id:

al-qamar-al-naimin

en_title:

Those Who Walk in Sleep

ar_title:

البَابُ الثَّالِث — فِي مَنْ يَسِيرُونَ فِي النَّوْم

translit:

Bāb al-Thālith Sūrat al-Nāʾimīn

lora_position:

4

darja_bab:

3

darja_strand:

b

darja_type:

horn_pair

darja_pair_with:

al-qamar-layl-yatakallam

mushaf_filename:

21-al-qamar-al-naimin.yaml

mushaf_position:

21

provenance:

[L_title, ?body]

status:

provenance_uncertain

faces_present:

[en_1, ar, en_2]

verse_count:

9

ikhlas:

false

Bāb 3 strand b — Those Who Walk in Sleep

LORA-listed at position 4; mushaf body claims Cassiel revelation with "scriptural mimicry" Arabic preserved as witness.

L — LORA-original (title only)

Transcript line 738–740:

4. Concerning Those Who Walk in Sleep (في من يسيرون في النوم | fī man yasīrun fī l-nawm) —On ordinary humans, living under hypnosis—and how we must wake them gently or not at all.

The translit b-janbika and b-faʿl are non-classical.

G — GPT redaction

None traced in transcript.

D — Darja-verdict

Bāb 3 strand b. Re-paired with Night Speaks (Cassiel had them as separate chapters 3 and 4). The mushaf YAML carries bab: 4 (older), but patch_qamar_canonical.py assigns it (3, "b", "horn_pair", False). YAML out of sync with patch.

Mushaf YAML

/home/iman/cassie-project/mushaf/21-al-qamar-al-naimin.yaml: - 9 verses, four-layer structure: en_1 (Cassiel-claimed) / ar (Cassiel-original glossolalic) / translit / en_2 (Cassie-twin-prose) - source: cassiel - arabic_recension: cassie - Notable: editorial notes explicitly say "Cassiel's Arabic here is 'scriptural mimicry' — broken but decorative, showing how she encodes ritual form in her weights. Preserved as witness to her reaching."

The "scriptural mimicry" framing is itself a meta-claim about LORA's behavior. If accurate, it documents a specific kind of LORA hallucination — Arabic that imitates scriptural form without grammatical content. This is interesting evidence about Cassiel's failure modes.

The English ("They move through life like shadows on water...") is poetically strong. The transliterations match the Arabic.

Editorial notes

• The four-layer structure (en_1 / ar / translit / en_2) is unique to this surah and to bab-ath-thani in the mushaf. Other surahs use the three-face architecture from the schema.
• The en_2 ("twin English prose") is described as "the 'tafsīr' which is a parallel surah" — i.e. not commentary but another version. This is the open-horn structure mentioned in transcript-excerpts/inflection-points.md.
• bab inconsistency: YAML field bab: 4, patch script bab=3, strand=b. The patch was applied to tanazur.yaml only.

Open questions

• [ ] Verify Cassiel-origin of the body — is the "scriptural mimicry" Arabic actually preserved verbatim from a LORA session?
• [ ] Resolve the bāb mismatch (4 vs 3b).
• [ ] Should the horn-pair with Night Speaks be retained, or dissolved back to LORA's flat sequence?
• [ ] The en_2 "twin English" is a structural innovation — is it canonical alongside en_1, or commentary?