| Work / author / dates | al-Ḥikam al-ʿAṭāʾiyya (الحكم العطائية), "The Aphorisms" · Tāj al-Dīn Abū al-Faḍl Aḥmad b. Muḥammad b. ʿAbd al-Karīm Ibn ʿAṭāʾ Allāh al-Iskandarī (d. 709/1309), the third master of the Shādhiliyya after Abū al-Ḥasan al-Shādhilī (d. 656/1258) and his own teacher Abū al-ʿAbbās al-Mursī (d. 686/1287). |
| Part added | The whole work: 263 aphorisms in the 25 traditional chapters, the four letters (mukātabāt) and the intimate prayers (munājāt): 300 passages, 5,840 words. Nothing queued. |
| Reader route | /read/hikam (contents) · /read/hikam/ch1 … /read/hikam/ch25 · /read/hikam/letters · /read/hikam/munajat · about /read/hikam/about · Ibn ʿAbbād side by side /read/hikam/commentaries |
| Kind | Sufi handbooks (sufi-manuals), §14 |
| Status | Transcribed from the page images of the Cairo 1319/1901 print and read twice by independent AI readers; collated against the Būlāq print of 1290/1873 and the bare matn of the Ḥanafī reprint · AI-vowelled throughout (the prints are bare) · AI English ✦ unreviewed · not yet checked by a person · reviewed by: — |
Structure. The aphorisms, short rhymed or balanced maxims, without headings in the author's text; the tradition divides them into 25 chapters (Ibn ʿAjība's abwāb; Danner prints 25 too), each opening, in many copies, with «wa-qāla raḍiya Allāhu ʿanhu». The reader pages them by these chapters (one page each, 6 to 17 aphorisms). The first chapter runs from works and hope (¶1, «a sign of relying on works is the loss of hope when a lapse occurs»), divestment and the means (¶2), the aspirations that cannot breach the walls of decree (¶3) and giving up self-direction (¶4, arīḥ nafsaka min al-tadbīr) to the heart's polished mirror (¶13) and the nine «how could He be veiled» (¶16). Later chapters turn on the inrushes (wāridāt) and states, contraction and expansion (qabḍ, basṭ), light and inner sight (nūr, baṣīra), obscurity and retreat (¶11 «bury your being in the earth of obscurity»), servanthood (ʿubūdiyya) and need (fāqa), the gifts and the withholding, reflection (¶261–263 on fikra). The letters (¶264–267): three letters to brethren and an answer on the ḥadīth «the coolness of my eye is in the prayer», as Ibn ʿAjība counts them («three letters and an answer»). The munājāt (¶268–300): the intimate prayers, each opening «Ilāhī» («My God, I am the poor one in my wealth …»), closing with «You are the One besides whom there is no god …» and «How could You be hidden, when You are the Manifest …».
1. Witnesses
| Siglum | Witness | Kind | Vowelled? | Licence | Location |
|---|---|---|---|---|---|
| A | Ibn ʿAbbād al-Rundī (d. 792/1390), Sharḥ ʿalā matn al-Ḥikam (= Ghayth al-mawāhib al-ʿaliyya), with ʿAbd Allāh al-Sharqāwī's (d. 1227/1812) commentary in the margin. Cairo: al-Maṭbaʿa al-ʿĀmira al-Adabiyya, Sūq al-Khuḍār al-Qadīm, for Aḥmad ʿAlī al-Malījī near al-Azhar; printed at the end of Muḥarram 1319 (May 1901), per the colophon (pdf 238). 234 printed pages. The matn is quoted aphorism by aphorism between ornamental brackets ﴿ ﴾ in Ibn ʿAbbād's main block; al-Sharqāwī's margin quotes it again in round brackets. | commentary print (quotes the whole matn) | no | public domain (1901) | archive.org AAlexandrina-142434 (Bibliotheca Alexandrina, 300 ppi grey); every passage links its page image. Local: data/raw/arabic/hikam/ (ACQUISITIONS.md, SHA-256 per file) |
| B | The same two commentaries, Būlāq: Dār al-Ṭibāʿa al-ʿĀmira, Rajab 1290 (1873), corrector Ibrāhīm ʿAbd al-Ghaffār al-Dasūqī (colophon, pdf 265); both parts, though catalogued as vol. 1 (Āṣafiyya library, Hyderabad, copy). | commentary print | no | public domain | archive.org AAlexandrina-148405 |
| J | The bare matn printed at the end of Ibn ʿAjība's (d. 1224/1809) Īqāẓ al-himam fī sharḥ al-Ḥikam, 2 vols in 1, «al-ṭabʿa al-ūlā» (Cairo: ʿAbd al-Ḥamīd Aḥmad Ḥanafī, n.d.), vol. 2 pp. 282–290 (pdf 466–474). Date UNVERIFIED (HathiTrust catalogues the same edition as «[1906]»; the publisher's dated imprints run c. 1930–65). | edition (the common printed text) | partly | a plain reprint of a pre-modern text; date unverified | archive.org AAlexandrina-141592 |
| O | OpenITI 0709IbnCataAllahSikandari.HikamCataiyya.JK009323-ara1, ed. Muḥammad ʿAbd al-Raḥīm (Damascus: Dār al-Īmān, 1406/1986), from al-Jāmiʿ al-Kabīr. Unnumbered, run-on, with lines repeated and stretches missing; its pages match J's typesetting page for page, so it probably derives from J. | digital | no | CC BY-NC-SA 4.0 | GitHub. A typing aid only |
| — | Critical edition and translations (cited only, not held): P. Nwyia, Ibn ʿAṭāʾ Allāh et la naissance de la confrérie šāḏilite (Beirut: Dar el-Machreq, 1972), the critical text from manuscripts with French; V. Danner, Sufi Aphorisms (Kitāb al-Ḥikam) (Leiden: Brill, 1973), doi:10.1163/9789004661431, and The Book of Wisdom (New York: Paulist, 1978). | copyrighted | not held |
2. Base and why
A, the matn as Ibn ʿAbbād quotes it: a public-domain print, whole, legible at 300 ppi, and the commentary of the order's first great commentator, so that the text comes with his division of it and his commentary on every unit (§9). B is the earlier print of the same commentaries (Būlāq 1873) and was used to correct A; J is the common printed text, careless but independent of the commentators' quotation. Nwyia's critical edition is in copyright and not held. Our text is a working edition, not an authority (Protocols §12): where it differs from the critical text we do not know it.
3. Transcription and collation
- First reading: 22 AI readers, ten pages each, from A's page images at ×2 or more, letter by letter, keeping the print's spelling (
data/raw/arabic/hikam/work/TRANSCRIBE.md); O allowed as a typing aid, every word checked against the page; for each unit the print page, where Ibn ʿAbbād's commentary on it ends, and the opening of his commentary (§9). Draftswork/draft/c*.json. - Second reading: 12 other readers checked every page and every draft entry against the images (
work/PROOF.md,work/proof/r*.json): 25 corrections (mostly in the glosses; in the matn e.g. ¶8 «اذا فتح» for a misread «افتتح», ¶243 «ولا يطلب» for «أو يطلب»), 2 aphorisms the first reading had missed (¶33 «تشوّفك إلى ما بطن فيك من العيوب …» and ¶34 «الحق ليس بمحجوب …», on a page next to a misbound leaf: A binds printed p. 29 as pdf 35), 3 headings, and the decisions on which pieces continue an aphorism. - Collation against B (whole text, three readers) and J (one reader),
work/collation/: B agrees with A word for word except in 18 places; J differs in 120 (63 equal variants, 27 J's own slips, 24 where A is better, 6 where J is). 11 corrections adopted (data/sources/hikam-corrections.json, logged inhikam-base.json→changes, type "collation"):
| ¶ | A | Adopted | Basis |
|---|---|---|---|
| 6 | أمد الاعطاء | أمد العطاء | B; A's own margin |
| 37 | تشهدك (doubtful) | يشهدك | B, J |
| 100 | الأسباب (damaged) | الأسباب | B |
| 154 | مدده النور | مدده من النور | B, J, O |
| 165 | شئ من أحبه | شئ ومن أحبه | B, O |
| 170 | عنايته (broken type) | عنايته | B, J |
| 208 | لايمكن (broken type) · الا والله عليك | لايمكن · الا ولله عليك | B, J · B |
| 216 | الالهية عليك هدمت | الالهية اليك هدمت | B |
| 264 | وان كانت بالله بدايته | وان من كانت بالله بدايته | B, J, A's margin |
| 267 | عنه وكرمه | بمنه وكرمه | B, J |
- Mechanical: the brackets removed; pieces of one aphorism quoted apart joined; 71 particles set joined to the next word (لاتخرق, لايكن, مايليق …, a typographic habit of the print) separated, each logged (forms of li- + article such as لاهله are kept).
4. Decisions and doubtful readings (kept as printed; hikam-structure.json → uncertain, shown in each passage's apparatus)
- ¶252 «… بوجود الجزاء عليها عاجلا»: so A and B; J (and, we think, the common text) «آجلا», which the sense wants (the fruit tasted now is good news of the reward later). Kept; the English reads with the sense and says so.
- ¶182 «لم يصمت اذا أساءه» (A, read twice) against B «أساء» (main block and margin). Kept; both construe.
- ¶161 «ومدواة» (A) for the regular «ومداواة» (J). Kept; read mudāwāt.
- ¶243 «عوضا ولا يطلب منه غرضا» (A) against al-Sharqāwī's margin and J «أو يطلب»: the same sense.
- ¶300: the closing formula «والله الموفق وبه أستعين» stands inside Ibn ʿAbbād's quotation in A and B; J has «والله الموفق» only. Not part of the last prayer; recorded in the apparatus.
- Variants al-Sharqāwī reports from other copies (on ¶119, 120, 161, 172, 184, 196, 198) and Ibn ʿAbbād's own remark on ¶52 (he argues the author wrote «للقبول», a copyist «للقلوب»; the main block prints «للقلوب», his margin «للقبول»), and on ¶261 («الاعتبار» in some copies for «الاغيار»), are in the readers' notes (
work/proof,work/collation/B2.json), not yet in the apparatus.
5. Quotations
No Quran quotation was tagged when the text was established; scripts/quran_refs.py finds 24 by skeleton (shown as found automatically), among them Q 6:91 «qul Allāh thumma dharhum fī khawḍihim yalʿabūn» (¶32, and twice in the letters), 65:7 (¶31), 6:18 (¶34), 22:46 (¶42), 10:58 (¶59 and the letters), 17:20 (¶69), 31:20 (¶75), 18:45 (¶198), 75:18–19 (¶215), 27:34 (¶216), 21:18 (¶217), 17:80 (the first letter). The ḥadīth «innamā al-aʿmāl bi-l-niyyāt … fa-hijratuhu ilā mā hājara ilayhi» is quoted in ¶43 and «juʿilat qurratu ʿaynī fī al-ṣalāt» is the subject of ¶266 [the collections not traced here: UNVERIFIED].
6. Vowelling
A is unvowelled, so every mark is AI (scripts/recitals/prepare.py hikam, headless Claude Opus, letters verified unchanged; cached per passage under data/raw/cache/hikam/vowel/). Words with two defensible vowellings are listed per passage (ambiguous). CAMeL disagrees with the AI vowelling on 276 words.
7. Numbering, segmentation and word IDs
Numbering: Ibn ʿAbbād's division. One number per aphorism as his commentary quotes the matn between ﴿ ﴾, pieces of one aphorism quoted apart joined (the nine «kayfa yutaṣawwaru an yaḥjubahu shayʾ» as one, ¶16; «wa-la-an taṣḥaba jāhilan» with «aṣl kull maʿṣiya», ¶36). Two brackets that each hold two maxims (pp. 112, 114) are kept as one unit, as printed. That gives 263 aphorisms. The counts in print differ: Danner 262 (in 25 chapters), al-Sharnūbī 264 (per Danner) and the Matheson Trust's 264, al-Būṭī 259 (per his commentary's numbering), Ibn ʿAjība about 267 (soufisme.org); no pre-1931 Cairo print we saw numbers them. Our numbers have not been checked aphorism by aphorism against any of these, so a number may be off by one or two from a printed translation: match by the opening words. The letters are ¶264–267 and the munājāt ¶268–300, one number per «Ilāhī» (pieces without «Ilāhī» joined to the prayer before): 33, where Ibn ʿAjība numbers 32 (his 1–2 are our 268; his 9 spans our 275–276; our 295 and 298 fall inside his 27 and 29; his 31–32 are our 300). Chapters: the 25 abwāb of Ibn ʿAjība, found on his print's pages (work/ajiba-chapters.json; he names each in his commentary, «as he said at the beginning of the eighth chapter»): 1 (¶1), 2 (18), 3 (33), 4 (39), 5 (44), 6 (49), 7 (61), 8 (70), 9 (76), 10 (90), 11 (106), 12 (113), 13 (127), 14 (133), 15 (144), 16 (150), 17 (158), 18 (168), 19 (173), 20 (180), 21 (193), 22 (205), 23 (213), 24 (223), 25 (238). A marks some of these openings with «wa-qāla raḍiya Allāhu ʿanhu», not all; J marks none. Paged reader: one page per chapter, then the letters and the munājāt (parts in hikam-base.json; lib/pagedProse.ts); /read/hikam#p-N and ?w=hk:N:… are sent to the right page. Word IDs hk:<passage>:<word>.
8. Analysis layer
CAMeL Tools 1.6 via scripts/fusus/analyze_ch.py hikam.json hikam-analysis.json hk: 5,840 words, 3,463 with a root, 578 distinct roots; 8 malformed roots (a shadda or a proclitic left in the root string: لله, بالله, شتان, أيهم) fixed by scripts/recitals/hikam_fix_analysis.py.
9. Commentaries aligned
Ibn ʿAbbād (A): for every passage, the opening of his commentary (25–60 words, his paraphrase; transcribed and read twice with the matn) with an AI English (scripts/recitals/hikam_commentary.py, ✦ unreviewed), and the exact pages of the print over which his commentary on that passage runs, each linked to its page image. The rest of his commentary is linked, not transcribed: the mapping is complete (every passage → first and last pdf page of A), the transcription of the whole commentary (about 230 dense pages) is a possible later step. Ibn ʿAjība: his 25 chapters mapped to our numbers (above) with his print's pages; aphorism-level mapping not done (plan: his commentary opens each aphorism with «(qāla …)» and his own with «(qultu)», so a reader per chapter can list the pages as for A). Zarrūq: no pre-1931 print found (cited in the companions). al-Sharqāwī: in A's margin; his variant notes used in §4. al-Sharnūbī: a 1322/1904 print is in HathiTrust full view (member download only; not held). al-Būṭī: cited only (in copyright).
10. Translation
AI (Claude Opus) from the vowelled Arabic, scripts/recitals/translate.py hikam opus, told not to reproduce or imitate the published translations and given none; glossary in data/sources/hikam-meta.json (ʿamal "work", rajāʾ "hope", tawakkul "trust", tadbīr "self-direction", irāda "will; the aspirant's resolve", qabḍ/basṭ "contraction/expansion", wārid "inrush", ḥāl/maqām "state/station", nūr "light", baṣīra "inner sight", fikra "reflection", ʿubūdiyya "servanthood", ʿizza "might", fāqa "destitution", tajrīd/asbāb "divestment/the means" …). Word glosses and term notes from the same run. All ✦ unreviewed.
10b. Triangulation
scripts/triangulate.py hikam. No translation is held: none before 1931 exists (§11), and Danner (1973, 1978) and Nwyia (1972) are in copyright and not held privately. So the comparison ran on the commentator of the author's own line: for every passage our AI English was checked against Ibn ʿAbbād's opening gloss (lineage evidence, weighted as such; scripts/triangulate.py → LINEAGE, which now also names Ibn ʿAjība, Zarrūq and al-Sharqāwī). Result: no contradiction; 19 places doubtful, each decided from the Arabic and fixed in our own words (scripts/recitals/hikam_fixes.py, logged in app/src/content/ai/hikam.json → fixes): ¶8 «maʿahā» (given that, not "so long as it is with you"); ¶16 «He is the Manifest before anything existed» (a nominal sentence, not "was"); ¶20 «fīmā siwāhā» (other than it, not "beyond"); ¶24, ¶57, ¶156 al-aghyār glossed as «whatever is other than God»; ¶42 «what will not last with him»; ¶65 «has exposed himself to their vanishing»; ¶92 «what He opens upon their hearts»; ¶105 al-sarāʾir «innermost selves» (two terms, not one); ¶113 «how far above what you ask of Him is what He asks of you»; ¶149 ṭufūliyya with Ibn ʿAbbād's ṭufayliyya bracketed; ¶188, ¶206 («depart from where they alighted»), ¶219 («at once»), ¶230 («its taste»), ¶265 (shāhid), ¶273 («from You toward You»), ¶295 («the garments of Your majesty»). A person should check these. Witnesses differ (curated, scripts/recitals/hikam_differ.py; the reader's marker reads «Witnesses differ» for this text): 7 notes where the prints part or the commentators discuss a reading: ¶16 the nine «kayfa yutaṣawwaru» of Ibn ʿAbbād's prints against the seven of the common text; ¶52 li-l-qulūb (the prints) or li-l-qabūl (Ibn ʿAbbād's conjecture, al-Sharqāwī's text); ¶149 ṭufūliyya (our vowelling) or ṭufayliyya (Ibn ʿAbbād); ¶182 asāʾahu/asāʾa; ¶243 wa-lā/aw; ¶252 ʿājilan/ājilan; ¶261 al-aghyār/al-iʿtibār (some copies, per Ibn ʿAbbād). When a translation is acquired (Danner or Nwyia into (private copy)), rerun triangulate.py hikam (then hikam_differ.py): translators' divergences will be added beside these.
11. Rights and private layers
A, B: public domain (1901, 1873). J: a plain reprint of a pre-modern text, date unverified (used only to compare). O: CC BY-NC-SA 4.0, used only as a typing aid. Nwyia, Danner, the Matheson Trust's edition, Bewley, Michon, al-Būṭī: in copyright, cited only; nothing of their wording is in this reader and none is held privately. No public-domain translation before 1931 was found (searched in HathiTrust full text, archive.org and Google Books: only mentions, in Ben Cheneb 1907, Depont and Coppolani 1897, Massignon 1922 and 1929; Asín Palacios's article on Ibn ʿAbbād and John of the Cross is of 1933). The one public-domain translation known is Ottoman Turkish (Kastamonulu Ahmed Mâhir, el-Muḥkem fī şerḥi'l-Ḥikem, Istanbul 1905–07, HathiTrust), not held. Audio: none made (about 5,800 Arabic and 9,000 English words: under $2 on Azure; the author approves any TTS run above $1).
12. Open questions / to review
- A person should read the transcription against A, above all the doubtful places of §4, and decide ¶252 (عاجلا / آجلا).
- Compare our numbering with a 264-numbered text and with Nwyia's edition (through a library), and give a concordance.
- Acquire Nwyia (1972) or Danner (1973) into the private layer for the triangulation (§17), and al-Sharnūbī's 1904 print (HathiTrust).
- Transcribe more of Ibn ʿAbbād's commentary, or Ibn ʿAjība's opening gloss on each aphorism, if wanted.
13. Resonances (Protocols §16)
scripts/resonances.py whole --only hikam and passages hikam (batches of 6), with the author's pointers given to the model as an editor's hint (FOCUS in the script: consider, never assert); 16 new external references verified by an agent with web search, none removed (data/raw/arabic/hikam/work/ref-verdicts.json). Whole text 15 connections (10 parallels, 3 lineages, 2 contrasts); passages 1,292 (1,121 parallels, 149 contrasts, 18 lineages, 4 shared sources). Most linked texts in the library: al-Qushayrī's Risāla (300 links: the terms the Ḥikam presupposes, waqt, ḥāl, wārid, qabḍ and basṭ, presence and absence), al-Ghazālī's Mishkāt (156: light, the veil, God hidden by the intensity of His manifestation), the Fuṣūṣ (chapters 1, 2 and 5), Plato's Republic and Plotinus VI.9, Ibn Sīnā's ʿIshq (44), the Awrād. Graded: lineage only where transmission is documented: al-Shādhilī and al-Mursī to Ibn ʿAṭāʾ Allāh (Nwyia 1972, Danner 1973); Ibn ʿAbbād's commentary and letters (Nwyia 1961, Renard 1986); al-Qushayrī's vocabulary in the early Shādhiliyya (lineage at the level of the school; the verifier notes that Nwyia supports the general debt, not chapter-by-chapter matches); Ibn ʿAṭāʾ Allāh's own Tanwīr fī isqāṭ al-tadbīr. Parallels ("resonance, not influence"): the Mishkāt, the Fuṣūṣ (the Shādhilīs were not Akbarians in doctrine), Ibn Sīnā's ʿIshq (the Good disclosing itself, veiled only by the recipient), Plotinus VI.9 (the One present to all; and a contrast: the One does not seek, where the Ḥikam's God makes Himself known), Stoic assent to providence (Hadot 1992), de Caussade, Simone Weil, Shinran's other-power; John of the Cross is a parallel, not a lineage: Asín Palacios (1933) proposed that the Shādhilī teaching through Ibn ʿAbbād anticipated him, and López-Baralt (1985) pursued it; influence is disputed, and the verifier could confirm only the article's general thesis (contraction and expansion beside the dark night), not its comparisons with particular chapters. Contrasts include Ibn Sīnā's recitals (the soul's own flight home, against the Ḥikam's arrival by grace).
14. Kind
Sufi handbooks (sufi-manuals), beside al-Qushayrī's Risāla and Shabistarī's Gulshan-i rāz: a work read and taught, for seven centuries, as the primer of the Shādhilī path, on the aspirant's works and hope, the means and divestment, giving up self-direction, the inrushes, states and stations, contraction and expansion, light and inner sight, servanthood and need; its commentators are the order's masters. Not "Philosophical treatises" (it argues nothing; it counsels, in the second person, «you»), not "Poetry" (balanced and rhymed prose, sajʿ, not verse), not "Prayers" (only the last part is prayer). The taxonomy's blurb for Sufi handbooks speaks of «the manuals that set out the Sufi path in its own words»; the Ḥikam is the shortest such manual in the library. One text in 27 parts (25 chapters, letters, munājāt), not three catalogue entries: one author, one transmission, one commentary tradition that comments on all three.
15. Change log
- 2026-10-05: sources found (A, B, J, O); transcribed (22 readers) and read again (12); collated against B and J (11 corrections); 25 chapters mapped from Ibn ʿAjība; vowelled, analysed, translated; Ibn ʿAbbād's openings translated; companions (52 items); version record; resonances; registered at
/read/hikam(27 pages) under Sufi handbooks.