# Tafsir Coran, Tier A Cognates & Verification Rubric

> **Source:** companion methodology document in the Tafsir Coran sibling project.
> **Status:** processed contribution, 8 verified cross-linguistic cognate pairs anchored in Quranic vocabulary, plus a 4-bar verification rubric and a Tier B ledger of evaluated candidates.
> **Relationship to other vault files:**
> - The 17 English↔Arabic cognate cases curated from Mazen Hammoude's "Beyond the Word" series are in [`beyond-the-word-examples.md`](beyond-the-word-examples.md). **Zero overlap** with the 8 pairs here, these are independent contributions from a Quranic methodology, not the popular-cognate corpus.
> - For the underlying letter-charge framework these cognates rest on, see [`../03-scholar-extracts/consensus-letter-charges.md`](../03-scholar-extracts/consensus-letter-charges.md).
> - The Arabic sound-shift laws (K↔Q, F↔P, …) referenced in the rubric overlap with خشيم's قوانين الإبدال in `Data raw/Languistic theories/علي فهمي خشيم/`.

---

## The principle

Some Arabic Quranic concept-words share a **consonant skeleton** with words in other language families that carry the **same nucleus of meaning**. When this happens, the methodology demands that we name the cross-linguistic resonance, both because it is true (the same human mouth made these shapes for the same human realities) and because it gives the modern English/French reader a pre-installed handle on what the Arabic is actually doing.

A faithful translation, on this methodology, is **the rendering that lets the consonant skeleton sound through**, even across languages.

## The canonical example: ك-ف-ء ↔ *copy* / *copie*

Sūrat al-Ikhlāṣ 112:4, «وَلَمْ يَكُن لَّهُ كُفُوًا أَحَدٌ»

| Language | Word | Skeleton | Meaning |
|---|---|---|---|
| Arabic | كُفء (kufu') | K-F-' | the matching second instance, the equivalent that lines up against another |
| Hebrew | כפל (kāphal) | K-F-L | doubled, folded-onto-itself |
| Latin | copia (medieval scribal) | K-P | the duplicate of an exemplary original |
| English | copy | K-P | a reproduction, a second instance that matches the first |
| French | copie | K-P | same as English |

The K-F nucleus across all five carries: *a fold-onto-itself, a layer beside its image, the second-instance that stands where the first stood.*

- **Conventional translation:** "And there is none equal to Him", accurate but flat.
- **Cross-linguistic translation:** "And there has never been a copy of Him", sharper, more truthful to what the consonant skeleton names. The modern reader who has known *copy* since childhood (photocopies, digital copies, file copies) knows exactly what this verse denies, before any Arabic is studied.

## The 4-bar verification rubric

A candidate must clear **all four bars** to be promoted to Tier A:

1. **Skeleton match**, the consonant skeleton is genuinely shared, allowing for known phonetic shifts (K↔Q, F↔P, ج↔K, ض↔D, …).
2. **Nucleus identity**, the semantic nucleus matches, not just superficial sound-similarity.
3. **Modern reader-recognition**, the cross-linguistic word is *common enough* in modern English/French that the reader will recognize it (no rare technical Latin).
4. **Dual-family confirmation**, at least one Semitic cognate (Hebrew, Aramaic, …) AND one Indo-European cognate (Latin, Greek, Germanic, …).

### Recognized phonetic shifts

- **K ↔ Q**, Arabic ق and PIE *kʷ-/k- share place-of-articulation (back of throat). Interchangeable for this methodology.
- **F ↔ P**, both labial fricatives/stops sharing the lip-shape. Well-attested across language families.
- **ج ↔ K/G**, well-attested (German *Kamel* = Arabic جمل, *Katze* = قط).
- **ض ↔ D**, emphatic-to-dental shift recognized.
- **ص ↔ K/G**, partially attested (Arabic *صين* = Greek/Russian *Kina*, China).

These shifts don't disqualify a cross-linguistic move when the rest of the rubric clears.

## Tier A, confirmed cognates (apply these now)

Each row has been verified for both consonant skeleton and semantic nucleus, with a designated **home-base verse** where the full apparatus is established.

| Arabic root | Cognate | Skeleton | Nucleus | Home-base verse |
|---|---|---|---|---|
| ك-ف-ء | copy / copie / kāphal (Heb) | K-F | the second-instance that lines up against the first | 112:4 ✅ shipped |
| س-ل-م | shalom (Heb) / shalmā (Aram) | S-L-M | bound completeness, peace-as-wholeness | 97:5 «سَلامٌ هِيَ» |
| ر-ب-ب | rabbi / rav (Heb) / rabbā (Aram) | R-B | the expansive holder, master-by-nurture | 1:2 «رَبِّ الْعَالَمِينَ» |
| ج-ن-ن | genie / genus / genus → kind, genus → born-from | G-N | born-from-a-hidden-source, the kin-bearing concealment | 114:6 «الْجِنَّةِ وَالنَّاسِ» |
| ك-ل-م | claim / clamor (Lat. *clamare*) | K-L-M | utter-with-intent, the spoken claim that lands | 2:75 «كَلَامَ اللَّهِ» |
| ق-و-ل | call (PIE *\*kel-*, Lat. *calare*) | K/Q-L | the summon-uttered, the audible-reach | 112:1 «قُلْ» |
| ش-ر-ب | sorb / absorb / syrup (Lat. *sorbere*) | S-R-B | the take-in of fluid, absorbing-flow | 2:60 «وَاشْرَبُوا» |
| ج-م-ع | Gemini / gemellus (Lat.) | G-M / J-M | gathered-into-pair, together-as-one | 2:148 «يَأْتِ بِكُمُ اللَّهُ جَمِيعًا» ✅ shipped |

## Tier B, verification ledger

Candidates evaluated against the 4 bars, with verdicts:

| Root | Proposed cognate | Skeleton | Nucleus | Reader recog. | Dual-family | Verdict |
|---|---|---|---|---|---|---|
| ك-ت-ب | Gr. κτύπος / Eng. *type* / *stamp* | K-T preserved in Gr.; *type* lost the K; *stamp* is S-T-M (different) | inscribed-mark ✓ | ✗, *type* has drifted, *stamp* has wrong skeleton | Heb. כתב (katab) ✓ but no clean IE | **KEEP TIER B**, needs a more legible English cognate |
| ع-ل-م | Lat. *signum* / Gr. σῆμα | S-N vs ʿ-L-M, fails | marking-as-knowing ✓ | n/a | n/a | **REJECT**, skeleton incompatible |
| ج-م-ع | Lat. *Gemini* / *gemellus*; Eng. *com-/con-* | G-M ↔ J-M ✓ (J↔G well-attested) | gathered-into-pair, together-as-one ✓ | ✓, *Gemini* universally known | Heb. גמל (gamal) Semitic ✓; Lat./IE Gemini ✓ | **PROMOTED TO TIER A**, Gemini route only |
| ق-ر-أ | Lat. *legere* / Fr. *lecture* | Q-R vs L-G, fails even with Q↔K | gather-into-utterance ✓ | n/a | n/a | **REJECT**, skeleton fails |
| ن-و-ر | Lat. *lumen* | N-R vs L-M, fails | light ✓ | n/a | n/a | **REJECT**, skeleton fails |
| خ-ل-ق | Lat. *colere* (cultivate) → *colony*, *cult* | K-L ↔ Kh-L ✓ | bring-into-being vs tend-into-growth, adjacent, not identical | partial | Heb. חלק ✓; IE *colere* ✓ | **KEEP TIER B**, nucleus drifts; revisit with sharper IE cognate |
| ك-ت-م | Lat. *occultus* / Gr. κρύπτω | occultus K-L-T; κρύπτω K-R-P-T, neither matches K-T-M | conceal ✓ | n/a | n/a | **REJECT**, no clean cognate |
| ر-أ-ي | Lat. *radius* / *retina* | R alone matches; rest diverges | seeing ✓ partial | n/a | n/a | **REJECT**, only R lines up |
| ض-ح-و | Lat. *dies* / Eng. *day* / PIE *\*dyeu-* | ض → D ✓; ḥ → ∅ in IE | midday-brightness, the open-light-of-day ✓ | ✓, *day* universally known | IE side **very strong**; Semitic side weak | **KEEP TIER B**, IE side dazzling; needs stronger Semitic anchor |
| ي-ت-م | Turk./Pers./Urdu *yatīm* | direct match | direct match | ✓ but **loanword from Arabic** | n/a, not parallel emergence | **NOT APPLICABLE**, historical loanword chain, not a phenomenological cognate |

### Verification pass summary

- **Promoted to Tier A:** ج-م-ع ↔ Gemini (1)
- **Kept in Tier B:** ك-ت-ب, خ-ل-ق, ض-ح-و (3)
- **Rejected (skeleton or nucleus fails):** ع-ل-م, ق-ر-أ, ن-و-ر, ك-ت-م, ر-أ-ي (5)
- **Not applicable (loanword):** ي-ت-م (1)

### Use of Tier B candidates

When a Tier B candidate is referenced, write it cautiously: *"The K-T skeleton appears also in the Greek κτύπος, though the connection bears further study"*, flagged as not-yet-codified. Never apply the full Tier A apparatus to a Tier B root.

### Rejected candidates: stay rejected

The five rejected candidates failed the skeleton or nucleus bar. Don't reintroduce them through a side door (e.g. a coincidental sound similarity in a passing context). The methodology's strength is its discipline.

## Anti-patterns

- **Don't force a sound-similar word that has a different nucleus.** Arabic «بَر» (land) is not cognate with English «bar» (rod, drinking establishment) despite sharing B-R; the nuclei diverge completely.
- **Don't use rare or specialized cognates the average reader won't recognize.** *Lumen* might be technically correct but most English readers don't have it ready-to-hand.
- **Don't dress up a coincidence as a discovery.** If the skeleton matches but the nucleus is forced, the cross-linguistic claim is *weaker* than no claim at all, it discredits the methodology.
- **Don't translate-down to fit the cognate.** The Arabic always governs. The cross-linguistic word is a *handle* for the modern reader, not a replacement for the Arabic semantic field.

## Phenomenological vs etymological, the framing

This is **not** an etymological claim about historical loan-words. It is a **phenomenological** claim: that the human mouth, across cultures, made certain shapes for certain realities, and that those shapes are still legible to anyone who listens.

If a careful Indo-European linguist insists the historical etymology of *copy* (Latin *copia* < PIE \*op-) does not connect to Semitic ك-ف-أ, that does not undo the move. The move is **not** "these words descend from one ancestor." The move is **"these words, in their use *now*, name the same human reality through the same mouth-shape."** Two converging arrivals to the same nucleus, regardless of route.

That is what makes the rendering "no copy of Him" *truthful* in a way "no equal to Him" is not. The English reader hears *copy* and immediately understands the impossibility of reproducing God. They never quite hear *equal* with that force.

## Methodology aphorism

> **الكَلِمةُ تَعبُرُ ألسُنَ البَشَر، ومَعناها يَبقى ثابِتاً في هَيكَلِها الصَّوتيّ.**
>
> *The word crosses the tongues of humans; its meaning stays fixed in its sonic skeleton.*

## How this complements the existing cross-linguistic corpus

| Source | Corpus | Style | Confidence model |
|---|---|---|---|
| [`beyond-the-word-examples.md`](beyond-the-word-examples.md) | 17 popular-cognate cases (rack/رق, burglar/برج, anchor/أنجر, cold/جليد, finger/بنصر, inspire/نفس, …) | Etymological pathway with sound-shift rules | Single-rubric: phonetic match + semantic match |
| **This file** (Tafsir Coran) | 8 Quranic-anchored Tier A pairs (kufu'/copy, salām/shalom, rabb/rabbi, jinn/genie, kalām/claim, qul/call, sharb/sorb, jam'/Gemini) | Phenomenological framing, home-base-verse protocol | 4-bar rubric + Tier A/B/Reject ledger |

The two sets are complementary, not redundant: the Mazen corpus is broad and popular; the Tafsir Coran corpus is narrow, Quran-anchored, and gated by a stricter rubric. Both are inputs to the future LV4 cross-linguistic layer described in [`../02-architecture/lv1-architecture.md`](../02-architecture/lv1-architecture.md).