Repository path: workshop/translations/kalila-nasik/collation.md · rendered 2026-09-09

Page metadata (front matter)

type	`note`
id	`kalila-nasik-collation`
status	`frozen`
created	`2026-08-16`
updated	`2026-08-16`
links	`workshop/translations/kalila-nasik/R34-v1/translation.md`, `workshop/translations/kalila-nasik/source-wikisource-nasik.txt`, `workshop/translations/kalila-nasik/witness2-amiriyya-1937.txt`, `workshop/translations/kalila-saih/collation.md`

Copy-text and collation — «باب الناسك والضيف»

1. The witnesses

	witness	provenance	state
copy-text	Arabic Wikisource, `كليلة ودمنة/باب الناسك والضيف`	no edition statement; unvocalised; modern orthography	clean text, 287 words of body in 19 sentences
witness 2	Arabic Wikisource, `كليلة ودمنة (الأميرية، 1937)/باب الناسك والضيف`	Cairo: al-Maṭbaʿa al-Amīriyya (Būlāq) 1937, pp. 299–300, page images present	vocalised; optical recognition heavily damaged, the same damage profile as the `kalila-saih` witness (ث→ن, ج→ح/خ, ي→ى, stray marginal digits)

Same pair of witnesses as ../kalila-saih/collation.md, two pages earlier in the same printing — this chapter immediately precedes «باب السائح والصائغ» in both. Louis Cheikho 1905 was already established as unusable (its scan carries no recoverable Arabic at all); not re-searched.

2. Sentence numbering

The copy-text carries no numbering. Sentences are cut at ., ! and ؟, giving 19 sentences, numbered [1]–[19] in copy-text order. Every citation on the translation page and in the figure inventory uses these numbers.

3. The collation

Read whole, both witnesses, in Arabic. One substantive adoption; everything else is optical noise or punctuation.

3.1 The one substantive correction, adopted from witness 2

sentence	copy-text	witness 2	adopted	why
`[12]`	وزهدت فيها لا تجد	وزهدت فما لا تجد	وزهدت فيما لا تجد	The copy-text's فيها is ungrammatical here — there is no feminine antecedent, and the clause is the matched partner of قنعت بالذي تجد. Witness 2's فما is فيما with the ي lost to the scan. The rendered English is "content to go without what you do not find".

3.2 Places where the copy-text is right and witness 2 is damaged

Recorded so the collation can be tested, not to pad it. In every one of these the copy-text reading is retained.

sentence	copy-text	witness 2 as scanned	reading
`[4]`	بتمر	بمر	ت lost
`[6]`	التمر	السمر	ت→س
`[8]`	بثمار	بنمار	ث→ن
`[9]`	يثقل	ينقل	ث→ن
`[10]`	الأثمار	الأنمار	ث→ن
`[10]`	فما حاجتها	فا حاجتها	م lost
`[10]`	إلى التمر مع وخامته	إلى التمر وخامته	مع lost; without it the chain is ungrammatical
`[12]`	سعيد الجدِّ	سعيد الخد	ج→خ
`[14]`	وعالج	وعالح	ج→ح
`[19]`	لما رأيت	ما رأيت	ل lost
`[19]`	ولم يؤدبه عليه	ولم يود به عليه	ؤدب broken across two tokens

3.3 Variants that are not errors

[1] دبشليم / دبسليم — both spellings of the frame-king's name are attested. The copy-text's دبشليم is retained, matching «باب السائح والصائغ» and this project's existing Dabshalīm.
Punctuation differs throughout (witness 2 prints : where the copy-text prints !, and ، where it prints .). Sentence division in §2 follows the copy-text and is editorial in both witnesses.
Witness 2 closes with a colophon, «انقضى باب الناسك والضيف», absent from the copy-text and not body text. Not rendered.

4. What this collation does not establish

Two Wikisource transcriptions are not a critical edition. Both descend from the Būlāq/Amīriyya tradition, so their agreement is weak evidence about the recension and none at all about what Ibn al-Muqaffaʿ wrote. The one adopted correction is adopted on grammar, with witness 2 as corroboration, not on witness authority.