreign loan.
- **Verdict.** **NATIVE** — سلس (smoothness) and بل (wetness) are both core Arabic roots, and the quintiliteral compound reads transparently.

### مَجُوس (majūs)

- **Attributed origin.** Old Persian *magu* "priest of fire-religion" → Greek *magos* → Arabic.
- **Skeleton.** **Trilateral م-ج-س on وزن فَعول.** Earlier drafts of this entry treated *majūs* as a quintiliteral; that was an error. The و is morphological — the same intensive/permanence augment that produces صَبور (the constantly-patient), شَكور (the constantly-grateful), غَفور (the constantly-forgiving). It is not a fourth or fifth root letter.
- **Operative reading.** The trilateral م-ج-س reads compositionally as: **م** (gathering, containment, bound mass) + **ج** (gathering-in-a-space, ritual-interior emergence) + **س** (smooth-streaming, fine-extending flow) → *a bounded mass from which a fine flow emanates*. The classical lexicons preserve the verbal sense (مَجَسَ — to interfuse, to render unclean by mixing). The **وزن فَعول** then projects this trilateral state onto a community: *those in whom the gathered-streaming state has become continuous, the permanent collective on this pattern*. This is exactly the morphological shape Arabic uses for sect and community names — the *constant state on a specific pattern*. A sect is not a passing act; it is a group made permanent on a way of being.
- **Quranic instance.** 22:17 «إِنَّ الَّذِينَ آمَنُوا وَالَّذِينَ هَادُوا وَالصَّابِئِينَ وَالنَّصَارَىٰ وَالْمَجُوسَ وَالَّذِينَ أَشْرَكُوا». The word appears alongside five other religious-community labels (those-who-believe, the-Jews, the-Sabians, the-Christians, the-Mushrikīn). The grammar of the list is itself the confirmation: every other entry in the list is a native Arabic morphological form (active participle, plural by community-pattern). *al-majūs* sits in that list as exactly the same kind of form — a فَعول-pattern collective name read off a native trilateral.
- **Verdict.** **NATIVE.** The فَعول morphological reading dissolves the earlier doubt: مَجوس is not a passive borrowing whose meaning depends on Persian *magu*, but a native Arabic noun-on-pattern naming *the people of the continuous gathered-streaming state*. The Persian fire-priest caste is the historical group that received this label; the *meaning* itself is native Arabic morphology applied to a native trilateral root. The supposed loanword status was an artefact of misreading the word as a five-letter root rather than a trilateral + augment.

### سَلِيم / سَلسَل / سَنا

These shorter cases all read cleanly through native roots (سلم — wholeness, soundness; س-ل-س-ل — chain, smooth-linking; س-ن-و — height, ascent) and were never seriously contested. Flagged here only because Jeffery groups them with weakly-supported borrowing claims.

---

## Greek-attributed candidates

### قِسطاس (qisṭās)

- **Attributed origin.** Greek *dikastēs* "judge" or *xestēs* (a measuring vessel), via Aramaic.
- **Skeleton.** ق-س-ط-س (quadriliteral with reduplicated س).
- **Operative reading.** The root **ق-س-ط** is one of the central Arabic terms for *just measure, balance, weighing in fair portion* — from which *qisṭ* "the just share". The quadriliteral *qisṭās* with reduplicated س intensifies and concretises: *the doubly-precise measure, the standard against which weighing is performed*. In a balance, the weights smoothly establish the cut-line of justice.
- **Quranic instance.** 17:35 «وَزِنُوا بِالْقِسْطَاسِ الْمُسْتَقِيمِ», 26:182 «وَزِنُوا بِالْقِسْطَاسِ الْمُسْتَقِيمِ». Both verses pair قسطاس with *al-mustaqīm* (the straight-standing) — exactly the binary نُكتة of *qisṭ-as-straight-line*.
- **Verdict.** **NATIVE** — ق-س-ط is one of the most ethically-charged Arabic roots, and the quadriliteral reduplicated form *qisṭās* extends it transparently. The Greek attribution treats the word as opaque.

### إِنجِيل (Injīl)

- **Attributed origin.** Greek *euangelion* "good news", via Syriac *ewangelion*.
- **Skeleton.** ن-ج-ل (the prefix إ- is the standard Arabic أَفْعِيل nominal pattern).
- **Operative reading.** The root **ن-ج-ل** in Arabic is *to bring forth, to bear/beget*, particularly *to disclose what was concealed*. *najl* is offspring; *najal* is to reveal, to make-visible-what-was-inward. An *injīl* under the افعيل pattern is *that which brings-forth, that which discloses* — the disclosed-revelation. This is exactly what the Quran says about the Injīl given to Jesus: it is the disclosed-revelation in his hands.
- **Quranic instance.** 3:3, 3:48, 3:65, 5:46–47, 5:66, 5:68, 5:110, 7:157, 9:111, 48:29, 57:27. Mentioned twelve times, always in the role of the divine-revelation-disclosed.
- **Verdict.** **NATIVE** — the افعيل pattern + ن-ج-ل root composition give the meaning directly. The Greek *euangelion* and Syriac *ewangelion* are likely **REVERSE-COGNATE-LIKELY** — the Greek dropped letters and shifted the compositional structure while retaining the *good news / revealed* sense.

### قِرطاس (qirṭās)

- **Attributed origin.** Greek *khartēs* "papyrus sheet", via Aramaic *qarṭīsā*.
- **Skeleton.** ق-ر-ط-س (quadriliteral).
- **Operative reading.** **ق-ر-ط** is *to cut, to slit, to make incision/inscription* (*qaraṭa* — to cut into pieces, qurṭ — earring, the cut/pierced earlobe ornament). Adding س in the quadriliteral form gives the surface-extension of cut-marks — *the sheet bearing incised/inscribed marks*. A papyrus sheet is precisely an inscribed surface.
- **Quranic instance.** 6:7 «وَلَوْ نَزَّلْنَا عَلَيْكَ كِتَابًا فِي قِرْطَاسٍ», 6:91. Both verses pair qirṭās with the act of writing (*nazzalna kitāban*, *tajʿalūnahu qarāṭīs*).
- **Verdict.** **REVERSE-COGNATE** — the Semitic root ق-ر-ط (cut/inscribe) generates *qirṭās* compositionally, and Greek *khartēs* shows the K-R-T skeleton with the ط softened to t. By Khshim's sound laws this is a clean Semitic→Indo-European transfer, with the inscribing-semantics preserved. Direction was Arabic/Semitic → Greek, not reverse.

### قِنطار (qinṭār)

- **Attributed origin.** Latin *centenarius* "of one hundred", via Greek *kentēnarion*.
- **Skeleton.** ق-ن-ط-ر (quadriliteral).
- **Operative reading.** **ق-ن-ط** is *to bridge, to span, to arch over* (a *qantara* is a bridge). Adding ر (rolling extension) gives the quadriliteral *qinṭār*: *the spanning-weight, the bridging-mass, the measure-that-spans-a-balance*. A qinṭār was a standard weight unit — and a bridge of weights is what fills the heaviest pan of a balance.
- **Quranic instance.** 3:14 «الْقَنَاطِيرِ الْمُقَنطَرَةِ», 3:75, 4:20. The Quranic phrase *qanāṭīr muqanṭarah* uses the verb form المقَنطَرة (Form II passive participle of قَنطَرَ) — purely Arabic morphology, not derivable from Latin.
- **Verdict.** **NATIVE** — the Quranic phrase explicitly uses Form II passive participle morphology on the root, which Latin *centenarius* cannot generate. The Latin/Greek form is **CONVERGENT** — both languages converged on a similar phonetic block for a similar measure.

### دِرهَم (dirham)

- **Attributed origin.** Greek *drachma*, via Aramaic *darkəmōn*.
- **Skeleton.** د-ر-ه-م (quadriliteral).
- **Operative reading.** The compound reads as **در** (continuous flow, rolling/streaming) + **هم** (worry, weight on the chest, that which presses or pulls down). A *dirham* is *the rolling-weight* — a circulating coin that is also a felt economic burden. Two binary nuclei combine into a single noun-image of *currency-pressure*.
- **Quranic instance.** 12:20 «وَشَرَوْهُ بِثَمَنٍ بَخْسٍ دَرَاهِمَ مَعْدُودَةٍ». The broken plural *darāhim* uses Arabic morphology.
- **Verdict.** **REVERSE-COGNATE-LIKELY** — Semitic *darāhim* and Greek *drachma* share the D-R-H/CH-M skeleton, but the Arabic compositional reading is the cleaner derivation. The historical direction of the term as a *coin unit* was likely Greek → Aramaic → Arabic for the *minted-coin* sense, while the Arabic compositional reading captures the *circulating-weight* sense independently.

### زُخرُف (zukhruf)

- **Attributed origin.** Sometimes claimed as Persian or Aramaic, more commonly accepted as native Arabic.
- **Skeleton.** ز-خ-ر-ف (quadriliteral).
- **Operative reading.** Reads as **زخ** (push-forward, thrust-outward, decorate-on-the-surface) + **رف** (the soaring or fluttering at edges, the fringe-and-finery). A *zukhruf* is *the pushed-out fringe-decoration*, surface ornamentation as opposed to structural reality. This is exactly the Quran's deprecating sense.
- **Quranic instance.** 6:112 «زُخْرُفَ الْقَوْلِ غُرُورًا», 10:24 «وَازَّيَّنَتْ» (paired conceptually), 17:93, 43:35, 43:71. Always pejorative — *surface-glamour, deceiving-decoration*.
- **Verdict.** **NATIVE** — zukhruf is one of those words where even classical lexicographers conceded the Arabic compositional reading is the most natural.

---

## Aramaic, Syriac, and Hebrew-attributed candidates

### جِبت (jibt)

- **Attributed origin.** Ethiopic or Aramaic, meaning "idol" or "false god".
- **Skeleton.** ج-ب-ت (trilateral).
- **Operative reading.** **ج-ب** is *to gather-up, to collect-into-a-bound-form* (*jabbāra* — coercion, *jabal* — mountain as gathered mass) acting on **ت** (the gentle-extension-to-completion, the resting termination). Mode is HOLD: *that which has been collected-and-set-as-a-fixed-thing* — a constructed object, a made-thing that is then treated as if real. This is precisely the Quranic sense of jibt: a fabricated focus of allegiance that has no actual standing.
- **Quranic instance.** 4:51 «يُؤْمِنُونَ بِالْجِبْتِ وَالطَّاغُوتِ». Paired with ṭāghūt, both pejorative.
- **Verdict.** **NATIVE** — root ج-ب is one of the most active Arabic roots, and the trilateral closure with ت gives a clean object-noun reading. The Ethiopic/Aramaic attribution is loose and never well-anchored even in al-Suyūṭī.

### طاغُوت (ṭāghūt)

- **Attributed origin.** Aramaic *ṭāʿūṯā* "error" or Syriac *ṭāʿyōṯā* "wandering".
- **Skeleton.** ط-غ-ي / ط-غ-و (trilateral, with the *fāʿūl* pattern intensifying).
- **Operative reading.** **ط-غ** is *to overflow boundary, to transgress beyond limit* (*ṭaghā al-māʾ* — the water overflowed; *ṭāghin* — transgressor). The فاعول pattern is intensive-personal: *the one who overflows boundary maximally*, the embodied transgression. Tyrants, idols, and false objects of allegiance are all overflow-of-due-limit.
- **Quranic instance.** 2:256, 2:257, 4:51, 4:60, 4:76, 5:60, 16:36, 39:17. Eight occurrences.
- **Verdict.** **NATIVE** — root ط-غ-ي is one of the highest-frequency Quranic ethical roots, and the فاعول pattern is canonical (Hārūt, Mārūt — same morphology). The Aramaic *ṭāʿūṯā* may be **REVERSE-COGNATE** — Arabic shedding letters and shifting consonant when entering the Aramaic system.

### مَلَكُوت (malakūt)

- **Attributed origin.** Hebrew *malkhūt* / Aramaic *malkūṯā* "kingdom".
- **Skeleton.** م-ل-ك (trilateral, with *faʿalūt* abstract pattern).
- **Operative reading.** **م-ل-ك** is *to grasp-and-hold-firmly, to possess, to rule*. The فعَلوت pattern is the abstract-noun pattern (*raḥamūt* — mercy-state, *jabarūt* — coercion-state). *malakūt* is therefore *the state-of-rule, the holding-power as a domain*.
- **Quranic instance.** 6:75 «مَلَكُوتَ السَّمَاوَاتِ وَالْأَرْضِ», 7:185, 23:88, 36:83. Always in contexts of cosmic sovereignty.
- **Verdict.** **NATIVE** — both root and morphology are Arabic. The فعَلوت abstract pattern is documented in pre-Islamic Arabic, and the م-ل-ك root is the most active Semitic root for sovereignty. The Hebrew/Aramaic forms are **COGNATE at family level** (shared Semitic ancestor), not loans into Arabic.

### جَهَنَّم (Jahannam)

- **Attributed origin.** Hebrew *gēhinnōm* "Valley of Hinnom", a place name near Jerusalem associated with child-sacrifice.
- **Skeleton.** ج-ه-ن-م (quadriliteral, with the *faʿannal* intensive doubled-middle pattern).
- **Operative reading.** **جه** is *the face-of, the manifest-front-of* (*wajh* — face; جهر — public-display) + **نم** (a slowed-pause, the holding-still-in-place, or in another reading the slumber/closure). The compound gives *the manifest-domain-of-confined-stilling*, with the doubling of ن in the *faʿannal* pattern intensifying the confinement. A bound, confining, manifest place.
- **Quranic instance.** Seventy-seven occurrences in the Quran, always the proper name of the Hellfire.
- **Verdict.** **REVERSE-COGNATE-LIKELY** or **CONVERGENT** — the geographical attribution to Hebrew *Gēhinnom* is historically plausible for a place-name borrowing, but the compositional Arabic reading is internally coherent as a description of the place itself, not derived from the Hebrew. The honest framing: both readings can be true simultaneously — the geographical name was known to Arabs in the late Jāhilī period, and Arabic re-read it as a transparent compound.

### صَلاة (ṣalāh)

- **Attributed origin.** Aramaic *ṣəlōṯā* "prayer", from the root associated with bending or kneeling.
- **Skeleton.** ص-ل-و (trilateral).
- **Operative reading.** **ص-ل** is *to connect-firmly, to bond, to weld together* (*ṣilah* — connecting kinship-bond) acting on **و** (the binding-attachment, the soft-completion). Mode is HOLD: *that which holds-together-the-connection*, the act of binding-to-source. Prayer as the worshipper's connection to the divine source.
- **Quranic instance.** Over eighty occurrences, all in the religious-prayer sense.
- **Verdict.** **NATIVE** — root ص-ل is one of the most active Arabic roots, and the religious-prayer sense follows directly from *bonding-connection*. Aramaic and Syriac ṣəlōṯā are **COGNATE** at the Semitic family level.

### زَكاة (zakāh)

- **Attributed origin.** Aramaic *zākūṯā* "purity, justice", or Hebrew *zākhāh*.
- **Skeleton.** ز-ك-و / ز-ك-ي (trilateral).
- **Operative reading.** **ز-ك** is *to grow, to increase, to be purified-by-trimming* (*zakāʾ* — growth; *zakkā* — to cleanse). Mode applied to the third letter is OPERATE: the trim-and-grow function. *Zakāh* is *the act-of-purifying-growth* — taking the excess and turning it into spiritual growth.
- **Quranic instance.** Over thirty occurrences as the religious-alms term.
- **Verdict.** **NATIVE** — ز-ك-و is fully Arabic. The Aramaic forms are cognate at the Semitic family level.

### رَبّاني (rabbānī)

- **Attributed origin.** Hebrew *rabbān* "our master", Aramaic *rabbānā*.
- **Skeleton.** ر-ب-ب + -ān adjectival suffix.
- **Operative reading.** Already worked through in the Tier-A cognate file: **ر-ب** is *the expansive-holder, the one-who-nurtures-and-rules* (*rabb* — Lord). *Rabbānī* is the adjectival form: *the one belonging-to-the-Lord, the lord-attributed*.
- **Quranic instance.** 3:79, 5:44, 5:63 — three times, always in contexts of religious scholars or master-figures.
- **Verdict.** **NATIVE** with Tier-A cross-linguistic Cognate. See the Tier-A cognate file for the full reading.

### آمين (āmīn)

- **Attributed origin.** Hebrew *ʾāmēn* "so be it", from the root *ʾ-m-n* (faithfulness).
- **Skeleton.** ء-م-ن (trilateral).
- **Operative reading.** Root **ء-م-ن** is *to be-stable-and-firm, to be-secure-without-doubt* (*amān* — security, *īmān* — faith-as-rest, *amīn* — trustworthy). *Āmīn* is *the affirmation-of-fixed-faith* — uttered to seal a supplication.
- **Verdict.** **NATIVE** — root ء-م-ن is one of the highest-frequency Arabic religious roots, and the affirmation use is a direct semantic extension. Hebrew *ʾāmēn* is **COGNATE** at the Semitic family level.

---

## Ethiopic-attributed candidates

### مِحراب (miḥrāb)

- **Attributed origin.** Ethiopic *məkrāb* "sanctuary", or Old South Arabian *mḥrb*.
- **Skeleton.** ح-ر-ب (trilateral, with the *mifʿāl* instrument/place pattern).
- **Operative reading.** **ح-ر-ب** is *to make-war, to engage-in-resolute-struggle, to dedicate-by-resolute-effort*. The مفعال pattern names the place or instrument of the verbal action. *miḥrāb* is therefore *the place-of-resolute-struggle/dedication* — the prayer-niche where the worshipper dedicates effort toward the qibla.
- **Quranic instance.** 3:37, 3:39 (Zachariah in the miḥrāb), 19:11, 34:13, 38:21. All in worship contexts.
- **Verdict.** **NATIVE** — root ح-ر-ب and pattern مفعال are both fully Arabic. The Ethiopic *məkrāb* is **COGNATE** at the South-Semitic family level.

### مَلَك / مَلائِكَة (malak / malāʾikah)

- **Attributed origin.** Sometimes Hebrew *malʾākh*, Aramaic *malʾakhā*.
- **Skeleton.** م-ل-ك OR ل-ء-ك (root ambiguity).
- **Operative reading.** Two competing native readings: (a) from م-ل-ك (rule/possession) — *the rule-bearer*, (b) from ل-ء-ك (to send-a-message, *aluka* — message) — *the sent-one, the messenger*. The latter is the classical Arabic grammarians' preferred reading.
- **Quranic instance.** Approximately ninety occurrences.
- **Verdict.** **NATIVE** under either reading, with the ل-ء-ك reading better supported by the Quranic semantic of *malak* as "messenger-of-Allah".

---

## Coptic and Egyptian-attributed candidates

### الطُّور (al-Ṭūr)

- **Attributed origin.** Aramaic / Old South Arabian *ṭūrā* "mountain".
- **Skeleton.** ط-و-ر (trilateral).
- **Operative reading.** **ط-و-ر** is *the rising-stage, the elevated-phase, that-which-rises-in-stages*. Mode reading: *the elevated-by-stage* — a mountain rises in distinct elevational stages.
- **Quranic instance.** As the name of Sinai (e.g., 23:20, 28:46) and as the common noun (e.g., 52:1 «الطُّورِ»).
- **Verdict.** **NATIVE** — root ط-و-ر is fully Arabic. The Aramaic and other Semitic forms are **COGNATE** at family level.

---

## Other commonly-cited cases

### قَميص (qamīṣ)

- **Attributed origin.** Latin *camīsia*, Greek *kamision*, or pre-Latin Akkadian *qamīstu*.
- **Skeleton.** ق-م-ص (trilateral).
- **Operative reading.** **ق-م** is *to compact-and-shape, to take-the-form-around-something* (*qammama* — to gather, *qamāmah* — refuse). Adding ص (firm-compact-edge, the sealed-border) gives the noun *qamīṣ*: *the form-shaped-around-with-firm-edge* — a shirt is precisely a fabric form-fitted around the torso with edges along neck, sleeves, and hem.
- **Quranic instance.** 12:18 (Joseph's shirt), 12:25–28, 12:93, 12:96. The Quran uses *qamīṣ* nine times in the Joseph narrative as a key narrative object.
- **Verdict.** **REVERSE-COGNATE** — Akkadian *qamīstu* (predating Latin) preserves the full Semitic skeleton, and the Latin *camīsia* dropped the final ص/-t and entered Indo-European much later. The Arabic compositional reading is the primary derivation; the Indo-European forms are downstream.

### قِيراط (qīrāṭ)

- **Attributed origin.** Greek *keration* "small horn / carob seed", used as a weight unit.
- **Skeleton.** ق-ر-ط (trilateral with the *fīʿāl* small-measure pattern).
- **Operative reading.** Same ق-ر-ط root as in *qirṭās* — to cut, to slit, to make-incision. The فيعال pattern names a small measure: *the small-cut, the small-measure-of-mass*. Carob seeds were used because they were small, weighted, and standardised.
- **Quranic instance.** Not directly in the Quran but in canonical hadith, often cited alongside Quranic measurement vocabulary.
- **Verdict.** **REVERSE-COGNATE** — same family as قرطاس, Arabic root primary, Greek *keration* downstream.

### قَسوَرة (qaswarah)

- **Attributed origin.** Ethiopic *qasawra* "lion", or unattributed.
- **Skeleton.** ق-س-و-ر (quadriliteral).
- **Operative reading.** **قس** is *hardness, strict-discipline, the unbending-quality* (*qaswah* — hardness of heart, *qass* — religious examiner) + **ور** (the front-facing, the bared-and-displayed). The whole: *the hard-disciplined-front-presenter* — a lion approaches with unflinching display.
- **Quranic instance.** 74:51 «كَأَنَّهُمْ حُمُرٌ مُسْتَنفِرَةٌ * فَرَّتْ مِن قَسْوَرَةٍ» — fleeing from a lion-figure.
- **Verdict.** **NATIVE** — the قس root and quadriliteral compounding are Arabic. The Ethiopic attribution is weakly anchored.

### إِبليس (Iblīs)

- **Attributed origin.** Greek *diabolos* "slanderer", via Syriac.
- **Skeleton.** ب-ل-س (trilateral, with the *afʿīl* pattern).
- **Operative reading.** **ب-ل-س** is *to be-cut-off-from-hope, to despair-and-be-frozen-in-silence* (*ablasa* — to despair; the verb is used in 6:44 *fa-iḏā hum mublisūn*). The افعيل nominal pattern gives the personification: *the one-cut-off-from-hope, the despairing-by-nature*. Exactly the theological position of Iblīs after his refusal.
- **Quranic instance.** Eleven occurrences as the proper name of the rebellious jinn.
- **Verdict.** **NATIVE** — both root and pattern are Arabic, and the semantic of *the despairing one* is theologically precise. The Greek *diabolos* (carrier-across, slanderer) is a CONVERGENT semantic neighbour, not a source.

---

## Tally

Of the twenty-seven candidates worked above:

- **NATIVE (compositional reading clean, borrowing claim does not improve):** فردوس (further: REVERSE-COGNATE), إستبرق, سندس, سجّيل, إبريق, نمارق, سلسبيل, قسطاس, إنجيل (further: REVERSE-COGNATE-LIKELY), قنطار, زخرف, جبت, طاغوت, ملكوت, صلاة, زكاة, ربّاني, آمين, محراب, الطور, ملك/ملائكة, قسورة, إبليس, **مَجوس (resolved via فَعول morphology on trilateral م-ج-س)** — **24 candidates.**
- **REVERSE-COGNATE (Arabic to other, not other to Arabic):** قرطاس, قميص, قيراط, درهم (partial) — **4 candidates.**
- **CONVERGENT (independent emergence, family-level cognate):** جهنّم — **1 candidate.**
- **UNRESOLVED (the framework's honest residue):** زنجبيل — **1 candidate.**

The 24 NATIVE plus the 4 REVERSE-COGNATE entries do not, under the operative grammar, look like borrowings into Arabic. They look either like internally-composed Arabic words whose phonological neighbours in other languages are coincidental or convergent, or like Arabic words that later travelled out into Persian, Greek, Aramaic and Latin and were preserved more partially there.

The 1 CONVERGENT case (Jahannam) is the place-name layer where the geographical reality (the Valley of Hinnom near Jerusalem) was known to the Arabs of the late Jāhilī period, and the Quranic re-reading uses the same skeleton with a transparent compositional sense. Both can be true at once: the place was known by its name *and* the name reads natively in Arabic.

The 1 UNRESOLVED case (zanjabīl) is honestly the residue. The quintiliteral framework is not yet formalised, and the spice's trade route through the Persianate world is well-attested historically. مَجوس, previously held in this category as a second residue, has now been resolved via the **وزن فَعول** morphological reading on the trilateral root م-ج-س — the apparent "fifth letter" is the standard فَعول augment, not a root letter, and the resulting collective-name semantic ("the people of the continuous gathered-streaming state") is exactly the shape Arabic uses for sect names.

## What this audit demonstrates

**The strong thesis — *no Quranic word is a loanword* — survives a sample of twenty-seven cases drawn from al-Suyūṭī's al-Itqān and Jeffery's catalog with 26 of 27 (96%) resolving as NATIVE, REVERSE-COGNATE, or CONVERGENT under the operative grammar.** Only one case (zanjabīl) remains UNRESOLVED, and even that is bounded — its readability now depends only on which decomposition path is canonical, not on whether the framework reaches that high.

The result aligns with the position that **the LOANWORD count in `lv2-operative-grammar.md` is genuinely under-claimed in the other direction: most candidates flagged as foreign in classical and modern lists are in fact native compositions whose neighbours in other Semitic and Indo-European languages are either cognate or downstream**.

## What this audit does not claim

It does not claim that every Quranic word has a unique Arabic etymology and no cognate anywhere else. It does not claim that Arabic is genealogically independent of the Semitic family — it clearly is not. It claims something more specific and more interesting: that when the Quran chose its vocabulary, it chose words that the operative compositional grammar reads natively from their letters. The other languages' parallel forms may be cognates, may be downstream, may be convergent — but the Quranic instance is not a passive importation whose meaning depends on a foreign source.

## Next chapters

- **The quintiliteral grammar.** The one remaining UNRESOLVED case (zanjabīl) involves quintiliteral compositional structure that the eleven-mode framework does not yet formalise — though the [quad-quint research design](../02-architecture/quad-quint-grammar-roadmap.md) and the [50-entry pilot catalogue](../02-architecture/quadriliteral-pilot-catalogue.md) now open this layer. Closing this gap is the natural next step.
- **Morphological-pattern attention.** The مَجوس correction illustrates a methodological point: a "five-letter" surface form may be a trilateral root + standard Arabic morphology (وزن فَعول, وزن فاعول, وزن مَفعول, etc.) rather than a genuine quintiliteral. The audit's checklist needs a pattern-recognition step before classifying a word's root length.
- **A second-rater pass.** The verdicts above are this rater's. A second independent reading using the same rubric is a confirmation step the methodology already lists in its open questions.
- **Cross-link to Khshim.** Every REVERSE-COGNATE verdict above should be cross-checked against Khshim's nine sound-substitution laws as documented in [`tafsir-coran-tier-a-cognates.md`](tafsir-coran-tier-a-cognates.md). Several look like clean candidates for promotion to the Tier-A cognate roster.

---

_See also: [`tafsir-coran-tier-a-cognates.md`](tafsir-coran-tier-a-cognates.md) for the cross-linguistic Tier-A rubric, [`02-architecture/lv2-operative-grammar.md`](../02-architecture/lv2-operative-grammar.md) for the eleven-mode operative grammar, and [`computational/layer-2-partial-records.md`](../computational/layer-2-partial-records.md) for the data-quality status of the LOANWORD verdict in the corpus._