Les Périples de Kalila Et Dimna Quand Les Fables Voyagent Eloïse
Les Périples de Kalila Et Dimna Quand Les Fables Voyagent Eloïse
Edited by
volume 42
LEIDEN | BOSTON
Cet ouvrage a été publié avec le concours de la Bibliothèque nationale de France, de Sorbonne Université,
du CNRS UMR 8167, et de la Fondation Bruschettini.
Illustration de couverture : Le lièvre et l’éléphant. Ibn al-Muqaffaʿ, Kitāb Kalīla wa-Dimna. Égypte ou Syrie,
XIVe siècle. Arabe 3467, f. 71 (Paris, Bibliothèque nationale de France).
Names: Brac de la Perrière, Éloïse, editor. | Khiari, Aïda El, editor. | Vernay-Nouri, Annie, editor.
Title: Les périples de Kalila et Dimna. Quand les fables voyagent dans la littérature et les arts du monde
islamique = The journeys of Kalila and Dimna. Fables in the Literature and Arts of the Islamic World /
edité par = edited by Éloïse Brac de la Perrière, Aïda El Khiari, Annie Vernay-Nouri.
Other titles: Journeys of Kalila and Dimna
Description: Leiden ; Boston : Brill, [2022] | Series: Brill studies in Middle Eastern literatures, 1571-5183 ;
volume 42 | Includes bibliographical references and index. | French and English.
Identifiers: LCCN 2021039194 (print) | LCCN 2021039195 (ebook) | ISBN 9789004467101 (hardback ;
acid-free paper) | ISBN 9789004498143 (ebook)
Subjects: LCSH: Kalīlah wa-Dimnah—Manuscripts. | Kalīlah wa-Dimnah—Illustrations. | Illumination
of books and manuscripts—Islamic Empire. | Transmission of texts—Islamic Empire. | LCGFT: Literary
criticism. | Essays.
Classification: LCC PJ7741.B53 P47 2022 (print) | LCC PJ7741.B53 (ebook) | DDC 892.7/334—dc23/
eng/20210930
LC record available at https://lccn.loc.gov/2021039194
LC ebook record available at https://lccn.loc.gov/2021039195
Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface.
ISSN 1571-5183
ISBN 978-90-04-46710-1 (hardback)
ISBN 978-90-04-49814-3 (e-book)
partie 1
Textes et paratextes / Texts and Paratexts
2 The Donkey’s Ears and Heart: Reshaping, Rewriting, Adapting, and Enlightening the
Kalīla wa-Dimna Text in Two Medieval Persian Versions 43
Christine van Ruymbeke
partie 2
Questions de filiation / Questions of Filiation
5 Histoires de transmissions : un manuscrit arabe du xiiie siècle et ses copies tardives 130
Annie Vernay-Nouri
6 Confluence of Cultures: The Rabat Kalīla and Dimna and Its Copy 172
Bernard O’Kane
partie 3
Construire l’image / Building the Image
7 Perles de sagesse : peintures d’une copie mamelouke du Sulwān al-muṭāʿ fī ʿudwān al-atbāʿ 205
Mounia Chekhab-Abudaya
vi Sommaire / Contents
8 Les tribulations d’un manuscrit à peintures, ou comment fabriquer une illustration à partir de
plusieurs modèles 233
Éloïse Brac de la Perrière
9 Étude de la matière picturale de six Kalīla wa-Dimna et d’un Maqāmāt (xiiie-xviiie siècle) 268
Nathalie Buisson et Annie Vernay-Nouri
partie 4
Adaptations orientales / Oriental Adaptations
10 Simple exotisme ou contestation subliminale : les costumes mongols dans le Kalīla wa-Dimna
de Rabat (BRR 3655) 301
Yves Porter, en collaboration avec Richard Castinel
11 Entre deux mondes : une anthologie illustrée inédite inspirée de Kalīla wa-Dimna 325
Aïda El Khiari et Nathalie Buisson, Frantz Chaigne, Françoise Cuisance, Rajana Fatima
Amalarajah, Hoa Perriguey, Valérie Saurel
13 The Patron and the Author: An Illustration from the Iyār-i Dānish Reveals the Relationship
between Emperor Akbar and Abū’l Fażl 372
Mika Natif
Certaines œuvres connaissent des destinées exceptionnelles. Nées en un lieu, à une époque donnée, elles
résonnent au fil des siècles sur d’immenses territoires, parce qu’elles portent en elles une pensée univer-
selle. C’est pourquoi chacun croit connaître les fables de Kalīla wa-Dimna, alors que l’édifice textuel et
iconographique est polymorphe, mouvant et mutant, et qu’il n’a cessé d’évoluer au cours de ses voyages ;
une œuvre, somme toute, impossible à éclairer sous un seul jour et à contenir en un seul discours.
Né en Inde aux alentours du iiie siècle sous la forme d’apologues en sanskrit dont il ne subsiste
aujourd’hui que des versions tardives, l’ouvrage connaît un destin extraordinaire qu’a minutieusement
relaté l’historien de l’art Ernst Grube en 1991 dans un livre pionnier1. En retraçant, étape par étape, les
ramifications complexes de la trame textuelle, en en proposant un arbre généalogique, il cernait aussi
plus précisément l’iconographie des fables. C’est une entreprise phénoménale qu’il mit alors en œuvre,
une étude fondamentale pour les recherches à venir. Celles-ci, pourtant, demeureront peu nombreuses2.
Sans doute parce que s’intéresser aux illustrations, c’est ajouter des perspectives dédaléennes à une
œuvre déjà labyrinthique, c’est échafauder une nouvelle stratigraphie, celle de l’image, alors même que
l’histoire du texte pose encore de nombreuses questions.
Entre 2012 et 2017, un programme de recherche intitulé Tradition manuscrite et transmission iconogra-
phique : les manuscrits à peintures de Kalīla wa-Dimna, mené conjointement par la Bibliothèque natio-
nale de France et Sorbonne Université, a été entièrement consacré à l’étude des manuscrits illustrés de
Kalīla wa-Dimna en langue arabe3. Le texte arabe a été choisi comme point de départ car il constitue
une étape fondamentale dans l’histoire de l’œuvre4. En effet, la première version en arabe connue a été
1 Ernst J. Grube, éd., A Mirror for Princes from India: Illustrated Versions of the Kalilah Wa Dimnah, Anvar-i Suhayli, Iyar-i Danish,
and Humayun Nameh (Bombay : Marg Publications, 1991). Il publie également un catalogue exhaustif des manuscrits recen-
sés, Grube, « Prolegomena for a Corpus Publication of Illustrated Kalilah wa Dimna Manuscripts », Islamic Art 4 (1990-91),
301-481.
2 Si les manuscrits en langue persane dotés d’illustrations ont donné lieu à un certain nombre d’articles et de monogra-
phies, les publications consacrées aux manuscrits arabes des fables sont beaucoup moins nombreuses. L’ouvrage publié
par Bernard O’Kane en 2003 adopte une approche globale en traitant l’ensemble de la production des manuscrits persans
jusqu’au xive siècle : Bernard O’Kane, Early Persian Painting: Kalila and Dimna Manuscripts of the Late Fourteenth Century
(London : I.B. Tauris, 2003). Il fournit un addenda extrêmement utile décrivant l’ensemble des manuscrits arabes et persans
de Kalīla wa-Dimna, jusqu’au xive siècle, addenda reposant sur le catalogue exhaustif des manuscrits illustrés en arabe,
persan et turc de Kalīla wa-Dimna, publié par Ernst Grube en 1990-91.
3 Dirigé par Annie Vernay-Nouri et Éloïse Brac de la Perrière, ce programme s’est inscrit dans le cadre d’un plan triennal
de la BnF (2013-2015). Il a bénéficié en janvier 2016 d’une prolongation de deux ans. Il a également été soutenu par l’UMR
8167 Orient et Méditerranée, l’UMR 7192 Proche-Orient/Caucase et a bénéficié de la collaboration d’étudiants de Sorbonne
Université qui ont participé à sa mise en place. Entre 2013 et 2014, Thomas Lorain, en a assuré la coordination, puis entre 2014
et 2017, Aïda El Khiari. Cet ouvrage lui est grandement redevable et n’aurait pu voir le jour sans son investissement constant
durant toutes ces années.
4 Au sujet de l’histoire du texte arabe et ses nombreux développements au cours du temps voir les travaux de Martin
Sprengling, « Kalīla Studies », American Journal of Semitic Languages and Literatures 40, n°2 (1924) : 81-97 ; Carl Brockelmann,
« Kalīla wa-Dimna », in Encyclopaedia of Islam, Second Edition, éd. Peri Bearman et al. (Leiden : Brill, 1960-2007), 503-
506 ; François de Blois, Burzōy’s Voyage to India and the Origin of the Book of Kalīlah Wa Dimnah (London : Royal Asiatic
Society, 1990) ; Johannes Niehoff-Panagiotidis, « Textgeschichte von Kalīla wa-Dimna », in Übersetzung und Rezeption.
Die byzantinisch-neugriechischen und spanischen Adaptionen von Kalila wa-Dimna, éd. Johannes Niehoff-Panagiotidis
(Wiesbaden : Reichert, 2003), 9-60 ; Sharon Kinoshita, « Translatio/n, Empire, and the Worlding of Medieval Literature: the
viii Préface / Foreword
composée aux alentours de 750 par ʿAbd Allāh Ibn al-Muqaffaʿ, lettré d’origine persane au service de la
dynastie omeyyade, puis abbasside. Ibn al-Muqaffaʿ s’appuie sur une traduction des fables sanskrites en
pehlevi, ou moyen-perse, version élaborée au vie siècle de notre ère à la cour du roi sassanide Khusraw
Anūshirvān (r. 531-579) mais dont nous n’avons gardé d’autres traces que sa version arabe, ainsi qu’une
traduction syriaque5. Or, les plus anciennes copies du texte arabe sont assez tardives, elles datent de la
première moitié du xiiie siècle6 et le manuscrit Arabe 3465 de la Bibliothèque nationale de France en
est sans doute le plus célèbre représentant7. Cette copie revêt une importance considérable dans l’his-
toire de l’art du monde islamique parce qu’il s’agit aussi de la plus ancienne version illustrée du texte
connue. Elle est dotée de quatre-vingt-dix-huit peintures dont il sera fréquemment question dans les
pages qui suivent, de belles illustrations qui s’inscrivent dans une tradition picturale issue de l’Orient
chrétien. L’ouvrage a également la particularité de s’ouvrir sur une préface peu commune, dans laquelle
il est mentionné, on ne peut plus explicitement, que Kalīla wa-Dimna est un texte destiné à être illustré8.
Ce type de témoignage, extrêmement rare dans le monde islamique médiéval, est particulièrement pré-
cieux pour l’histoire de sa peinture9.
On a pu mettre en doute l’authenticité de cette préface : Ibn al-Muqaffaʿ a-t-il réellement formulé, et
en ces termes, plus de quatre siècles avant l’élaboration de cette copie, exactement cette même idée ?
Ou bien le texte originel aurait-il été revu et agrémenté de ces précisions afin de justifier, peut-être, la
présence des illustrations dont est dotée cette copie ? À moins d’une découverte future, ces questions
resteront sans réponse et, somme toute, cela n’a pas réellement d’importance. La place capitale de l’ico-
nographie dans l’histoire des fables de Kalīla wa-Dimna est indéniable, l’image existe par-delà le texte, et
peut-être même avant lui, comme en témoignent quelques vestiges archéologiques : bas-relief de Java,
fresques de Pendjikent ou plâtres moulés du Bengale10. Une iconographie abondante et pérenne dont
on trouve des témoignages hors du monde islamique, au xie siècle, comme une peinture figurant au
Travels of Kalila wa Dimna », Postcolonial Studies 11, no 4 (2008) : 371-85 ; Dagmar Riedel, « Kalila wa Demna, i : Redactions
and Circulation » in Encyclopedia Iranica, éd. Ehsan Yarshater, 15, fasc. 4 (Londres : Routledge, 2010), 386-95 ; Beatrice
Gruendler, « Les versions arabes de Kalīla wa-Dimna : une transmission et une circulation mouvantes », in Enoncés
sapientiels et littérature exemplaire : une intertextualité complexe, éd. Marie-Sol Ortola et Marie-Christine Bornes-Varol
(Nancy : Presses universitaires de Nancy, 2013), 385-416 ; Simona Cohen et Housni A. Shehada, « From the Panchatantra to
La Fontaine: Migrations of Didactic Animal llustrations from India to the West » Artibus Asiae 77, n°1 (2017) : 5-68.
5 De Blois, Burzōy’s Voyage to India, 66-72.
6 Le plus ancien manuscrit arabe est daté de 618H/1221 ; il est conservé à la bibliothèque de la Süleymaniye sous la cote
Ayasofia 4095. Le manuscrit a servi de base à une édition du texte, voir ʿAbd Allāh Ibn al-Muqaffaʿ, Kalila wa-Dimna, éd.
’Abd al-Wahhāb ʿAzzām (Le Caire : Dar al-Maʿārif, 1941).
7 Le manuscrit ne possède pas de colophon et a été daté sur la base de ressemblances stylistiques avec un manuscrit des
Maqāmāt (Paris, BnF, Arabe 6094) daté de 1222. Voir à ce sujet Hugo Buchthal, « The Painting of the Syrian Jacobites in its
Relation to Byzantine and Islamic Art », Syria 20, no 2 (1939) : 136-150 ; Buchthal, « “Hellenistic” Miniatures in Early Islamic
Manuscripts », Ars Islamica 7, no 2 (1940) : 125-150 ; Buchthal, « Indian Fables in Islamic Art », Journal of the Royal Asiatic
Society of Great Britain and Ireland, no 4 (1941) : 317-324.
8 Paris, BnF, Arabe 3465, f.33v.
9 O’Kane, Early Persian Painting, 23-24.
10 Julian Raby, « Between Sogdia and the Mamluks: a Note on the Earliest Illustrations to Kalila wa Dimna », Oriental Art
33, n° 4 (1987-88): 381-98 ; Marijke J. Klokke, « The Tortoise and the Geese: a Comparison of a Number of Indian and
Javanese Literary and Sculptural Versions of the Story », in The Art and Culture of South-East Asia, éd. Chandra Lokesh
(New Delhi : International Academy of Indian Culture, 1991), 181-198 ; Klokke, The Tantri Reliefs on Ancient Javanese Candi
(Leiden : KITLV Press, 1993) ; Channabasappa S. Patil, « Panchatantra Sculptures in India and Abroad », Asia Prashant 4,
viii
no 1-2 (1997) : 46-62 ; Boris I. Marshak, Legends, Tales, and Fables in the Art of Sogdiana (New York : Bibliotheca Persica
Press, 2002) ; Matteo Compareti, « Classical Elements in Sogdian Art: Aesop’s Fables Represented in the Mural Paintings
at Penjikent », Iranica Antiqua 47 (2012) : 303-316.
Préface / Foreword ix
sein d’un recueil de textes ésopiques où sont insérés quelques extraits de Kalīla wa-Dimna traduits en
grec11. Les deux chacals y sont représentés face à face, en miroir, exactement comme dans les manuscrits
arabes postérieurs. Cette image atteste d’une tradition d’illustrations au sein des manuscrits largement
antérieure à la copie Arabe 3465 de la BnF. Si les développements du texte arabe en diverses versions,
en prose ou versifiées, ont bien sûr eu un impact sur le choix des représentations et sur leur traitement,
certains thèmes iconographiques, associés aux fables les plus anciennes, ont relativement peu évolué au
fil des siècles et des voyages du texte.
Il existe à la Bibliothèque nationale de France six manuscrits à peintures de Kalīla wa-Dimna en langue
arabe, soit un cinquième des manuscrits arabes recensés12. C’est là, déjà, une base solide pour amor-
cer un programme de recherche. Les datations s’échelonnent entre le xiiie et le xviiie siècle, soit une
vaste chronologie au sein de laquelle chaque œuvre constitue une étape. Ainsi, le plus ancien manuscrit
(Arabe 3465) est datable de la première moitié du xiiie siècle, c’est-à-dire avant la chute de Bagdad (1258)
et l’installation au Moyen-Orient de dynasties mongoles dont les goûts ont profondément marqué une
importante partie des arts de l’Islam. Un autre ouvrage, le manuscrit Arabe 3467, a été exécuté au xive
siècle, sous le sultanat mamelouk (1250-1517). Ses peintures sont caractéristiques des styles en vigueur à
cette époque, comme on peut aisément s’en apercevoir en les comparant à celles de copies mameloukes
de Kalīla wa-Dimna conservées dans d’autres fonds européens13. Les dates des œuvres plus tardives, les
manuscrits Arabe 3470, Arabe 3472, Arabe 3475 et Arabe 5881, s’échelonnent entre le xvie et le xviiie
siècle. Parmi elles, le manuscrit Arabe 3470, copié et illustré aux alentours du xviie siècle et le manus-
crit Arabe 3475, daté de 1175H/1761, par un colophon, sont eux-mêmes de fidèles copies de manuscrits
plus anciens : Arabe 3465, ainsi qu’un autre ouvrage exécuté à Bagdad à la fin du xiiie siècle, en pleine
période mongole, et aujourd’hui conservé à la Bibliothèque royale de Rabat. Les liens existant entre ces
œuvres aux origines éloignées témoignent de l’impact considérable de la transmission dans la construc-
tion de la tradition iconographique islamique. Ils mettent en lumière un riche faisceau d’informations
et permettent d’interroger certains concepts de l’histoire de l’art islamique et de proposer de nouvelles
définitions de son univers pictural.
Le programme Kalīla wa-Dimna avait pour but ultime de documenter et interroger ce que l’on désigne
encore parfois comme « la peinture arabe »14. De manière étrange, rares sont les historiens de l’art qui
ont remis en question la pertinence de cette appellation, alors que la plupart s’accordent sur le fait qu’elle
est historiquement peu valable et ne peut être artistiquement cantonnée à un style défini, ou à un centre
de production avéré. La définition selon laquelle la peinture « arabe » est celle qui orne les ouvrages
en langue arabe est bien sûr caduque, mais force est de reconnaître qu’il est difficile, à ce stade de nos
connaissances, de proposer une terminologie plus précise pour classer ces œuvres15. Depuis les toutes
premières études consacrées au sujet, les historiens de l’art islamique semblent avoir volontairement
11 La copie ésopique a sans doute été exécutée en Italie du Sud, elle est aujourd’hui conservée à New York au sein de la
Pierpont Morgan Library (ms. 397) ; voir à ce sujet Myrtilla Avery, « Miniatures of the Fables of Bidpai and of the Life of
Aesop in The Pierpont Morgan Library », The Art Bulletin 23, no 2 (1941) : 103-116.
12 Le recensement effectué au cours du programme de recherche a permis d’identifier plusieurs exemplaires illustrés qui
n’étaient pas inclus dans le catalogue de Grube ; ainsi, huit manuscrits ont été redécouverts, soit un total de vingt-sept
manuscrits illustrés, trois folios illustrés et sept manuscrits destinés à être illustrés. Une grande partie des peintures figu-
rant dans ce livre sont publiées pour la première fois.
13 Munich, BSB, Cod.Ar.616 ; Oxford, OBL, Pococke 400 ; Cambridge University, TS-AR-051-060.
14 Richard Ettinghausen, La peinture arabe (Genève : Skira, 1962).
15 A. Contadini revient avec justesse sur ces questions inhérentes à la définition de la peinture arabe dans Anna Contadini,
éd., Arab Painting: Text and Image in Illustrated Arabic Manuscripts (Leiden: Brill, 2007). Au sein de ce volume collec-
tif, les contributions suivantes s’attachent tout particulièrement à déconstruire la taxinomie de la « peinture arabe » :
x Préface / Foreword
négligé les périodes tardives, les percevant comme des productions commerciales de moindre valeur,
tout en s’inscrivant là dans le sillage de R. Ettinghausen qui regrettait qu’après la période mamelouke,
la peinture arabe soit tombée « à des niveaux indignes »16. Il ne s’agit pas de vilipender l’une des plus
grandes figures de l’histoire de l’art islamique, mais il est temps de penser en d’autres termes et de définir
de nouvelles bases, une réflexion à laquelle le programme Kalīla wa-Dimna s’est en partie consacré. Les
articles rassemblés dans le présent ouvrage explorent des voies nouvelles en se concentrant sur l’analyse
d’œuvres inédites, ou méconnues et en proposant d’approfondir des pistes de réflexion rarement visitées.
Filiation, reproduction, mutation sont les maîtres mots de cette recherche collective et nous avons voulu
qu’ils demeurent les fils conducteurs de la publication qui en résulte. Le livre de Kalīla wa-Dimna en
effet, par son histoire hors du commun, ses voyages dans le temps et dans l’espace, se prête parfaitement
à l’exercice. Continuellement, son étude pose la question de la transmission, littéraire et artistique, qui
elle-même nous ramène à l’humain : contextes de production, goût du lectorat, choix des artistes, sou-
haits des mécènes, moyens matériels et procédés artistiques.
Toutes ces problématiques sont mises en œuvre dans cet ouvrage qui réunit des spécialistes parmi les
meilleurs du domaine. Sa première partie renouvelle les études sur le texte de Kalīla wa-Dimna en remet-
tant en question les paradigmes dominants concernant l’histoire des versions arabes et persanes du
recueil. À l’initiative d’un projet d’édition digitale du texte arabe de Kalīla wa-Dimna à la Freie Universität
de Berlin, Beatrice Gruendler montre l’existence de multiples variantes textuelles à partir de l’étude de
la fable « Le chat et le rat » dans sept manuscrits s’échelonnant entre 1221 et 1699. Récusant l’idée même
d’une reconstruction du texte original, elle s’interroge sur les ressemblances et les divergences de chaque
version et met en avant le rôle fondamental des copistes dans des processus de réécriture destinés à
des publics spécifiques. Cette approche de l’œuvre par le texte se poursuit grâce à l’article de Christine
van Ruymbeke qui se penche sur les deux adaptations persanes les plus célèbres. S’appuyant sur un
minutieux travail de comparaison des versions d’Ibn al-Muqaffaʿ, de Naṣr Allāh Munshī et de Kāshifi
de la fable « L’âne, le lion et le chacal », Christine van Ruymbeke conteste l’idée communément admise
selon laquelle le substrat indien constitue la trame originelle de Kalīla wa-Dimna et met en lumière
l’approfondissement des thèmes apporté au texte arabe dans les adaptations persanes. Enfin, l’article de
Jean-Charles Coulon aborde un aspect inédit lié à l’utilisation des manuscrits à peintures. Son intérêt se
porte sur le paratexte, à savoir les marques de géomancie présentes dans les marges et sur les peintures
d’un manuscrit mamelouk de Kalīla wa-Dimna conservé à la BnF sous la cote Arabe 3467 et il émet plu-
sieurs hypothèses sur le sens à leur donner en mettant en lumière un usage encore inconnu du texte.
La deuxième partie du volume est consacrée aux questions de transmission iconographique, pierre
angulaire du corpus étudié dans le programme de recherche. Plusieurs copies tardives sont ainsi claire-
ment redevables à des exemplaires antérieurs. L’article d’Annie Vernay-Nouri montre comment le plus
ancien manuscrit illustré connu de Kalīla wa-Dimna (Paris, BnF, Arabe 3465) constitue un jalon majeur
dans la chaîne de transmission. Deux volumes du xviie siècle (Paris, BnF, Arabe 3470 et Oxford, OBL,
E.D. Clarke.Or.09) y puisent directement selon des modalités infiniment complexes ; d’autres copies, plus
lointaines, en sont également tributaires. L’article suivant que l’on doit à Anna Contadini resitue les pre-
miers manuscrits illustrés de Kalīla wa-Dimna dans le contexte plus large de la peinture arabe des xiiie
et xive siècles. Elle se penche également sur la question de l’intertextualité et de l’existence d’un réper-
toire iconographique commun. Bernard O’Kane, quant à lui, exploite une copie datée de 1761 (Paris, BnF,
Contadini, « The Manuscript as a Whole », in Arab Painting, 3-16 ; Oleg Grabar, « What Does “Arab Painting” Mean ? », in
Arab Painting, 17-22.
16 Ettinghausen, La peinture arabe, 180.
Préface / Foreword xi
Arabe 3475) pour reconstituer les images manquantes du cycle iconographique d’un des plus anciens
manuscrits illustrés, datant de la période ilkhanide (Rabat, BRR, ms. 3655) qui jusqu’ici n’a fait l’objet que
de très peu d’études.
La troisième partie de l’ouvrage s’attache à analyser les mécanismes de construction et de production
des images. L’article de Mounia Chekhab-Abudaya montre la relation étroite entre anecdotes et peintures
au travers de l’étude d’un manuscrit du Sulwān al-mutā (Doha, MIA, ms. 27.1997). Apparenté au texte de
Kalīla wa-Dimna, ce manuscrit comporte vingt-six peintures dont la qualité en fait l’un des plus beaux
spécimens illustrés de l’époque mamelouke. L’article d’Éloïse Brac de la Perrière démontre comment
plusieurs manuscrits ont été utilisés de manière successive, sur un temps long et dans divers contextes,
pour donner naissance à une œuvre hétéroclite et déroutante (New York, MMA, 1981.373.51), véritable
patchwork d’images. Elle souligne le rôle important joué par l’utilisation des calques et des poncifs dans
l’élaboration des peintures. Enfin, les analyses physico-chimiques menées par Nathalie Buisson et Annie
Vernay-Nouri viennent clore cette partie et éclairent à un autre niveau un corpus de sept manuscrits de
la BnF réalisés entre le xiiie et le xviiie siècle (dont six Kalīla wa-Dimna). Elles ont permis de différencier
les palettes de couleurs selon les époques et jettent une lumière nouvelle sur l’attribution de certains
groupes de manuscrits à des ateliers communs.
La quatrième partie explore les mutations profondes qui transforment texte et images lorsque l’œuvre
s’exporte au-delà de ses frontières spatiales ou temporelles. Dans le premier des articles, Yves Porter étu-
die un aspect original du manuscrit de Rabat (Rabat, BRR, ms. 3655), copié au début de la domination il
khanide sur l’Iraq. Se penchant sur l’influence mongole, il s’interroge sur la signification des « marqueurs
ethnoculturels » dans un contexte de profonds bouleversements politiques et culturels. Le chapitre qui
suit, fruit d’un travail collectif ici synthétisé et analysé par Aïda El Khiari, aborde les problématiques
complexes liées à l’« identité » et au concept de « peinture arabe » à travers l’étude d‘un manuscrit iné-
dit. Conservée dans une collection particulière, cette adaptation versifiée en arabe de Kalīla wa-Dimna
dont ce manuscrit est le seul exemplaire illustré connu, constitue un témoin exceptionnel de son temps.
Il apporte des éléments nouveaux à notre connaissance des activités picturales de la première moitié
du xvie siècle, période encore largement ignorée par les historiens de l’art islamique. Francis Richard,
quant à lui, étudie une version turque de Kalīla wa-Dimna différente du Hümāyūnnāme et dont seules
quelques pages sont préservées à la BnF et à la BULAC. Leurs illustrations, de très belle facture, se rat-
tachent à la production impériale ottomane de la seconde moitié du xvie siècle. Enfin, Mika Natif, dans
le dernier article du volume, s’intéresse à la manière dont le texte de Kalīla wa-Dimna ainsi que son ico-
nographie sont réinvestis par l’Inde moghole dans Iyār-i Danīsh, une nouvelle adaptation commanditée
en 996H/1587-8 par l’empereur Akbar auprès de son ministre Abū al-Faẓl ibn Mubārak.
Offertes au sein d’un même ouvrage, ces contributions visent à saisir dans sa globalité une œuvre poly-
sémique qui n’a eu de cesse, au gré des siècles et des langues, de se transformer. Embrassant d’un même
regard ses différents aspects, les articles conjuguent l’étude du texte, dans son histoire et sa narratolo-
gie, aux approches stylistiques et iconographiques des peintures. Ils bénéficient tout autant des apports
complémentaires de l’analyse matérielle (paléographie, codicologie, études physico-chimiques …) et
démontrent ainsi la fécondité de l’approche multidisciplinaire.
Remerciements / Acknowledgements
Le programme de recherche consacré aux manuscrits à peintures de Kalīla wa-Dimna, dont cet ouvrage
est le fruit, a bénéficié depuis ses débuts de l’aide et du concours de nombreux acteurs institutionnels
et académiques, de personnalités scientifiques et d’étudiants. Ce programme a vu le jour à l’initiative de
la Bibliothèque nationale de France qui l’a hébergé et soutenu durant cinq années consécutives (2012-
2017) ; Sorbonne Université a apporté son appui scientifique et pédagogique. À la Bibliothèque nationale
de France, nous voulons remercier tout particulièrement Isabelle Le Masne de Chermont dont le sou-
tien a été constant. L’équipe Islam Médiéval, au sein de l’UMR 8167 – Orient et Méditerranée, et l’UMR
7192 – Proche-Orient, Caucase, Langues, Archéologie, Cultures, nous ont accompagnées depuis la genèse
du projet jusqu’à la publication de cet ouvrage. Celui-ci n’aurait pu voir le jour sans le concours de la
Bibliothèque nationale de France, de Sorbonne Université et de la Fondation Bruschettini. Nous tenons
à leur exprimer ici toute notre gratitude.
Nous exprimons notre plus vive reconnaissance au collectionneur privé qui nous a permis d’étudier
un manuscrit jusque là inédit et qui a souhaité garder l’anonymat. Nous adressons également nos plus
sincères remerciements au personnel des différentes institutions qui nous ont apporté leur pleine assis-
tance : en particulier Helga Rebhan (Bayerische Staatsbibliothek) ; Mohammed Binebine (Bibliothèque
Royale du Maroc) ; Alasdair Watson (Bodleian Library) ; Ursula Sims-Williams et Daniel Lowe (British
Library) ; Marie-Lise Tsagouria et Francis Richard (BULAC) ; Yasmin Faghihi (Cambridge University
Library) ; Elaine Wright et Moya Carey (Chester Beatty Library) ; Romain Pingannaud (anciennement
chez Christie’s) ; Yahya b. M. Junaid (King Faysal Center for Research and Islamic Studies) ; Elizabeth
Gow (John Rylands Library) ; Anaïs Salamon (McGill University Library) ; Sheila Canby et Deniz Beyazıt
(Metropolitan Museum of New-York) ; Christoph Rauch (Staatsbibliothek zu Berlin).
À toutes celles et ceux qui ont accepté de contribuer à cet ouvrage, Nathalie Buisson, Mounia
Chekhab-Abudaya, Franz Chaigne, Anna Contadini, Jean-Charles Coulon, Françoise Cuisance, Rajana
Fatima Amalarajah, Béatrice Gruendler, Mika Natif, Bernard O’Kane, Hoa Perriguey, Yves Porter, Francis
Richard, Valérie Saurel et Christine van Ruymbeke nous voulons dire encore notre profonde reconnais-
sance pour ce travail d’équipe, extrêmement stimulant. Il nous faut aussi rappeler toute l’implication
des étudiants et anciens étudiants de Sorbonne Université dans la phase préparatoire du programme
de recherche : Nala Aloudat, Suzanne Babey, Pauline de Keukelaere, Ghazaleh Esmaïlpour Qouchani,
Giovanni Guzzeta, Isabelle Imbert, Adib Kassas, Sarah Lakhal-Kermani et Loïc Lesvignes.
Nous remercions Thomas Lorain pour le travail de coordination qu’il a effectué durant la première
année du programme de recherche, et pour l’organisation d’un workshop qui s’est tenu à Paris en
novembre 2013. Ce livre est aussi le fruit des très riches discussions que nous avons eues durant cette ren-
contre scientifique et nous exprimons notre gratitude aux institutions qui en ont permis la réalisation :
le FIR de Sorbonne Université, la chaire de dialogue des cultures (Université Paris 1 Panthéon-Sorbonne),
la Fondation Bruschettini et l’École doctorale 124 (Sorbonne Université). En 2014, Aïda El Khiari a repris
la coordination du projet et sa profonde implication a beaucoup contribué aux résultats présentés dans
cet ouvrage.
Plusieurs de nos collègues ont soutenu ce projet en nous permettant de bénéficier de leur expertise
et de leurs conseils, nous tenons à les remercier très sincèrement : Balsam Abdul-Rahman Saleh, Ekin
Akalın, Sandra Aube, Francesca Bellino, Samer Ben Brahim, Nourane Ben Azzouna, Mathieu Bonicel,
Remerciements / Acknowledgements xiii
Eric Delpont, Sylvie Denoix, François Déroche, Maxime Durocher, Jean-Louis Estève, Marcus Fraser,
Marie Geneviève Guesdon, Laurent Héricher, Karim Ifrak, Adeline Laclau, Rachel Milstein, Jean-Michel
Mouton, Antoine Saliba, Jacques Sicre, Aurélia Stréri, Jean-Pierre van Staëvel, Ségolène Walle, Elaine
Wright, Sara Yontan.
Enfin, tous nos remerciements vont à Teddi Dolls chez Brill qui a suivi la publication jusqu’à son abou-
tissement final, à Carina van den Hoven qui en a assuré la production, à tous ceux qui, à la BnF ont œuvré
pour que cette coédition voit le jour, particulièrement David Guillet et Marie-Caroline Dufayet, et aux
deux relecteurs de l’anglais et du français, Edward Street et Cécile Martin.
Note au lecteur / Note to the Reader
Translittération
Le système de translittération utilisé dans cet ouvrage est celui établi par la revue International Journal of Middle
Eastern Studies.
Abréviations
Figures
1.1 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 5881, fol. 2v. Incipit with short table of contents 35
1.2 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 5881, fol. 79v. Beginning of the chapter of “The Cat and
the Rat” 36
1.3 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3466, pp. 26-27. Beginning of the table of contents
including the synopsis of the sixth chapter, “The Cat and the Rat,” within the preface of ʿAlī b. Shāh 37
1.4 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3466, pp. 273-74. Beginning of the chapter of
“The Cat and the Rat” with unfilled blank for an illustration 38
1.5 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, London, British Library, Or. 4044, fol. 97v. Beginning of the chapter
of “The Cat and the Rat” (By permission of the British Library) 39
3.1 Un texte de géomancie attribué à Abū Muḥammad ʿImrān al-Munajjim al-Jabsūyī al-Faylasūf, Paris BULAC,
ms. 579, f. 47v-48r (Photo : BULAC) 62
3.2 Le roi et le philosophe, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 3v 73
3.3 Le roi et le philosophe, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 110r 75
3.4 Le lion et trois chacals, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 3r 77
3.5 Le lion et sa mère, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 46r 78
3.6 Kalīla et Dimna discutent, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 21r 79
3.7 Kalīla et Dimna discutent, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 41v 80
3.8 Le lion tue le taureau, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 40v 81
3.9 L’associé malhonnête et le voleur, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 26v 82
3.10 Le pauvre surprend le voleur, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 28v 83
3.11 Le moine et les deux bouquetins tuant un renard dans leur combat, Ibn al-Muqaffaʿ, Kalīla wa-Dimna,
Paris, BnF, Arabe 3467, f. 15v 85
3.12 L’empoisonneuse à la sarbacane, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 16r 85
3.13 Le lion, prêt à se jeter sur son propre reflet, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467,
f. 30r 86
3.14 Un singe se saisit de l’oiseau pour le jeter à terre, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467,
f. 42v 87
3.15 Le rat délivre les colombes du filet, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 63r 88
4.1 The two jackals: Kalīla and Dimna. Ibn al-Muqaffaʿ, Kalīla wa-Dimna; a) Egypt or Syria, ca. 1220. BnF
Arabe 3465, fol. 48r; b) Egypt or Syria, ca. 1300-10. BSB Cod. Arab. 616, fol. 44v (Courtesy of the Bayerische
Staatsbibliothek) 98
4.2 Burzoy and physicians setting off on their journey. Ibn al-Muqaffaʿ, Kalīla wa-Dimna; a) Egypt or Syria,
ca. 1300-10. BSB, Cod. Arab. 616, fol. 21r (Courtesy of the Bayerische Staatsbibliothek); b) Egypt or Syria,
dated 755H/1354. Bodleian Pococke 400, fol. 12v (Courtesy of the Bodleian Library) 99
4.3 “The Heron and the Crab”; a) Giuliano Da Sangallo, step-end from the Gondi Palace in Florence, ca. 1490-
1501. London, Victoria & Albert Museum, inv. no. 35-1891 (Courtesy of the Victoria & Albert Museum); b) Ibn
al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca. 1300-10. BSB Cod. Arab. 616, fol. 52r (detail) (Courtesy of the
Bayerische Staatsbibliothek) 102
xvi Figures et tableaux / Figures and Tables
4.4 “Lion and Hare at the Well”; a) Sketch of fresco, first register, Room 1/XXI, Panjikent, near Samarkand,
early seventh century (After Raby, “The Earliest Illustrations,” fig. 5); b) Ibn al- Muqaffaʿ, Kalīla wa-Dimna,
Egypt or Syria, ca. 1300-10. BSB Cod. Arab. 616, fol. 53r (detail) (Courtesy of the Bayerische Staatsbibliothek);
c) Egypt or Syria, ca. 1220. BnF Arabe 3465, fol. 58v (detail) 103
4.5 “The Monkey on the Carpenter’s Work Horse”; a) Relief from Mallikarjuna Temple, Pattadakal, Karnataka,
mid-eighth century (After Cohen and Shehada, “From the Panchatantra to La Fontaine,” detail of fig. 32);
b) Moulded terracotta plaque from the temple of Paharpur, Bengal, probably tenth century ce (After Raby,
“The Earliest Illustrations,” fig. 16); c) Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca. 1300-10. BSB Cod.
Arab. 616, fol. 43r (detail) (Courtesy of the Bayerische Staatsbibliothek) 104
4.6 a) Ruler-prince, enthroned with two attendants. Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca.
1220. BnF Arabe 3465, fol. 34r; b) Ruler-prince, enthroned with two attendants. Ibn Bakhtīshūʿ, Kitāb Naʿt
al-Ḥayawān (“Book on the Characteristics of Animals”), probably Baghdad, datable to ca. 1224-25. London,
British Library, Or. 2784, fol. 2v (reconstructed fol. 3v) (By kind permission of the British Library Board);
c) Dioscorides as scholar-prince (with Aristotle and Luqmān), frontispiece. Dioscorides, Hayūlā ʿIlāj al-Ṭibb,
Syria?, dated 642H / 1245. Bologna, Biblioteca Universitaria, Cod. arab. 2954, fol. 141r (Courtesy of the
Biblioteca Universitaria, Bologna); d) Joseph of Arimathea claiming the body of Christ. Gospel, Damietta,
Egypt, dated 1178-80. BnF Copte 13, fol. 131r 107
4.7 Pigeons trapped in the net by the fowler, and rat gnawing the net, while the crow looks down from
the tree. Ibn al-Muqaffaʿ, Kalīla wa-Dimna; a) Egypt or Syria, ca. 1220. BnF Arabe 3465, fol. 84v; b) Egypt
or Syria, mid-fourteenth century. BnF Arabe 3467, fol. 63r; c) Egypt or Syria, dated 755H/1354. Bodleian
Pococke 400, fol. 82v (Courtesy of the Bodleian Library) 109
4.8 a) Pigeons trapped in the net, b) they fly off while still in the net, c) and rat gnawing through the net. Ibn
al-Muqaffaʿ, Kalīla wa-Dimna; Egypt or Syria, ca. 1300-10. BSB Cod. arab. 616, fols. 76r, 76v and 77v respec-
tively (Courtesy of the Bayerische Staatsbibliothek) 110
4.9 Pigeons fly off while still in the net. Al-Kāshifī, Anvār-i Suhaylī (“Lights of Canopus”), South Asia, Mughal,
dated 978H/1570. London, SOAS University of London, SOAS Library Special Collections, MS 10102, fol. 123r
(Courtesy of the School of Oriental and African Studies) 111
4.10 Geese carrying the tortoise in flight. a) Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca. 1220. BnF
Arabe 3465, fol. 67r; b) Egypt or Syria, ca. 1300-10. BSB Cod. Arab. 616, fol. 60v (Courtesy of the Bayerische
Staatsbibliothek); c) Egypt or Syria, dated 755H/1354. Bodleian Pococke 400, fol. 61v (Courtesy of the
Bodleian Library) 112
4.11 a) The tortoise falls to the ground. Naṣrullāh, Kalīla wa-Dimna, Herat, ca. 1510. Rampur, Raza Library, MS
2982; b) The tortoise, having fallen to the ground, is beaten to death. Stone relief, Mathura, third century ce.
Mathura Museum, no. J 36 (After Grube, A Mirror for Princes, figs. 27 and 23 respectively) 113
4.12 Dimna in prison. Ibn al-Muqaffaʿ, Kalīla wa-Dimna; a) Dimna visited by Kalīla. Egypt or Syria, ca. 1300-10.
BSB Cod. Arab. 616, fol. 70v (Courtesy of the Bayerische Staatsbibliothek); b) Dimna visited by Kalīla. Egypt
or Syria, dated 755H/1354. Bodleian Pococke 400, fol. 73v (Courtesy of the Bodleian Library); c) shaʿhar/
Rūzabī carrying off a heavy bag given to it by Dimna. Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca.
1220. BnF Arabe 3465, fol. 80v 113
4.13 a) The hares. Ibn Bakhtīshūʿ, Kitāb Manāfīʿ al-Ḥayawān (“Book of the Usefulness of Animals”), probably
Damascus, dated 755H/1354. San Lorenzo del Escorial, Biblioteca Real, Ar. 898, fol. 31v (Copyright of the
Biblioteca Real, San Lorenzo del Escorial); b) Fairuz the Hare, addressing the hares. Ibn al-Muqaffaʿ,
Kalīla wa-Dimna, Egypt or Syria, dated 755H/1354. Bodleian Pococke 400, fol. 98r (Courtesy of the
Bodleian Library) 114
Figures et tableaux / Figures and Tables xvii
4.14 a) Two goats. Ibn Bakhtīshūʿ, Kitāb Naʿt al-Ḥayawān (“Book on the Characteristics of Animals”), probably
Baghdad, ca. 1224-25. London, British Library, Or. 2784, fol. 113r (reconstructed fol. 22r) (By kind permission
of the British Library Board); b) Two goats. Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca. 1220. BnF
Arabe 3465, fol. 110r 115
4.15 Similar-shaped plants in: a) Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca. 1220. BnF Arabe 3465,
fol. 107v; b) Gospel, Damietta, Egypt, dated 1178-80. BnF Copte 13, fol. 167v 116
4.16 Ribbed blue domes on arches in: a) Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca. 1220. BnF Arabe
3465, fol. 30v; b) Gospel, Damietta, Egypt, dated 1178-80. BnF Copte 13, fol. 138v 116
4.17 Similar-shaped conceptual architecture in: a) Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca. 1220. BnF
Arabe 3465, fol. 131v; b) Al-Ḥarīrī, Maqāmāt (“Assemblies”), Syria(?), dated 619H/1222. BnF Arabe 6094, fol.
180r 117
4.18 “The merchant and the musician” (right, fol. 24v) and “The dog and the bone” (left, fol. 25r). Ibn
al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, late fourteenth century (?). Cambridge, Corpus Christi Library,
MS 578 (Courtesy of the Corpus Christi College, Parker Library) 120
4.19 Qurʾan produced in the Great Umayyad Mosque in Damascus, ca. 731-741H/1330-40, London, Khalili
Collection, QUR 807, fols. 295v-296r 120
4.20 a) Ibn Ẓafar al-Siqillī, Sulwān al-Muṭāʿ fī ʿUdwān al-Atbāʿ (“The Consolation of the Prince with Hostile
Subjects”), probably Damascus, second quarter of the fourteenth century. Doha, Museum of Islamic
Art, MS.27.1999, fol. 1r (After Melikian-Chirvani, Sulwān al-Muṭāʿ); b) Al-Ḥarīrī, Maqāmāt (“Assemblies”),
probably Damascus, dated 734H/1334. Vienna, Nationalbibliothek, A.F. 9, fol. 8v (Copyright of the
Nationalbibliothek); c) Ibn Bakhtīshūʿ, Kitāb Manāfīʿ al-Ḥayawān (“Book of the Usefulness of Animals”),
probably Damascus, dated 755H/1354. San Lorenzo del Escorial, Biblioteca Real, Ar. 898, fol. 80r (Copyright
of the Biblioteca Real, San Lorenzo del Escorial); d) Al-Ḥarīrī, Maqāmāt (“Assemblies”), probably Damascus,
dated 738H/1337. Bodleian Marsh 458, fol. 45r (Courtesy of the Bodleian Library) 121
4.21 Glass vase, gilded and enameled. Probably Damascus, first half of the fourteenth century. Kuwait, the
Al-Ṣabāḥ Collection, Dār al-Āthār al-Islāmiyyah, LNS 69 G (After Atıl, Arte islamica, p. 213) 122
4.22 Ibn Ẓafar al-Siqillī, Sulwān al-Muṭāʿ fī ʿUdwān al-Atbāʿ (“The Consolation of the Prince with Hostile
Subjects”), probably Damascus, second quarter of the fourteenth century. Doha, Museum of Islamic Art,
MS.27.1999, fol. 44r (After Melikian-Chirvani, Sulwān al-Muṭāʿ) 123
5.1a Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3465, f. 114v 134
5.1b Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3465, f. 139v 134
5.2 Frontispice, Paris, BnF, Arabe 3465, f. 34 135
5.3 Kalīla et Dimna, Paris, BnF, Arabe 3465, f. 48 136
5.4 Le moine, le voleur, le renard et l’épouse du cordonnier, Paris, BnF, Arabe 3465, f. 55 137
5.5 Burzūya, Paris, BnF, Arabe 3465, détail f. 20v 138
5.6 Alexandre consultant les astrologues, Paris, BnF, Arabe 3465, f. 3 139
5.7 Histoire du roi Nuʿamān, Mille et une Nuits, Manchester, John Rylands Library, Arabic 646, f. 78v (Courtesy of
the University of Manchester) 140
5.8 Le saint homme et son hôte, Paris, BnF, Arabe 3465, f. 139v 141
5.9 Le saint homme et son hôte, Paris, BnF, Arabe 3470, f. 116v 142
5.10 Duel entre Poros et Alexandre, Oxford, Bodleian Library, E.D. Clarke Or. 9, f. 3v (Courtesy of the Bodleian
Library) 146
5.11 Duel entre Poros et Alexandre, Manchester, John Rylands Library, Arabic 487, f. 4v (Courtesy of the
University of Manchester) 147
xviii Figures et tableaux / Figures and Tables
5.12 Fanzah refusant de revenir vers le roi, Oxford, Bodleian Library, E.D. Clarke Or. 9, f. 103 (Courtesy of the
Bodleian Library) 148
5.13 Les chiens que la sobriété a empêchés de fuir leurs maîtres et d’être dévorés, Oxford, Bodleian Library,
E.D. Clarke Or. 9, f. 103v (Courtesy of the Bodleian Library) 149
5.14 Ibn Ẓafar, Sulwān al-muṭāʿ, Paris, BnF, Arabe 3511, détail f. 24v 150
5.15 Khusraw et Burzūya, Paris, BnF, Arabe 3465, f. 23 151
5.16 Khusraw et Burzūya, Paris, BnF, Arabe 3470, f. 13 152
5.17 Khusraw et Burzūya, Oxford, E.D. Clarke Or. 9, f. 18 (Courtesy of the Bodleian Library) 153
5.18a Détails de Paris, BnF, Arabe 3465, f. 20v 154
5.18b Détails de Paris, BnF, Arabe 3465, f. 22 155
5.18c Détails de Paris, BnF, Arabe 3470, f. 11v 155
5.19 Le moine, le voleur, le renard et la femme du cordonnier, Paris, BnF, A3470, f. 26 163
5.20 L’éléphant et l’alouette, Paris, BnF, Arabe 3465, f. 6 166
5.21 Bataille entre Alexandre et Poros, Oxford, E.D. Clarke Or. 9, f. 3 (Courtesy of the Bodleian Library) 167
5.22 Le lièvre et l’éléphant, Paris, BnF, Arabe 3470, f. 77v 168
6.1a Architectural scene. Paris, BnF, Arabe 3475, f. 10b 188
6.1b “Kisra Orders the Treasury to be Thrown Open to Burzuya”. Rabat, BRR, ms. 3655, f. 2b (Courtesy of the
Bibliothèque Royale) 188
6.2a “The Thief and the Moonbeam: The Thief Approaches the Bed”. Rabat, BRR, ms. 3655, f. 9a (Courtesy of
the Bibliothèque Royale) 189
6.2b “The Thief and the Moonbeam: The Thief is Beaten as He Falls Through the Roof”. Rabat, BRR, ms. 3655,
f. 9b (Courtesy of the Bibliothèque Royale) 189
6.3a “The Imprudent Lovers: The Wife and Her Lover Arguing”. Rabat, BRR, ms. 3655, f. 10a (Courtesy of the
Bibliothèque Royale) 190
6.3b “The Imprudent Lovers: The Husband Beats the Lover”. Rabat, BRR, ms. 3655, f. 10b (Courtesy of the
Bibliothèque Royale) 190
6.4 “The Condition in Which Man is Born”. Rabat, BRR, ms. 3655, f. 15b (Courtesy of the Bibliothèque
Royale) 191
6.5a “The Ascetic and the Robe: The Madam’s Stratagem Backfires”. Paris, BnF, Arabe 3475, f. 50b 192
6.5b “The Merchant’s Wife and the Slave Make Love”. Paris, BnF, Arabe 3475, f. 90a 192
6.6a “The Crow and the Snake: The Tortoise and the Crab”. Rabat, BRR, ms. 3655, f. 31a (Courtesy of the
Bibliothèque Royale) 193
6.6b “The Crow and the Snake: The Tortoise and the Crab”. Paris, BnF, Arabe 3475 f. 54a 193
6.7 “The Hare and the Lion: The Hare and the Lion at the Well”. Rabat, BRR, ms. 3655, f. 33b (Courtesy of the
Bibliothèque Royale) 194
6.8a “The Horse Whose Good Qualities Were Its Ruin”. Rabat, BRR, ms. 3655, f. 37a (Courtesy of the
Bibliothèque Royale) 194
6.8b “The Hunter Kills the Cubs”. Rabat, BRR, ms. 3655, f. 109a (Courtesy of the Bibliothèque Royale) 194
6.9a “The Bee Imprisoned in the Waterlily”. Rabat, BRR, ms. 3655, 38a (Courtesy of the Bibliothèque
Royale) 195
6.9b “The Bee Imprisoned in the Waterlily”. Paris, BnF, Arabe 3475, 67a 195
6.10 “The Waterfowl and the Spirit of the Sea”. Paris, BnF, Arabe 3475 f. 75a 195
6.11a “The Simpleton and the Rogue: The Two Men before the Judge”. Rabat, BRR, ms. 3655, 47b (Courtesy of
the Bibliothèque Royale) 196
Figures et tableaux / Figures and Tables xix
6.11b “The Simpleton and the Rogue: The Rogue’s Stratagem is Discovered”. Rabat, BRR, ms. 3655, 49a
(Courtesy of the Bibliothèque Royale) 196
6.12a “The Mutual Enmity Between the Lion and the Elephant”. Rabat, BRR, ms. 3655, f. 67b (Courtesy of the
Bibliothèque Royale) 197
6.12b “The Wild Elephant That is Captured with the Aid of a Tame One”. Paris, BnF, Arabe 3475, f. 160b 197
7.1 Prince à sa cour tenant une coupe contenant une perle, f.1r, MS.27.1999, MIA 209
7.2 Scène abstraite illustrant le moment où Mufawwaḍ brûle Ẓālim vivant, f.15v, MS.27.1999, MIA 210
7.3 Les rats en conversation puis rencontrent la gerboise, ff.61v et 62r, MS.27.1999, MIA 210
7.4 Ẓālim et Mufawwaḍ en conversation, f.12v, MS.27.1999, MIA 221
7.5 La gazelle fuit l’antilope, f.44r, MS.27.1999, MIA 222
7.6 La couronne de Perse est placée entre Khusraw l’Usurpateur et Bahrām Gūr, f.90r,
MS.27.1999, MIA 222
7.7 Le cheval en joug libéré par le sanglier, f.39v, MS.27.1999, MIA 223
7.8 L’ours feint d’être mort tandis que les singes s’approchent de lui, f.76r, MS.27.1999, MIA 224
7.9 Le voleur tombe à travers la trappe dans la chapelle, f.77r, MS.27.1999, MIA 224
7.10 ʿAyn Ahlīh attaché en compagnie de la vieille femme mutilée, f.37r, MS.27.1999, MIA 225
7.11 Le prince indien donne audience à l’ambassadeur de Khusraw, f.56r, MS.27.1999, MIA 226
7.12 Le roi de l’Allān, l’archer et le vizir, f.100v, MS.27.1999, MIA 226
7.13 Shāpūr à la cour de l’Empereur byzantin, f.31r, MS.27.1999, MIA 227
7.14 Nuʿmān b. Imruʾ al-Qays en compagnie de ʿAdī b. Zayd à la shaqīqa, f.97r, MS.27.1999, MIA 227
8.1 Détail du f. 1v, MMA, 1981.73 (Courtesy of the Metropolitan Museum of Art) 235
8.2 Page de titre, f. 1, MMA, 1981.73 (Courtesy of the Metropolitan Museum of Art) 236
8.3 Khusraw et Burzuya, f. 8v, MMA, 1981.73 (Courtesy of the Metropolitan Museum of Art) 238
8.4 Le lion et sa mère, f. 47v, MMA, 1981.73 (Courtesy of the Metropolitan Museum of Art) 239
8.5 Khusraw et Burzuya, f. 5v-6, MMA, 1981.73 (Courtesy of the Metropolitan Museum of Art) 240
8.6 Fable de Shedram, Iblad et Irakht, détail du f. 96, MMA, 1981.73 (Courtesy of the Metropolitan Museum of
Art) 240
8.7 Le cobra et les grenouilles, f. 73, MMA, 1981.73 (Courtesy of the Metropolitan Museum of Art) 241
8.8 Le moine, le voleur et l’épouse du cordonnier, f. 31v, MMA, 1981.73 (Courtesy of the Metropolitan Museum
of Art) 243
8.9 Le moine, le voleur et l’épouse du cordonnier, f. 15v, BnF, Arabe 3467 244
8.10 Le moine, le voleur et l’épouse du cordonnier, f. 48v, BSB, Cod. Arab. 616 (Courtesy of the Bayerische
Staatsbibliothek) 245
8.11 Le moine, le voleur et l’épouse du cordonnier, f. 46, Bodl. Libr., Pococke 400 (Courtesy of the Bodleian
Library) 246
8.12 Superposition de la figure du f. 86 du Pococke 400 sur le f. 58 du ms. du MMA 251
8.13 Superposition de la figure du f. 86 du Pococke 400 sur le f. 73 du ms. du MMA 252
8.14 Le jugement de Dimna, f. 75v, Bodl. Libr., Pococke 400 (Courtesy of the Bodleian Library) 254
8.15 Le jugement de Dimna, f. 51, MMA, 1981.73 (Courtesy of the Metropolitan Museum of Art) 255
8.16 Le lièvre et le lion, f. 35v, MMA, 1981.73 (Courtesy of the Metropolitan Museum of Art) 256
8.17 Shedram, Iblad et Irakht, f. 101, MMA, 1981.73 (Courtesy of the Metropolitan Museum of Art) 257
8.18 Détail du f. 25 observé à la feuille lumineuse, Bodl. Libr., Pococke 400 (Cliché d’Aïda El Khiari) 258
8.19 Le pauvre et le voleur, f. 15, MMA, 1981.73 (Courtesy of the Metropolitan Museum of Art) 259
8.20 Le chien et l’os, f. 38v, (BSB, Cod. arab. 616 (Courtesy of the Bayerische Staatsbibliothek) 262
8.21 Le chien et l’os, f. 20v, MMA, 1981.73 (Courtesy of the Metropolitan Museum of Art) 263
xx Figures et tableaux / Figures and Tables
9.1 Pelage rose du deuxième lapin, Paris, BnF, Arabe 3467, f.70 277
9.2a-b Seconde nuance de rose montrant une réponse orange clair en IRFC, Paris, BnF, Arabe 3465, P1,
f.20v 277
9.2c-d Mauve montrant une réponse rose orangée en IRFC, Paris, BnF, Arabe 3465, P1, f.86 278
9.3a Spectres de réflectance des bleus mesurés sur les feuillets 34, 57 et 84 de Paris, BnF, Arabe 3465, comparés
au spectre de réflectance de l’indigo de référence 279
9.3b Spectres de réflectance des bleus mesurés sur les feuillets 52, 52, 71 et 84 de Paris, BnF, Arabe 3465, com-
parés au spectre de réflectance du lapis-lazuli de référence 279
9.4a Photographie en lumière directe 282
9.4b Image composite infrarouge fausses couleurs de la robe de Paris, BnF, Arabe 3467, f.61 282
9.5a Jaune d’orpiment mélangé à d’autres pigments, P2, Paris, BnF, Arabe 3465, f.25v 283
9.5b Jaune d’orpiment pur, Paris, BnF, Arabe 3467, f.30 283
9.6 Orange du f.133v de Paris, BnF, Arabe 3472 283
9.7 Brun de la toque du gouverneur, Paris, BnF, Arabe 6094, f.133v 284
9.8a Or pur en paillettes dispersées dans un liant, Paris, BnF, Arabe 6094, f.49 285
9.8c Or pur en paillettes dispersées dans un liant, Paris, BnF, Arabe 3470, f.50 285
9.8b Or pur en paillettes dispersées dans un liant, Paris, BnF, Arabe 3467, f.98v 285
9.8d Particules d’or et de cuivre dispersées dans un liant, Paris, BnF, Arabe 3465, P2, f.25v 285
9.9 Tracé préparatoire en rouge, mis en évidence par la détérioration de la matière picturale,
Paris, BnF, Arabe 6094, f.13 286
9.10 Dernière arcature du dais tracé à la pointe sèche, Paris, BnF, Arabe 3465, f.20v 287
9.11a Nuances de bleu, Paris, BnF, Arabe 3465, f.20v 288
9.11b Nuances de vert, Paris, BnF, Arabe 3465, f.43v 288
9.12a Visage, Paris, BnF, Arabe 3465, f.71 289
9.12b Visage, Paris, BnF, Arabe 6094, f.13 289
9.13 Paris, BnF, Arabe 3467, f.30v 290
9.14 Paris, BnF, Arabe 3467, f.61 291
9.15 Diversité des couleurs de Paris, BnF, Arabe 3475, f.196v 292
9.16 Diversité des couleurs de Paris, BnF, Arabe 5881, f.14v 293
9.17a-b Photographie en lumière directe et image composite infrarouge fausses couleurs du f.25 de
Paris, BnF, Arabe 3465 295
9.17c-d Photographie en lumière directe et image composite infrarouge fausses couleurs du f. 143 de
Paris, BnF, Arabe 3465 295
9.18 Comparaison des lions réalisés par le peintre de Paris, BnF, Arabe 3470 avec celui du peintre de
Paris, BnF, Arabe 3465 295
10.1 Le roi et Bilad, Fable de Shedram, Iblad et Irakht, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR,
ms. 3655, f.104 (Photo : Bibliothèque Royale) 303
10.2 Khusraw ouvre son trésor pour Burzuya, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.2
(Photo : Bibliothèque Royale) 304
10.3 Le roi et Bilad, Fable de Shedram, Iblad et Irakht, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR,
ms. 3655, f.105 (Photo : Bibliothèque Royale) 305
10.4 Fable du marchand et du joueur de cymbales, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR,
ms. 3655, f.12 (Photo : Bibliothèque Royale) 306
10.5 Les hommes tuent le cobra, Fable du corbeau, du cobra et du chacal, Ibn al-Muqaffaʿ, Kalīla wa-Dimna,
Rabat, BRR, ms. 3655, f.32v (Photo : Bibliothèque Royale) 307
Figures et tableaux / Figures and Tables xxi
10.6 Fable de Shedram, Iblad et Irakht, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.99v (Photo :
Bibliothèque Royale) 308
10.7 Le roi repousse Irakht pour avoir choisi la couronne, Fable de Shedram, Iblad et Irakht, Ibn al-Muqaffaʿ,
Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.108v (Photo : Bibliothèque Royale) 308
10.8 Fable de Shedram, Iblad et Irakht, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.108 (Photo :
Bibliothèque Royale) 311
10.9 Fable du fourbe et du benêt, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.46v (Photo :
Bibliothèque Royale) 312
10.10 Fable du fourbe et du benêt, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.47 (Photo :
Bibliothèque Royale) 313
10.11 Fable du fourbe et du benêt, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.47v (Photo :
Bibliothèque Royale) 314
10.12 Fable de l’amant surpris par le mari, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.10v (Photo
: Bibliothèque Royale) 316
10.13 Fable de l’amant surpris par le mari, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.11 (Photo :
Bibliothèque Royale) 317
10.14 Fable du qadi de Merv, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.14 (Photo : Bibliothèque
Royale) 318
10.15 Fable du saint homme et son hôte, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.111v (Photo :
Bibliothèque Royale) 321
11.1 Plat supérieur des reliures des volumes 1 et 2 (Clichés de Françoise Cuisance) 329
11.2 Réfection des fonds de cahier à l’aide de papier japon (Cliché d’Aïda El Khiari) 330
11.3 Exemple d’une réclame grattée (Cliché d’Aïda El Khiari) 330
11.4 Colophons des trois textes (Cliché d’Aïda El Khiari) 332
11.5 Frontispice (vol. 1, f. 64) et finispice (vol. 2, f. 29) du Miftāḥ al-fiṭna fī naẓm Kalīla wa-Dimna, collection
particulière (Cliché d’Aïda El Khiari) 333
11.6 Frontispice d’al-Ṣādiḥ wa-l-bāghim, vol. 2, f. 1 (Cliché d’Aïda El Khiari) 335
11.7 Frontispice d’al-Laṭāʾim wa-l-ashnāf, vol. 1, f. 2 (Cliché d’Aïda El Khiari) 336
11.8 Finispice du manuscrit, vol. 1, f. 63 (Cliché d’Aïda El Khiari) 336
11.9 Fable du marchand et du joueur de harpe : vol. 1, f. 19 (Cliché d’Aïda El Khiari) 341
11.10 Fable du marchand et du joueur de harpe, f. 24, 1388 ?, ms. 578, Cambridge, Parker Library (Courtesy of
the Parker Library, Corpus Christi College) 341
11.11 Fable du lion, du loup, du corbeau, du chacal et du chameau, vol. 1, f. 26 (Cliché d’Aïda El Khiari) 342
11.12 Fable du lion, du loup, du corbeau, du chacal et du chameau, f. 49v, 1388 ?, ms. 578, Cambridge, Parker
Library (Courtesy of the Parker Library, Corpus Christi College) 342
11.13 Fable du voleur et du rayon de lune, vol. 1, f. 18v (Cliché d’Aïda El Khiari) 343
11.14 Shukrī Bey Biltisī, Selīmnāma, 1527, Yahuda Ms.Ar.1116, Jérusalem, National Library of Israel (Courtesy of
the National Library of Israel, Jerusalem) 345
11.15 Fable de l’ermite et du vizir, vol. 2, f. 26v (Cliché d’Aïda El Khiari) 346
11.16 Représentation d’une ville miniature, détail, vol. 1, f. 9 (Cliché d’Aïda El Khiari) 347
11.17 Fable du chameau et du chamelier, détail, vol. 2, f. 5v (Cliché d’Aïda El Khiari) 347
11.18 Fable du fils du roi et ses compagnons, vol. 1, f. 59v (Cliché d’Aïda El Khiari) 349
11.19 Fable de Shedram, Iblad et Irakht, vol. 1, f. 53v (Cliché d’Aïda El Khiari) 350
xxii Figures et tableaux / Figures and Tables
12.1 Fable du peintre et de la jeune fille, version turque ottomane de Kalīla wa-Dimna, Paris, BULAC, ms.
Turc 205, f.6v (Photo : BULAC) 360
12.2 Fable du renard et du tambour, version turque ottomane de Kalīla wa-Dimna, Paris, BULAC, ms. Turc 205,
f.7 (Photo : BULAC) 362
12.3 Fable du singe et du menuisier, version turque ottomane de Kalīla wa-Dimna, Paris, BULAC, ms. Turc 205,
f.8 (Photo : BULAC) 363
12.4 Kalila et Dimna discutent, version turque ottomane de Kalīla wa-Dimna, Paris, BULAC, ms. Turc 205, f.11v
(Photo : BULAC) 364
12.5 Fable du médecin ignorant et de la fille du roi, version turque ottomane de Kalīla wa-Dimna, Paris, BnF,
Supplément turc 1243, f.18 365
12.6 Fable des deux perroquets et l’épouse du souverain, version turque ottomane de Kalīla wa-Dimna, Paris,
BnF, Supplément turc 1243, f.22v 367
12.7 Fable de la colombe au collier, version turque ottomane de Kalīla wa-Dimna, Paris, BnF, Supplément turc
1243, f.26 368
12.8 Fable de la colombe au collier, version turque ottomane de Kalīla wa-Dimna, Paris, BnF, Supplément turc
1243, f.29v 369
12.9 Fable du chasseur, version turque ottomane de Kalīla wa-Dimna, Paris, BnF, Supplément turc 1243,
f.33 370
13.1 Burzoy presented before King Anushirvan, by Basawan, Iyar-i Danish, fol. 5, Lahore, 1596-7, Mughal India.
Varanasi, Bharat Kala Bhavan, Ms. 9069 375
13.2 Detail: Burzoy presented before King Anushirvan, by Basawan, Iyar-i Danish, fol. 5, Lahore, 1596-7,
Mughal India. Varanasi, Bharat Kala Bhavan, Ms. 9069 377
13.3 Detail: Burzoy presented before King Anushirvan, by Basawan, Iyar-i Danish, fol. 5, Lahore, 1596-7,
Mughal India. Varanasi, Bharat Kala Bhavan, Ms. 9069 379
13.4 King Anushirvan and Burzoy, fol. 6a, Anvar-i Suhayli, by Kashifi, Allahabad, 1604-10, Mughal India.
London, British Library, Add. MS 18579 (By kind permission of the British Library) 380
13.5 Akbar and a dervish, ca. 1580-90, inscribed to Abd al-Samad, Mughal India. Toronto, Aga Khan Museum
Collection, AKM14 384
13.6 Abu’l Fazl Presents Akbar with the second volume of the Akbar Nama, by Govardhan, fol. 176v, ca. 1603-5,
Mughal India (right hand side of a double-page illustration) Dublin, Chester Beatty Library, ms. 3 385
13.7 Borzoy before Anushirwan, Anvar-i Suhayli, by Kashifi, first half of the sixteenth century, Central Asia.
Tashkent, Uzbek Academy of Sciences, Catalogue of Oriental Manuscripts 9109 387
Tableaux / Tables
1.1 Comparative chart of text units in the chapter of “The Cat and the Rat” in seven manuscripts 10
1.2 Chart of shared text segments in the chapter of “The Cat and the Rat,” in manuscripts of
Kalīla wa-Dimna 13
1.3 Comparative chart of text units in Ibn al-Muqaffaʿ’s preface of “The Purpose of the Book,” in seven
manuscripts 14
1.4 Synoptic edition of unit 20, “Rat describes common problem” based on seven manuscripts of Kalīla
wa-Dimna 20
3.1 Exemple de tableau géomantique d’un tirage avec pour maisons mères (I à IV) les figures suivantes :
, , et 63
Figures et tableaux / Figures and Tables xxiii
Sauf mention contraire, les illustrations de ce volume sont issues des collections de la Bibliothèque natio-
nale de France et les clichés en ont été réalisés par le département Images et Prestations numériques.
Elles sont disponibles à la consultation sur Gallica : <http://gallica.bnf.fr>, sur la banque d’images :
<http://images.bnf.fr>, et à la vente auprès du département Images et Prestations numériques : <repro-
[email protected]>.
Notes sur les contributeurs / Notes on Contributors
Nathalie Buisson
est ingénieure physico-chimiste, anciennement en charge du laboratoire Richelieu à la Bibliothèque
nationale de France. Après avoir travaillé au C2RMF (Centre de Recherche et de Restauration des Musées
de France), elle a rejoint la Bibliothèque nationale de France où elle était notamment chargée de la
caractérisation des matériaux de peinture.
Frantz Chaigne
est professeur agrégé de chimie à l’École Boulle. Il est également docteur en histoire de l’art de l’université
de Paris-Sorbonne. Sa thèse porte sur l’enluminure dans l’empire il-khanide. Il est l’auteur de plusieurs
articles sur les arts du livre du monde indo-persan de l’époque médiévale et s’intéresse aussi à la céra-
mique de la même période.
Mounia Chekhab-Abudaya
est conservateure responsable de la collection de l’Occident musulman et des manuscrits au Musée d’art
islamique de Doha (Qatar). Titulaire d’un doctorat en histoire de l’art et archéologie du monde islamique
de l’Université Panthéon Sorbonne, ses recherches portent sur le Maghreb, en particulier le Sahara, ainsi
que sur les manuscrits du monde islamique. En 2017-2018, elle a entrepris un projet de recherche à l’uni-
versité de Harvard en tant que chercheuse associée, dans le cadre de l’Aga Khan Program for Islamic
Architecture. Elle est entre autres l’auteure de Qajar Women, Images of Women in 19th-century Iran (avec
Nur Sobers-Khan, 2016) et Hajj, The Journey Through Art (avec Cécile Bresc, 2013).
Anna Contadini
est professeure en histoire de l’art islamique à la School of Oriental and African Studies de Londres. Elle
a précédemment exercé en tant que chercheuse associée de la Baring Foundation au Victoria and Albert
Museum (Londres) et conservateure à la Chester Beatty Library (Dublin) de 1994 à 1997. Elle est spécia-
liste de la culture matérielle avec un intérêt particulier pour les manuscrits illustrés arabes et persans
ainsi que pour l’étude des liens culturels et artistiques entre le Moyen-Orient et l’Europe. Elle a publié,
entre autres, A World of Beasts: A Thirteenth-Century Illustrated Arabic Book on Animals et a dirigé la
publication de Arab Painting: Text and Image in Illustrated Arabic Manuscripts, parus respectivement en
2012 et 2007 chez Brill ; son dernier ouvrage porte sur le Griffon de Pise et le Lion Mari-Cha. Elle mène
actuellement des recherches sur les manuscrits médiévaux à peintures produits en al-Andalus.
Notes sur les contributeurs / Notes on Contributors xxv
Jean-Charles Coulon
est docteur de l’Université de Paris IV Sorbonne en histoire médiévale et en études arabes et chargé de
recherche à la section arabe de l’Institut de Recherche et d’Histoire des Textes (IRHT-CNRS). Il est égale-
ment directeur adjoint et secrétaire de rédaction de la revue Arabica. Ses recherches portent sur l’histoire
de la magie et des sciences occultes dans le monde musulman médiéval. Il a publié La Magie en terre
d’islam au Moyen Âge (Paris, CTHS, 2017).
Françoise Cuisance
est restauratrice au département des Manuscrits de la Bibliothèque nationale de France.
Aïda El Khiari
est doctorante contractuelle à Sorbonne Université depuis septembre 2017. Elle achève une thèse de doc-
torat portant sur un corpus largement inédit de manuscrits arabes à peintures produits en Égypte et en
Syrie entre le xvie siècle et le xviiie siècle. Chargée de recherches à la Bibliothèque nationale de France,
elle a coordonné de 2014 à 2018 le programme sur les fables de Kalīla wa-Dimna.
Beatrice Gruendler
(Ph.D. Harvard University, 1995) est spécialiste de paléographie et de littérature arabe médiévale, et pro-
fesseure au Département d’Études Arabes à la Freie Universität de Berlin. Lauréate du Leibniz Prize de
la German Research Foundation en 2017, elle a également reçu le prestigieux Advanced Grant de l’ERC
pour mener un projet d’édition digitale du texte de Kalīla wa-Dimna, « AnonymClassic ». Elle enseignait
auparavant à l’université de Yale où elle est demeurée de 2002 à 2014. Ses publications incluent The
Development of the Arabic Scripts (Harvard Museum, 1997), Medieval Arabic Praise Poetry (Routledge,
2003), The Life and Times of Abū Tammām (NYU Press, 2015). Son nouvel ouvrage, Arabic Book Revolution,
est paru en 2020 à Harvard University Press.
Mika Natif
(Ph.D., New York University – Institute of Fine Arts, 2006) est historienne de l’art, spécialiste de
l’art pré-moderne et de l’Inde moghole en particulier, et Associate Professor à l’université Georges
Washington. Elle a auparavant exercé au Metropolitan Museum of Art et au Harvard Art Museum en
tant que conservateure adjointe. Elle a dirigé la publication de Eros and Sexuality in Islamic Art (Ashgate,
2013) et a récemment publié un ouvrage traitant des liens entre la peinture moghole et les arts occiden-
taux, Mughal Occidentalism: Artistic Encounters between Europe and Asia at the Courts of India, 1580-1630
(Brill, 2018).
Bernard O’Kane
est professeur d’art et d’archéologie islamiques à l’Université américaine du Caire où il exerce depuis
1980. Il était auparavant directeur adjoint du British Institute of Persian Studies à Téhéran. Il est l’auteur
de très nombreux ouvrages, le plus récent étant The Mosques of Egypt, Le Caire : American University in
Cairo Press, 2016.
Hoa Perriguey
est conservatrice-restauratrice spécialisée en art islamique. Elle travaille pour des institutions publiques
françaises. Elle est titulaire d’un master en conservation-restauration des biens culturels de l’université
Sorbonne-Nouvelle, et d’un master en histoire de l’art islamique de l’université Paris-Sorbonne.
xxvi Notes sur les contributeurs / Notes on Contributors
Yves Porter
est professeur d’art islamique au département d’Histoire de l’Art et d’Archéologie d’Aix Marseille
Université (AMU) et membre du Laboratoire d’Archéologie Méditerranéenne Médiévale et Moderne
(LA3M). Spécialiste des arts et techniques dans les domaines de l’Iran, de l’Inde musulmane et de l’Asie
centrale, il a publié de nombreux articles dans des revues scientifiques comme dans des publications à
grande diffusion, ainsi que des ouvrages, dont Peinture et arts du livre, Téhéran, 1992 ; L’Inde des sultans,
Paris, 2009 ; Le prince, l’artiste et l’alchimiste, Paris, 2011. Il a été nommé en 2018 à l’Institut Universitaire
de France où il mène un projet sur les céramiques lustrées de Kashan.
Francis Richard
est un éminent spécialiste des manuscrits persans. Il a été conservateur général au département des
Manuscrits de la Bibliothèque nationale de France, de 1974 à 2003. Il a ensuite dirigé le département des
Arts de l’Islam au musée du Louvre avant de prendre la direction scientifique de la Bibliothèque univer-
sitaire des langues et civilisations. Il a publié de nombreux ouvrages sur les manuscrits persans, dont
Le siècle d’Ispahan (Gallimard, 2007), Le Livre persan (BnF, 2003) ainsi que le Catalogue des manuscrits
persans de la BnF (1989-2013).
Annie Vernay-Nouri
est spécialiste des manuscrits arabes, et elle a été conservateure en chef au service Orient du départe-
ment des Manuscrits à la Bibliothèque nationale de France. Commissaire des expositions « Enluminures
d’islam » (BnF, 2011) et « Livres d’Arménie » (BnF, 2007) et auteure de leurs catalogues, elle a partagé le
commissariat des expositions « Paroles de bêtes (à l’usage des princes) : les fables de Kalila et Dimna »
(Institut du monde arabe, 2015), « Livres de Parole. Torah, Bible, Coran » (BnF, 2005), « Chevaux et cava-
liers arabes » (Institut du monde arabe, 2002), et « L’art du livre arabe » (BnF, 2001).
Elle a codirigé avec Éloïse Brac de la Perrière le programme de recherche sur les manuscrits à pein-
tures de Kalīla wa-Dimna.
partie 1
Textes et paratextes / Texts and Paratexts
∵
1
A Rat and Its Redactors: Silent Co-Authorship in Kalīla wa-Dimna
Beatrice Gruendler
Kalīla wa-Dimna, one of the most popular pre- Above all, it is a cosmopolitan work. From its
modern books, is a pioneering work of Arabic lit- Sanskrit origins, it traveled to Europe and Asia,
erature in several ways:1 first, it has two layers of via an intermediate Arabic phase, and became the
meaning, one overt and one hidden, which has most widely spread European medieval text. The
to be decoded by the reader; second, it displays a archaeologist James Henry Breasted, founder of the
sophisticated framing structure with replaceable Chicago Oriental Institute, once described it as “an
modular elements, making it easily adaptable to ancient text which, next to the Bible, has become
new cultural contexts; third, it is one of the ear- the most widely distributed and translated book in
liest Arabic books and reflects on how books are the entire history of literature,” and he made it one
to be read and how their content is to be applied of the pilot projects of his Institute, created in 1919
to real life; finally, it is one of the most frequently with the support of John D. Rockefeller as “a labo-
illustrated books of Arabic literature and discusses ratory for the study of the rise and development
the function of illustration in one of its prefaces.2 of civilization.”3 But this remained a prelude to an
unfinished edition.
1 Author’s note: this publication has been made possible via The Arabic version of Kalīla wa-Dimna is the
the Kalīla and Dimna – AnonymClassic research project direct or indirect source of all later translations,
at Freie Universität Berlin, funded by an Advanced Grant the point of departure for its global dissemina-
of the European Research Council within the European
Union’s Horizon 2020 research and innovation program
tion, and all research on its subsequent textual
under grant agreement No. 742635 (www.geschkult.fu- history must engage with it. However, the quest to
berlin.de/en/e/kalila-wa-dimna/index.html). I thank the untangle the torturous textual history of this ver-
members of the team for their comments on an earlier sion is as complex as the book itself. The extant
draft of this paper. Preparatory research was supported
witnesses of the Arabic version or, more correctly,
by an E-Learning/E-Research grant of the Center for
Digital Systems (CeDiS) of Freie Universität Berlin. Both versions, vary to such a degree that any expecta-
have used the LERA digital edition tool, developed in tion of reconstructing an “original” is unrealistic,
the SaDA-Project at Martin Luther University Halle- and indeed no critical edition has been produced
Wittenberg (sada.uzi.uni-halle.de), with support from the to this day. Inversely, the work’s textual history
German Ministry for Education and Research.
2 See Beatrice Gruendler, “Les versions de Kalīla wa-
offers a prime example of mutable and fluctuating
Dimna : une transmission et une circulation mouvantes,”
in Énoncés sapientiels et littérature exemplaire : une inter- Arabic translator-redactor, see István T. Kristó Nagy, La
textualité complexe, ed. M. Ortola and Marie-Christine pensée d’Ibn al-Muqaffaʿ : un “agent double” dans le monde
Bornes-Varol (Nancy: Presses Universitaires de Nancy – persan et arabe (Paris: Éditions de Paris, 2013) and Michael
Éditions Universitaires de Lorraine, 2013), 385–416. For Cooperson, “Ibn al-Muqaffaʿ,” in Arabic Literary Culture,
introductions to the work, see François de Blois, Burzōy’s Dictionary of Literary Biography, vol. 311, ed. Michael
Voyage to India and the Origin of the Book of Kalīlah Cooperson and Shawkat M. Toorawa (London: Thomson
Wa Dimnah (London: The Royal Asiatic Society, 1990); Gale, 2005), 150–63.
Ulrich Marzolph, and Heinz and Sophia Grotzfeld, “Kalīla 3 James H. Breasted, “The Oriental Institute in Chicago – a
und Dimna,” Enzyklopädie des Märchens, vol. 7 (1993): Beginning and a Program,” American Journal of Semitic
888–95; and Carl Brockelmann, “Kalīla wa-Dimna,” in Languages and Literatures 38, no. 4 (1922): 233–328, esp.
Encyclopaedia of Islam, second edition, ed. P. Bearman 314–19 (“The Tales of ‘Kalīla wa Dimna’ and the Ancestry
et al. (Leiden: Brill, 1960–2007), vol. 4, 503–6. On the of Animal Fables”).
written transmission, which is remarkable for a Arabic and Middle Arabic; and between rational-
book that once belonged to the adab corpus of ity and triviality. Regarding its authorship, there
classical Arabic literature and was counted as a were no true authors to begin with: the book
model of eloquent prose. As the centuries went began as a translation from Middle Persian, in
by, the book’s classical Arabic register became which it was already a selection, redaction, and
more and more hybridized in some manuscripts translation of anonymous and mutable Indian
(although the precise degree of non-classical fea- works (Pañcatantra and Mahābhārata). This
tures varies in each), and its content was trans- made it even easier to pull Ibn al-Muqaffaʿ off
formed in multiple ways. The purpose of this essay his authorial pedestal, and in fact most Arabic
is to trace these transformations. sources refer to Kalīla wa-Dimna as Kitāb al-Hind,
Moreover, Ibn al-Muqaffaʿ (d. 139/757), the erasing Ibn al-Muqaffaʿ altogether. One famous
Arabic translator-redactor of Kalīla wa-Dimna, French descendent, which served as a model for
has all but disappeared behind repeated rewrit- the fables of Jean de La Fontaine, is attributed to
ings. We can no longer tell what he himself may “le sage Pilpay.”5 This is none other than Baydabāʾ,
or may not have composed in the mid-eighth the philosopher who tells the Indian king instruc-
century, because his words have been overwrit- tive fables in the frame tale; he is turned into the
ten by countless later copyist-redactors, many of author of the entire book. Such an attribution
whom did not sign their names. Kalīla wa-Dimna is like declaring Shahrazad the author of The
once belonged among the most widespread texts Thousand and One Nights.
of premodern times. It is a classic, but its author- This essay can offer only a preliminary diagno-
ship has become obscure, even though we glimpse sis of the textual problems and a proposal for their
here and there the name of one Fulān b. Fulān, solution, because research on Kalīla wa-Dimna
who declared his name (not his interference) in has shown that unexpected puzzles lie hidden at
a colophon, as did Ḥasan al-Rabbāṭ, the owner each turn. For instance, the relations among the
of a library of popular literature in Damascus at manuscripts vary from chapter to chapter, and
the turn of the nineteenth century.4 To attempt a even within the same chapter, as does the degree
conventional critical edition of Kalīla wa-Dimna is of proximity (or distance) between the manu-
not only pointless but would ignore and distort its scripts. Therefore, until the totality of the extant
eventful textual history. specimens has been surveyed and compared, one
This work is therefore the site of manifold
polarities: between Ibn al-Muqaffaʿ and his com-
peting anonymous co-authors; between classical
5 After the French translation of Gilbert Gaulmin, Le livre
des lumières ou la Conduite des rois, composée par le sage
4 This library (which was turned into a lending library Pilpay indien, traduit en français par David Sahib d’Isfahan
by the owner’s descendants) has recently been recon- (Paris: Simeon Piget, 1644). See Jean de La Fontaine,
structed by Boris Liebrenz; see idem, “The Library of Fables, préface de Jean-Charles Darmon, dossier et notes
Aḥmad al-Rabbāṭ: Books and Their Audiences in 12th to par Jean-Charles Darmon et Sabine Gruffat (Paris: Librairie
13th/18th to 19th Century Syria,” in Marginal Perspectives Générale Française), 203 and 473 (another French ver-
on Early Modern Ottoman Culture: Missionaries, Travelers, sion of the sage’s name is Bidpaï). Baydabāʾ is also given
Booksellers, ed. Ralf Elger and Ute Pietruschka (Halle: as author on the title pages of some Arabic manuscripts.
Zentrum für Interdisziplinäre Regionalstudien Vorderer The later preface of ʿAlī b. Shāh tells the story of Baydabāʾ’s
Orient, Afrika, Asien der Martin-Luther-Univ. Halle- composition of the book for the Indian king, which is
Wittenberg, 2013), 17–59; and Ibrahim Akel, “Aḥmad ahistorical, since the redaction and the title derive from
ar-Rabbāṭ : sa bibliothèque et son rôle dans la réception, the Middle Persian phase. However, the appearance of
diffusion et enrichissement des Mille et une nuits,” doctoral Baydabāʾ as an author and narrator within the work has
thesis, Littératures, Université Sorbonne, Paris Cité, 2016. certainly contributed to the crediting of him as its author.
A Rat and Its Redactors 5
must refrain from generalizing initial and partial scholarship has hardly moved forward since then.
findings.6 Robert Irwin10 described the problem in 1992 in
There are several ways to access the text of the same manner: “Its [Kalīla wa-Dimna’s] text
Kalīla wa-Dimna, one may choose from (in chron- was not treated with respect by copyists and adap-
ological order): its indirect transmission, begin- tors … it has so far proved impossible to recon-
ning with the ninth century; its translations in struct the exact text written by Ibn al-Muqaffaʿ.”
Europe and the Near East, starting in the eleventh Although Sprengling fell short of his main goal,
century; and complete manuscripts of the work, he identified the sequence of the chapters as a
extant since the thirteenth century. Each of these parameter of classification. The number of chap-
paths is paved with surprises and, paradoxically, ters is relatively stable, but Sprengling found six
although this work appears to be well known, the different sequences in which these occurred,
closer one looks at it, the less familiar it becomes. which he labeled A–F.11 A further sequence has
Here I will take the third path – complete Arabic since been discovered, which I label G.12 François
manuscripts. de Blois finally edited the short and long versions
Any investigation of Kalīla wa-Dimna nec- of Burzoy’s voyage (two of the prefaces) from a
essarily relies on the foundational studies by number of manuscripts, and presents a useful sur-
Antoine Sylvestre de Sacy (who produced the first vey of dated and undated manuscripts preserved
European printed edition), Carl Brockelmann,7 and in European libraries.13
Theodor Nöldeke.8 A pioneering step was taken In the current state of research on this manu-
by Martin Sprengling in 1924 with his attempt at script tradition and its drastically differing ver-
a conventional critical edition and reconstruc- sions, only a synoptic critical digital edition of
tion of the original, which Theodor Nöldeke at the selected specimens can do the work justice, juxta-
time called “a colossal task.” But this project was posing and comparing versions, and investigating
not completed, and subsequently, as Christine van them in their received forms without any attempt
Ruymbeke formulates it, “the enormous excite- at reconstruction. The goal is not to turn back the
ment for the early discoveries around the extant wheel of time, but rather to document the history
versions of the text in several cultural and linguis- of the text in its preserved state, and to analyze the
tic areas came to a brutal standstill several decades context and factors of its textual development. On
ago,” which she ascribes to “disappointing results” this basis, an array of further questions arises. How
and an “overly-complicated field.”9 The state of can a work change so much and yet remain the
same? How exactly did it change? Who changed
6 Many of the phenomena bear comparison with manu- it? And why?
script variation in Medieval European literature, as
treated in the following recent studies: R. Howard Bloch,
Alison Calhoun, Jacqueline Cerquiglini-Toulet, Joachim 10 Robert Irwin, “The Arabic Beast Fable,” Journal of the
Küpper, and Jeanette Patterson, eds., Rethinking the New Warburg and Courtauld Institutes 55 (1992): 36–50.
Medievalism (Baltimore, MD: Johns Hopkins University 11 Martin Sprengling, “Kalīla Studies,” American Journal of
Press, 2014); Stephen Nichols, ed., “The New Philology,” Semitic Languages and Literatures 40 (1924): 81–97. See
Speculum special issue, 65, no. 1 (1990); and Bernard also a short description of the sequences in Gruendler,
Cerquiglini, L’éloge de la variante : histoire critique de la 2013, 397–98 and note 31 below.
philologie (Paris: Seuil, 1986). 12 These are MSS Rabat, Bibliothèque royale (BRR) 3655,
7 His survey of the worldwide spread of Kalīla wa-Dimna dated to 1265–80 CE with its near verbatim copy Paris,
(see note 2) has yet to be superseded. BnF, arabe 3475, dated 1175/1761, Istanbul, Ayasofya
8 For his some of his many articles, which are still valuable 4213, dated 880/1475, and Chicago, Oriental Institute
today, see the bibliography. A 11991 in which the chapters deriving from the two
9 Christine van Ruymbeke, Kashefi’s Anvar-e Sohaili: Rewrit- Indian sources (Pañcantantra and Mahābhārata) are
ing Kalila and Dimna in Timurid Herat (Leiden: Brill, 2016), dovetailed with each other.
321. 13 See de Blois, Burzōy’s Voyage, 66–72 and 81–95.
6 Gruendler
However, before these larger questions are They were selected based on their early date, their
broached, the textual witnesses must be surveyed distribution across different chapter sequences
and two immediate issues resolved. First, how do (according to Martin Sprengling’s classification15),
the manuscripts relate to each other? And second, and their substantial amount of rewriting,16 since
what is unique and specific about each manu- the goal was to arrive at a maximum of variation
script? These initial questions are explored here, to detect as many types of changes as possible. The
based on a small sample, in an attempt to refine following early manuscripts (dating from the thir-
them and discover every possible aspect of com- teenth to the fifteenth century) were consulted:
parison in this fluctuating tradition. Only then can – Istanbul, Süleymaniye Kütüphanesi, Ayasofya
the larger questions about the reasons for this vari- 4095, the earliest known manuscript, dated
ation, its connection with the work’s linguistic reg- 618/1221,17 chapter sequence C var., copyist
ister (which shifts between classical and Middle ʿAbdallāh b. Muḥammad al-ʿUmarī (hereafter,
Arabic), and the agency behind it be addressed. A4095);
In the following discussion, one should bear in – Paris, Bibliothèque nationale de France, arabe
mind that the variation of Kalīla wa-Dimna is the 3465 (P13),18 dated by Bernard O’Kane19 to the
result of mainly written (not oral) transmission,14 early thirteenth century (circa 1220) based on its
and that one cannot automatically privilege some illustrations, chapter sequence A; the beginning
manuscripts over others based on their earlier dat-
ing, since even the earliest ones are separated by 15 See note 11.
five centuries from the Arabic translator-redactor. 16 For the purpose of this analysis, those manuscripts
Conversely, late manuscripts may be based on lost that agree nearly verbatim with some used here (and
which must, with potentially intervening lost copies,
versions from several centuries before. In the com- derive from the same Vorlage) have been excluded.
parison below of a selection of manuscripts, their Nearly identical are P3465 with MS Munich, Bayerische
relative chronology was therefore not considered Staatsbibliothek, Cod. arab. 615, and Rabat, BRR 3655
an a priori factor, since their dating need not nec- with Paris, BnF arabe 3475. P5881 is likewise nearly
identical with MS Istanbul Süleymaniye Kütüphanesi
essarily coincide with that of the versions they Ayasofya 4214 and MS Cairo, al-Maktaba l-Markaziyya
contain. Nonetheless, early manuscripts are still li-l-Makhṭūtāt al-Islāmiyya 1169. Such close resem-
important, since they give the ante quem for the blances appear only among a minority of the manu-
contained redactions. scripts inspected so far, though their exact proportion
remains to be assessed. MSS Riyadh, King Fayṣal Center
2407, P3473, Montreal, McGill University Library 117953
(McLennan no. 94), and Berlin, Staatsbibliothek zu
1 The Manuscript Sample Berlin Wetzstein II 672 are less close to each other but
still share most units. More common is the case where
The seven Arabic manuscripts investigated here a manuscript shows a clear overlap with others while
exhibiting more or less evidence of individual rewrit-
represent a sample from the roughly one hun- ing. Such rewriting is, for instance, moderate between
dred and forty extant specimens so far identified. P3466 and MS Beirut USJ 00022(2), but extensive
between P3466 and Tunis Bibliothèque nationale de
Tunisie 2281.
17 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, ed. Ṭ. Ḥusayn and
14 Scant evidence of oral telling has been found so far, ʿA. ʿAzzām (Cairo: Dar al-Maʿārif, 1941). Although it is
such instances of the narration of selected enframed a critical edition, it contains unacknowledged inter-
tales (oral communication from Heba Tebakhi, ventions and suppletions of lacunae by the editor; the
March 28, 2019). Indeed, manuscripts such as Paris, manuscript has therefore been consulted instead.
BnF arabe 3593, which only contains selected subtales, 18 These sigla refer to the commented list of manuscripts
may have served as a repertoire for oral retelling (for a in de Blois, Burzōy’s Voyage, 66–72.
list of the tales, see René Basset, “Notice sur un manu- 19 Bernard O’Kane, Early Persian Painting: Kalila and
scrit des fables du Kalilah et Dimna,” Journal asiatique Dimna Manuscripts of the Late Fourteenth Century
9, no. 16 (1900): 360–69. (London: I.B. Tauris, 2003), 38.
A Rat and Its Redactors 7
and end of the manuscript have been lost and Two later manuscripts (from the seventeenth
restored (hereafter, P3465); century) have been added for comparison with
– Paris, Bibliothèque nationale de France, arabe the early group, because upon inspection they
3466 (P1), dated to 854/1450 based on a reader’s likewise exhibit substantial rewriting:
note (which is only an ante quem), with unfilled – Paris, Bibliothèque nationale de France, arabe
spaces for illustrations, chapter sequence D 5881, (P6), dated 1092/1681, chapter sequence C,
(hereafter, P3466); illustrated (hereafter, P5881); and
– London, British Library, Or. 8571 (L1), dated – Paris, Bibliothèque nationale de France, arabe
799/1369, chapter sequence C var.; Kalīla wa- 3473 (P7), dated 1110/1699, chapter sequence B
Dimna is followed by an emulation of it in var., with the script alternating between three
verse, entitled al-Ṣādiḥ wa-l-bāghim, by Ibn colors and containing legends of illustrations,
al-Habbāriyya (d. 509/1115–16; hereafter, L8751); but no spaces for the illustrations themselves
and (hereafter, P3473).
– London, British Library, Or. 4044 (L5), dated by The rewriting of Kalīla wa-Dimna continued with
Charles Rieu to the fifteenth century, illustrated, verve during the eighteenth and nineteenth cen-
chapter sequence A var.; Kalīla wa-Dimna is turies. The ever-increasing degree of narrative
followed by Sulwān al-muṭāʿ by Muḥammad and linguistic variation (extending into the dia-
b. ʿAbdallāh b. Ẓafar al-Ṣiqillī, (d. 565/1169 or lectal register) in inspected manuscripts (Paris,
568/1172),20 written in the same hand; the Bibliothèque nationale de France, arabe 3478 with
beginning and end of the manuscript are miss- unfilled blanks for illustrations and dated to the
ing (hereafter, L4044). eighteenth century; and Berlin, Staatsbibliothek,
The label “early” is relative here, since even the Wetzstein II 672, dated 1246/1830, with marginal
oldest extant textual witnesses fall half a millen- legends but no spaces for illustrations22) requires
nium after the original Arabic version. Whatever a separate investigation, because this accelerated
Ibn al-Muqaffaʿ’s pen in fact produced must be redaction process needs to be placed within the
considered lost; the proliferating textual tradi- context of changing sociocultural parameters,
tion overwriting him up to the nineteenth century such as an increased readership and the rise of
should be studied in its own right. The interven- authorship in segments of the population below
ing “dark period” from the mid-eighth to the early the educated elite.23
thirteenth century may be filled in partly by frag-
mentary indirect transmission and partly by trans-
lations. Although both require their own specific
approaches, one major source of indirect trans-
22 This is an autograph by the owner of a library in
mission from the tenth century has been adduced Damascus; see Liebrenz, “The Library of Aḥmad
here as an early point of orientation.21 al-Rabbāṭ,” 19 and 27–28 and note 4 above.
23 See Dana Sajdi, The Barber of Damascus: Nouveau
Literacy in the Eighteenth-Century Ottoman Levant
20 A mirror consisting of beast fables (like Kalīla wa- (Stanford, CA: Stanford University Press, 2013),
Dimna) but also historical anecdotes, composed in reviewed by Jeffrey Sacks, “The Philological Present:
545/1150 and dedicated to Abū ʿAbdallāh Muḥammad Reading the Arabic Nineteenth Century. Review Essay
al-Qurashī, qāʾid of Sicily. For a recent study, see of Four Books,” Journal of Arabic Literature 47 (2016):
Francesca Bellino, “Animal Fables in the Sulwān 169–207; Nelly Hanna, In Praise of Books: A Cultural
al-muṭāʿ by Ibn Ẓafar al-Ṣiqillī,” in Mirella Cassarino, History of Cairo’s Middle Class, Sixteenth to Eighteenth
ed., Islamic Sicily: Philological and Literary Essays, Century, Syracuse, NY: Syracuse University Press, 2003;
special issue of Quaderni di Studi Arabi, no. 10 (2015): and Konrad Hirschler, The Written Word in the Medieval
103–22. Arabic Lands: A Social and Cultural History of Reading
21 See note 26. Practice (Edinburgh: Edinburgh University Press, 2012).
8 Gruendler
2 The Sample Chapter But lest we take the second step before the
first, a brief summary is in order: like most other
For comparing the manuscripts, the short chapter chapters of Kalīla wa-Dimna, “The Cat and the
of “The Cat and the Rat” has been chosen.24 This Rat” opens with a frame dialog between an Indian
belongs to the block of three tales deriving from king (Dabshalīm and variants) and a philosopher
the Mahābhārata25 and lends itself to analysis (Baydabāʾ and variants). Here the king requests
because the manuscript evidence is extremely an example of a situation in which a ruler is sur-
diverse; the chapter also figures prominently in rounded by enemies and is on the verge of perish-
one of the most copious sources for the work’s ing, yet prevails by concluding a pact with one of
indirect transmission, the Muḍāhāt amthāl Kitāb them and honoring it.
Kalīla wa-Dimna bimā ashbahahā min ashʿār In response, the philosopher tells a story in
al-ʿarab by al-Yamanī (d. 400/1009).26 This chapter which a rat cornered by an owl and a weasel
seems to have been of particular interest to later seeks refuge with a cat, who is himself caught
readers; the topic of instrumental, or strategic, in a hunter’s net nearby. The rat proposes a pact
friendship aroused visible reactions, as copyist- of mutual help and survival: if the cat spares the
redactors added, cut, or changed significant por- rat and offers him protection, the rat will gnaw
tions of the text. through the ropes confining the cat. So it comes
to pass. The cat agrees and rescues the rat, and the
rat fulfills his own part – not without some delay,
however (the versions differ as to the rat’s motiva-
24 It is found in the manuscripts on the following folios tions for the delay). Finally, both animals escape to
or pages: A4095 fols. 223v,17–226v,4; L4044, fols. 97v,4– safety. True to the book’s original purpose as a mir-
101v,2; L8751, fols. 41v,11–44r,2; P3465, fols., 117r,3–120r,6; ror for princes (Fürstenspiegel), there is a follow-
P3466, pp. 273,13–282,16; and P5881, fols. 79v,3–81v, 1. up to the tale of mutual rescue, explaining the
25 Within the Mahābhārata, they belong to Book XII,
“The Book of Peace,” which marks a moment of retar-
underlying principle of strategic friendship, i.e.,
dation after the great battle between the Kaurava and friendship that is conditional, useful, and tempo-
Pānḍava families. In it the hero Yudhiṣṭhira, pierced rary. As soon as the danger has passed, the cat calls
by so many arrows that his body rests on the arrows’ the rat back, assuming that their newly concluded
shafts when he is laid down, spends a year giving
friendship will last, but the rat informs the cat that
moral instruction to King Bhīsma and persuades him
not to abandon the throne to become an ascetic but this is not the case. The rat has already explained
to shoulder his duties as a ruler; for a synopsis, see in the second dialog that there are two types of
Moriz Winternitz, Geschichte der indischen Literatur friendship (4327), the first serving a purpose, and
(Leipzig: Amelangs Verlag, 1909), vol. 1, 259–403, esp. the second unselfish. Here the rat identifies the
315, 348–62, and 363. The Mahābhārata reached its
final form between 400 BCE and 400 CE and consti-
present case as the first type: their friendship was
tutes a conglomerate of literary genres whose compo- formed out of a need, which has ended, and there-
sitional entity is still a matter of debate; see Madeleine fore the friendship is over, as it has fulfilled its goal.
Biardeau, Le Mahabharata (Paris: Seuil, 2002). The rat further explains that any outward friend-
26 Abū ʿAbdallāh Muḥammad b. al-Ḥusayn b. ʿUmar
ship of the cat can only be hidden enmity: the sole
al-Yamanī, Muḍāhāt amthāl Kalīla wa-Dimna bimā
ashbahahā min ashʿār al-ʿarab, ed. M. Yūsuf Najm use the cat may have for him is as food (73). The
(Beirut: Dār al-Thaqāfa, 1961). This partisan of pure
Arab lore was irked by the popularity of Kalīla wa-
Dimna and felt impelled to show that its parables 27 The numbers refer to the semantic units into which
(amthāl s. mathal) and wise sayings (ḥikam s. ḥikma) the text has been divided for alignment in the XML/TEI
had antecedents in early Arabic poetry. In the process files, upon which the synoptic edition is based. See the
of juxtaposing both, he cites 165 passages from the list in tab. 1.1 below and the distribution of the units
book. across the manuscripts in tab. 1.2.
A Rat and Its Redactors 9
rat therefore tersely bids the cat “Goodbye” (82 in (all “g”), and the chapter’s closure phrase (“tm”
P3473 and P3466). In L4044 alone, the rat elabo- for tamma “the end”).29
rates: “The gist of the matter, which is the last word The story is related mainly through direct speech.
(wa-l-jumlatu llatī yanqaṭiʿu ʿanhā l-manṭiq), is In an initial monologue, the rat meditates on his
that no path leads to reunion unless a catastrophe precarious situation (units 12–16). Four dialogs
occurs like the one which previously brought you then follow: in the first (17–31), he proposes the
and me together.” pact to the cat; in the second (36–50), he explains
his delay in severing the cat’s ropes; in the third
(54–55), he announces that the time to fulfill his
3 Macroanalyis commitment has come; and in the fourth and lon-
gest (59–82), he explains to the cat the principles
To comprehend and assess the manuscripts’ full of strategic friendship. The following chart (tab.
scope of variation, the analysis of the narrative 1.1, on the next three pages) shows the plot’s units
is conducted on two levels. The divergence is too (listed by their XML-tags), their type (according to
large for an immediate close-up look. Instead, we the above list), and their distribution in each of
first need to obtain a bird’s-eye view of the narra- the seven manuscripts analyzed.
tive structure. The tale is thus observed in its over- At this scale, P3465 and, to a lesser degree, P5881
all composition. Only once this has been done, can lack a number of units, which may be the result
we take a detailed look at the formulation within either of abridgment in these manuscripts or lack
each element. At the first level of analysis, the of expansion, as in other manuscripts.30 In all
sample chapter has been subdivided into mean- manuscripts, the variation is strongest at the chap-
ingful segments (eighty-five units in total, of which ter’s end, where a number of units appear in only
the minimum appearing in a single manuscript is few of the manuscripts (74, 76, 78–81, and 85). In
forty-nine, in P3465; the maximum is seventy-two, terms of the number and sequence of units, P3465
in L4044; this translates into a difference of three in particular differs from the rest in that it is much
and a half folios of thirteen lines vs. four and a shorter and places right at the beginning the rat’s
half folios of fifteen lines). This subdivision takes explanation of his means of self-protection against
account of the framing structure and the diverse the cat, namely leaving one last rope intact until
ingredients of the narrative: the hunter’s arrival (30). Then he begins to gnaw
– Frame dialog (abbreviated in tab. 1.1 as “fd”); through the cat’s ropes (34), at which point the
– Third-person narrative within the fable (“n”); owl and the weasel leave in disappointment (33).
– Interior monologue (“ml”) or dialog (“d”) of the In all other manuscripts, the rat proposes another
characters; strategy, that of displaying their friendship (29), so
– Analogical images (“m” for mathal28) and max- the cat should show joy at the rat joining him (32),
ims (“ḥ” for ḥikma) used by a character to illus- at which point the rat’s other foes leave in
trate a point; and
– Paratexts, such as the title of the chapter’s fable, 29 Already the copyists noticed these structural elements,
as many manuscripts mark inquit formulae and begin-
presented within the frame dialog (“t”), preced-
nings of subtales by colored ink, overstrike, or para-
ing abstract, final summary, and commentary graph symbols (colored dots, inverted apostrophes, or
parallel oblique lines). The modular structure of the
text obviously prompted such copyists to flag salient
passages.
28 In this work, the word mathal, pl. amthāl, means both 30 Abridgment seems more likely, since some of the units
“parable” and “analogical image”; other meanings are absent from P3465 appear in the tenth-century work of
“proverb” and “example,” all of which share the basic al-Yamanī (21, 39, 70, 74, and 75), as do the some of the
meaning of analogy. units absent from P5881 (15, 74, and 76).
10 Gruendler
table 1.1 Comparative chart of text units in the chapter of “The Cat and the Rat” in seven manuscripts
Unit Unit Unit XML Id Muḍa̅ha̅t P3465 P3473 P5881 A4095 L8571 P3466 L4044
Nr. Type
1 g McAbstract 1u
2 fd McKingsQuestionOnPact 1 1 1 1 1 1 2
3 fd McPhilAnswersEmotionsChange par. 85 2 2 2 2 2 2 3
4 fd McPhilAnswersIntelligentResponds par. 85 3 3 3 3 4
5 fd McPhilAnswersIntelligentAsks UQ 4 3 3 4 4 4 5
6 fd McPhilAnswersResoluteReaches 5 4 4 5 5 5 6
7 t McPhilAnswersParabCatAndRat 6 5 5 6 6 6 7
8 fd McKingRequestsStoryPhilNarrates 7 6 6 7 7 7 8
9 n McRatsHole 8 7 7 8 8 8 9
10 n McCatIsCaught 9 8 8 9 9 9 10
11 n McRatSurroundedByFoes 10 9+ 9 10 10 10 11
12 ml McRatKeepsAlert 11 10 10 11 11 11 12
13 ḥ McIntelligentAlwaysKeepsAlert 12 11 11 12 12 12 13
14 m McIntelligentIsLikeSea 13 12 12 13 13 13 14
15 h McIntelligentNotBlindedBySituation par. 59 14 14 14 14 15
16 ml McRatWillSeekPactWithCat 15≠ 13 13 15 15 15 ≠ 16 ≠+
17 d McRatApproachesAndQuestionsCat 16 14 14 16 16 16 17
18 d McCatAdmitsBeingTrapped 17 15 15 17 17 17 18
19 ml McRatResolvesToBeSincere 16u
20 d McRatDescribesCommonProblem 18≠ 16 17 18 18 18+ 19
21 ḥ McFailWithoutMutualTrust par. 60 17 18 19 19 19 20
22 d McRatPromisesFaithfulness 18 19 20 20 20 21
23 ḥ McIntelligentDoesntPostpone 19 20 21 21 21 22
24 d McCatRatCommonRescue 20 22 22 22 23
25 m McSailorsShipCommonGoal 19 21 21 23 23 23 24
26 d McCatRatMutualHelp 24u
27 ḥ McThusFateOccurs 25u
28 d McCatTrustsRat 20 22 22 24 24 25 26
A Rat and Its Redactors 11
table 1.1 Comparative chart of text units in the chapter of “The Cat and the Rat” in seven manuscripts (cont.)
Unit Unit Unit XML Id Muḍa̅ha̅t P3465 P3473 P5881 A4095 L8571 P3466 L4044
Nr. Type
29 d McRatExplainsStrategy 23 23 25 25 26 27
30 d McRatExplainsLastRope 21u
31 d McCatAcceptsPact 28u
32 n McCatFulfillsHisPart 24 26 26 27 29
33 n McFoesLeaveFrustrated 23 24 27 27 28 30
34 n McRatBeginsCutRopeA 22 25 25 28 28 29 31
35 n McRatBeginsCutRopeB 26u
36 d McCatDeemsFulfillmentSlow 24 27 26 29 29 30 32
37 ḥ McNobleMustHelp 25 28 27 30 30 31 33≠
38 d McCatStatesRatOwesReciprocation 26 29 28 31 31 32 34
39 ḥ McNobleSubstitutesFriendship par. 79 30 29 32 33 35
Enmity
40 ḥ McBetrayalIsPunished par. 62 27 31 30 33 32 34+ 36
41 d McRatMustFulfillPact 35p
42 d McRatBeginsDefence 28 32 31 34 33 36 37
43 ḥ McFreeVsForcedFriendship par. 153 29≠ 33 32≠ 35 34 37 38
pt. A
44 ḥ McIntelligentTradesFavors par. 153 30 34 33 36 35 38 39
pt. B
45 ḥ McFriendshipForProfit 35 34 37 36 39 40
46 d McRatExplainsSelfProtection 36 35 38 37 40 41
47 n McRatBeginsCutRopeC 31u
48 m McIlltimedHarvest 37p
49 ḥ McEverythingHasItsTime 38 39 38 41 42
50 d McRatExplainsCutRopesButOne 39 36 40 39 42 43
51 n McRatCutsAllExceptOneRope 40 41 40 43 44
52 n McHunterApproaches 32 41 37 42 41 44 45
53 n McCatExpectsToDie 45 46
54 d McCatRequestsFulfillment 33u
55 d McRatAnnouncesCutLastRope 42 38 43 42 46 47
56 n RatBeginsCutLastRope 48u
57 n McRatFulfillsPact 34 43 39 44 43 47 49
58 n McCatRatEscape 35 44 40 45 44 48 50
59 d McCatCallsBackRat 36 45 41 46 45 49 51
60 d McRatsRefusal 42u
61 d McCatRemindsRatOfFriendship 37 46 43 47 46 50 52
62 ḥ McGivingUpFriendLoosingOut par. 154 38 47 44 48 47 51 53
12 Gruendler
table 1.1 Comparative chart of text units in the chapter of “The Cat and the Rat” in seven manuscripts (cont.)
Unit Unit Unit XML Id Muḍa̅ha̅t P3465 P3473 P5881 A4095 L8571 P3466 L4044
Nr. Type
63 d McCatSuggestsReward 39 48 45 49 48 52 54
64 d McRatDistinguishesHiddenOvert par. 102 40 49 46 50 49 53 55+
Enmity
65 m McRidingElephantTusk 41 50 47 51 50 54 56
66 ḥ McEnemyBefriendedInNeed par. 104 42 51 48 52 51 55 57
67 m McYoungCattleNeedParents 43 52 49 53 52 56 58
68 m McCloudsRainOrNot 53 50 54 53 57 59
69 ḥ McEnmityTurnsFriendshipInNeed par. 105, 44 54 51 55 54 58 60
106
70 m McHeatedWaterCoolsDown par. 106 55 52 56 55 59 61
71 d McRatsNeedForFriendshipEnded 56 53 57 56 60 62+
72 ḥ McWeakMustAvoidStrong 57 54 58 57 61 63
73 d McCatNeedsRatForFood 58 55 59 58 62 64+
74 ḥ McCautiousWeakBetterThan par. 108 60 59 63 65
DeludedStrong
75 ḥ McIntelligentEndsForcedFriendship par. 109 59 56 61 60 64 66
76 h McIntelligentDistrustsForced par. 103 45+ 60 62 61 65 67
Friendship
77 h McIntelligentMustLeaveForced 68u
Friendship
78 ḥ McBestThingIsDistanceFromEnemy 61 63 62 69
79 d McRatAwayFromCatCatAwayFrom 62 64 63 66+ 70
Hunter
80 d McRatLovesCatFromFar 46 63 57 65 64 67
81 d McRatDoesntNeedReward 47 64 68 71
82 d McLastVerdictNoReunion 48 65 58 66 65 69 72+
83 g McPhilFinalCommentary 59u
84 g McSummaryGist 66p
85 tm McEndOfChapter 49 66 60
table 1.2 Chart of shared text segments in the chapter of “The Cat and the Rat,” in manuscripts of Kalīla wa-Dimna
Distribution of manuscripts based on sum of units and average degree of overlap (Mc chapter)
BWII Istanbul, Ayasofya 4095
58 (1221 CE)
52
Average degree of unit overlap
frustration (33). The self-protection follows here right in tab. 1.1: A4095, L8751, P3466, and L4044)
as part of the rat’s subsequent strategy, which share a body of units to which unique ones are
is moved to the second dialog and serves as his added in most cases. The overall impression is that
defense when the cat complains about the rat’s of a fluctuating and cumulative process, which
slowness in gnawing though the ropes (46, 50). I describe as a redactional continuum.
Also evident is the lack of anything one might The cumulative trend is visualized in tab. 1.2,
term “groups.”31 Even in the large-scale compari- which locates the manuscripts on a chart showing
son, no manuscript shares all units with any other. their total number of units (horizontal axis) and
Rather, some manuscripts (the four last cols. on the the proportion of their units that they share with
other manuscripts (vertical axis). For instance,
the short and idiosyncratic P3465 is on the lower
31 This term has been used by Martin Sprengling (see
left and shares few units with other manuscripts,
note 11) to classify manuscripts by the sequence of
their chapters. However, on close inspection, the chap- whereas the long and inclusive P3466 and L4044
ter sequence does not always tally with the relative are on the upper right.
similarity (or difference) between manuscripts based The particular continuum in the sample chap-
on shared text units. The designation of the sequence ter, however, cannot be generalized to the full
has been retained as merely one factor of classification
and is recorded in tab. 1.1, line 2. Sprengling’s assertion
manuscripts under investigation. In this chapter,
that sequences A and C dominate among Arabic man- A4095, L8751, P3466, and L4044 are closer to each
uscripts has so far proved true. other than to the remaining manuscripts. The
14 Gruendler
table 1.3 Comparative chart of text units in Ibn al-Muqaffaʿ’s preface of “The Purpose of the Book,” in seven manuscripts
Unit Unit Unit XML Id P3465 P3473 P3466 P5881 A4095 L8751 L4044
Nr. Type
1 n ImBaydabasComposition
2 g ImUniqueBookOfWisdom 1u
3 n ImIndiansCodeAnimalSpeech 1 2 1 1 1 1
4 n ImBookReunitesAspects 2 3 2 2 2 2
5 m ImFarmerObtainsWeeds 4u
6 n ImBooksUsesAreWisdomPlay 3 5 3 3 3 3
7 m ImBequeathedTreasure 4 6 4 4 4
8 n ImKnowledgeManyBranches 5 5 5
9 n ImReaderIgnoringCodeNoBenefit 5 7 4 6
10 n ImReaderToReflect 1 6 6 7
11 n ImFastReaderNoBenefit 6 8 5 2 7 7 8
12 n ImIndiscriminateReaderNoBenefit 7 9 6 3 8 8 9
13 m ImBuriedTreasureLost 8 10 7 4 9 9 10
14 ml ImDiscovererHiresHelp 9 11 5 10 10 11
15 m ImDiscovererLoosesTreasure 10 12 6 11 11 12
16 ḥ ImReadingNotInDepthNoBenefit 11 13 7 12 12 13
17 m ImNutCrackedForBenefit 12 14 8 13 13 14
18 n ImGleanSecretKnowledge 9 14 14 15
19 m ImMemorizedSheetNoKnowledge 13 15 8 10 15 15 16
20 n ImUnderstandingReaderMustAct 14 16 9 11 16 16 17
21 ḥ ImReasonSelfDenialNeverTooMuch 17 10 12 17 18
22 ḥ ImLifeNeedsCultureKnowledge 13 18 19
23 m ImNoScorningCultureDeath 19 17
24 m ImCultureIsLikeSpark 18 11 14 20 18 20
25 ḥ ImCultureRaisesKnowledgeForUseA 19 12 15 21 19 21
26 m ImManThrowsBallHigh 22 22
27 ḥ ImBuildingRestsOnBasis
28 n ImCultureRaisesKnowledgeForUseB 23u
29 m ImManNoticingBurglarFallsAsleep 15 20 13 16 24 20 23
30 ḥ ImUnusedKnowledgeIsIncomplete 16 21 14 17 25 21 24
31 m ImKnowledgeActionIsTreeFruit 17 22 15 18 26 22 25
32 ḥ ImNoActionNoScholar 18 23 19 27 23 26
A Rat and Its Redactors 15
table 1.3 Comparative chart of text units in Ibn al-Muqaffaʿ’s preface of “The Purpose of the Book” in seven manuscripts (cont.)
Unit Unit Unit XML Id P3465 P3473 P3466 P5881 A4095 L8751 L4044
Nr. Type
33 n ImManLearnsOfLostTraveller 20 28 24 27
34 m ImInformedManTakesRiskyRoad 19 24 16 21 29 25 28
35 n ImManErrsDespiteInfo 20 25 22 30 26 29
36 m ImInformedPatientTakesBadFood 21 26 23 31 27 30
37 ḥ ImWhoDistinguishesHasNoExcuse 22 27 24 32 28 31
38 m ImSeeingManInDitchNoExcuse 23 28 17 25 33 29 32
39 ḥ ImScholarMustTeachHimself 24 29 18 26 34 30 33
40 m ImSourceBenefitsOnlyOthers 25 19 27 35 31 34
41 m ImSilkWormBenefitsOnlyOthers 26 30 20
42 ḥ ImLearnerMustTameSoul 27 31 21 32 35
43 ḥ ImThingsNeededForWorldlyLife 28 32 22 28 36 33 36
44 ḥ ImNoBlamingFaultsOneHasToo 29 33 23
45 ḥ ImActionPursuedForBenefit 29 37 34 37
46 m ImSomeReadersErrInThis 38 35 38
47 m ImPassiveReaderLikeBlind
48 m ImBlindBlamingBlind 30 34 24 39 36 39
49 n ImNoBenefitByOthersHarm 30 40 37 40
50 n ImPursuitMustHaveLimit 31 35 38 48 45 48
51 ḥ ImWhoStrivesAimlesslyFails 39u
52 ḥ ImJourneyWithoutLimitKills 32 36 40 49 46 49
53 n ImNoSeekingTheUnreachable 33 37 41 50 47 50
54 n ImNoPreferringHereToHereafter 34 38 42 51 48 51
55 ḥ ImDetachmentEasesDeath 35 39 43 52 49 52
56 ḥ ImNoDespairDespiteHardship 53 50 53
57 ḥ ImTwoThingsSuitAll 36 40 25 44 54 51 54
58 ḥ ImTwoThingsDoNotSuitAll 37 41 26
59 m ImFireBurnsAll 38 42 27 45 55 52 55
60 m ImFireWaterDoNotJoin 39 43 28
61 n ImGodsGiftUnexpected 40 44 46 56 53 56
62 m ImPauperWinsBurglarsGarb 41 45 47 57 54 57
63 ml ImThiefTakesJar 42 46 48 58 55 58
64 m ImPauperCatchesThief 43 47 49 59 56 59
65 m ImIntelligentMustNotDespair 50 60 57 60
66 n ImStrivingNotExpectingLuck 44 48 51 61 58 61
67 n ImFortunateFewNoModel 45 49 52 62 59 62
68 n ImStrivingMassesAreModel 53 63 60 63
69 n ImStrivingForBenefitNotHarm 46 50 54 64 61 64
70 m ImDoveRepeatsLossOfChick 47 51 55 65 62 65
71 m ImJudgeOfLegalTricks
72 n ImGuardingGodsMeasure 48 29 56 66 63 66
16 Gruendler
table 1.3 Comparative chart of text units in Ibn al-Muqaffaʿ’s preface of “The Purpose of the Book” in seven manuscripts (cont.)
Unit Unit Unit XML Id P3465 P3473 P3466 P5881 A4095 L8751 L4044
Nr. Type
73 ḥ ImStrivingForHereHereafter 49 52 30 59 67 64 68
74 ḥ ImStrivingForHere 50 53 31 58 65 69
75 ḥ ImStrivingForHereafter 57 67
76 ḥ ImThreeThingsForWorldlyMan 51 54 32 60 68 66 70
77 ḥ ImThingsMarkingTheUseless 52 55 33 61 69 67 71
78 n ImManyAreCredulous 53 56 34 62 70 68 72
79 n ImBelievingInformation 63 71 69 73
80 n ImCheckingOnesDesire 54 57 35 64 72 70 74
81 n ImCheckingInfoForTruth 55 58 36 65 73 71 75
82 m ImContinuingOnWrongPath 56 59 37 66 74 72 76
83 m ImLoosingItchyEye 57 60 38 67 75 73 77
84 n ImHeedingDecreeResolve 58 61 39 68 76 74 78
85 n ImDoUntoOtherAsToOneself 59 62 40 69 77 75 79
86 n ImReaderMustHeedPreface 70 78 76 80
87 n ImUnderstandingBookBenefits 71 79 77 81
88 n ImThrowingStoneInDark 80u
89 n ImTransAddsChapterToExplain 72 81 78 82
90 m ImMerchantStealsFromPartner 60 63 41 31 41 38 41
91 ml ImMerchantDevisesPlot 61 64 42 32 42 39 42
92 m ImMerchantExecutesPlot 62 65 43 33 43 40 43
93 ml ImPartnerFindsGarb 63 66 44 34 44 41 44
94 m ImMerchantStealsOwnBale 64 67 45 35 45 42 45
95 m ImMerchantsSeesErrorAtHome 65 68 46
96 m ImMerchantSeesErrorInShop 36 46 43 46
97 m ImMerchantSilentAboutBetrayal 37 47 44 47
98 ml ImPartnerFearsAccusationMono 66 47
99 d ImPartnerFearsAccusationDialog 67 69 48
100 d ImMerchantAdmitsTreachery 68 70 49
101 d ImPartnerRequestsExplaining 69 71 50
102 d ImPartnersParableMerchantThief 70 72 51
103 m ImThiefTakesGrainForGold 71 73 52
104 m ImThiefEyesJar
105 m ImThiefTakesGrain
106 d ImMerchantForfeitsTrust 72 74 53
107 n ImReadingForParables 73 75 54
108 m ImTwoBrothersSquander 74 76
109 ml ImYoungestReflectsOnWealth 75 77
110 ḥ ImMisspendingRichIsPoor 76 78
111 ḥ ImRightSpendingEarnsPraise 77 79
112 ḥ ImFalseSpendingBringsRegret 78 80
A Rat and Its Redactors 17
table 1.3 Comparative chart of text units in Ibn al-Muqaffaʿ’s preface of “The Purpose of the Book” in seven manuscripts (cont.)
Unit Unit Unit XML Id P3465 P3473 P3466 P5881 A4095 L8751 L4044
Nr. Type
113 ml YoungestSpendsOnSiblings 79 81
114 n ImReaderMustTakeTime 80 82 55
115 m ImFisherFindsEmptyShell 81 83 56
116 m ImFisherMissesPearl 82 84 57
117 n ImSuperficialReadersMissMeaning 83 85
118 n ImPleasureSeekingReaders 84 86
119 m ImGardnerNeglectsHarvest 85 87
120 d ImReaderNeedsStaminaFocus 88 58
121 m ImPhilDiscipleTwoOwners 89u
122 m ImFriendsReachAgreementBy
Counseling
123 d ImPrefaceToExplainHighAudience 59u
124 n ImBookHasFourGoals 86 90 60
125 tm ImEndOfPrefaceBegContents 87 61
126 tm ImEndOfPrefBegLvOrBu 82u
127 ḥ ImCultureLightsUpHeart
128 m ImSunLightsUpEarth
129 ḥ ImBestNatureHonestGain
130 m ImDeedsRewardedEncounterSure
131 n ImReaderToConsultBook
modalities of combination vary, and the question copyist between the thirteenth and the seven-
requires further investigation. teenth century understood the fable, his implicit
A third phenomenon is that most manuscripts33 verdict varies.
show unique units at different points, which means Returning to the immediate questions posed in
that individual copyists selectively expanded the introduction, the macroanalysis has helped to
the tale. These unique units per manuscript are refine the first of them. In the search for the interre-
extracted here from tab. 1.1 and arranged by type: lationships between the manuscripts, one discerns
no discrete groups but rather a fluid development
in which each specimen marks a point along a
P3465 P3473 P5881 L8751 P3466 L4044 process of accretion, excision, and reformulation,
a sort of mouvance par écrit, to adapt a term Paul
n 30, 47 35 26, 41 56 Zumthor has coined for oral medieval French lit-
ml/d 54 19, 60 31 erature.34 As to the second question, the bird’s-eye
m/ḥ 48 27, 77 view shows that there exists a unique quality to
g 83 84 1 most manuscript versions in the form of certain
Total 3 2 3 1 2 5 text segments that occur there and nowhere else.35
To proceed further, the changes within each of the
units now need to be observed close up.
Most of these units add more detail to the action
or speech to the characters, although new analo-
gies, maxims, and paratexts appear as well. L4044 4 Microanalysis
stands out by containing more instances of rewrit-
ing than any other manuscript. The following analysis was aided by the alignment
In terms of the distribution of the changes of corresponding text segments in the different
across the chapter, the synopsis of units (tab. 1.1) manuscripts with the synoptic LERA-tool,36 which
shows the greatest fluctuation in the fourth dialog, allows easy comparison across all manuscripts
in which the rat explains the difference between and immediately shows the different amount of
a friendship formed by choice versus a friendship text per segment in each specimen.
induced by outer forces. He further explains why At first glance, the formulation of P3465 within
the second type of friendship necessarily ends shared units is often shorter than elsewhere. At
when these forces disappear and the situation times P3465 merely provides a kernel of text that is
has changed. Scribal intervention intensifies sur- integrated into a longer passage in the other man-
rounding this controversial topic. What is friend- uscripts (notably 63 and 69, and to a lesser degree
ship? Can it be instrumentalized? These questions 34, 40, and 57). Identical phrases recur through
necessarily generate different answers, based on many manuscripts in multiple combinations, but
the situation to which they are applied. A strategic often phrases appearing in either P3473 or P5881
pact between rulers is guided by a different set of are contained in both P3466 and L4044 (e.g., 11,
morals than a personal relation between individu-
als. According to the context in which a particular 34 Paul Zumthor, Essai de poétique mediévale (Paris:
Éditions du Seuil, 2000), 84–96 and 610.
35 The designation of such passages as unique remains
expanded version is noted on the margins, partly preliminary, because parallels may appear in manu-
resembling P3466 and L4044 and partly unique to this scripts still to be inspected.
manuscript. MS Oxford, Bodleian Library, Pococke 400 36 On LERA, see note 1. The manuscripts were tran-
likewise shows signs of cross-copying. scribed and digitized by the team of Kalīla and Dimna –
33 Even the oldest dated manuscript, A4095, see tab. 1.3. AnonymClassic, all tables designed by Mahmoud Kozae.
A Rat and Its Redactors 19
20, 69, 71, and 75). In these units, the accretion is will focus on intentional, i.e., redactional changes,
additive from P3473 or P5881 via A4095, L8751, and which substantially affect the text in the chosen
P3466 to L4044; the manuscripts have therefore chapter. Rewriting within the units can best be
been ordered in this sequence from left to right in shown by example, for which one scene has been
the chart, since this shows the cumulative process selected, that in which the rat, having assessed his
of redaction. The continuum from A4095 to L4044 situation, approaches the cat to propose a joint
(hereafter, the “London continuum,” after the loca- strategy. This is unit 20, labeled “Rat describes
tion of two of the manuscripts) tends to reunite common problem,” and counts among the strongly
formulations that appear scattered among other fluctuating units of the fable (see tab. 1.4).
manuscripts. The pair of P3466 and L4044 at the In this unit, every manuscript has been com-
right end in tab. 1.2 is most inclusive in assembling pared with every other, and a total of nineteen
a majority of units extant in diverse other speci- different types of correspondence appear, though
mens (although both manuscripts also contain only the most frequent relationships of similarity
much unique material, as will be shown). have been marked in color. Passages shared by all
In terms of the interrelationships between the manuscripts (marked in gray) are relatively few
manuscripts, the microanalysis across all units in number. A greater portion is taken up by pas-
confirms the results of the large-scale view, namely sages shared by some manuscripts (marked in
that of a redactional continuum in which formu- shades of blue according to the key in the graph;
lations fluctuate. The amount of change within for instance, light blue marks passages coinciding
the different units is not stable and constant, but from left to right in all manuscripts from P3473 to
varies considerably. There are some in which the L4044, i.e., columns b to g). A substantial portion
text changes little, although full verbatim repeti- of some manuscripts is constituted by those pas-
tion across all manuscripts is rare. Where it does sages that appear in only one manuscript (marked
occur, it often results from a binary syntactic or in magenta).
semantic figure (parallelism, antithesis) whose These unique passages represent either altered
tight structure stabilizes the formulation. In other or added formulations (from single words to full
units, massive rewriting occurs, especially in those sentences). These “edits” can be subdivided into
essential to the chapter’s main message – and it is five types:
characteristic of fables to include their own com- − Synonymous paraphrase;
mentary. This selectiveness of the rewriting pro- − Reformulation with a changed sense, although
cess proves that the intervention of the copyists by means of minor alterations;
was intentional and targeted, and not random or − Reinterpretation or adaptation of the conso-
thoughtless. nantal skeleton (rasm), yielding anything from
a paraphrase to the introduction of a new
meaning with more or less rewriting to contex-
5 Types of Changes tualize it;
− Proliferation of the text through the rephrasing
In surveying the changes the text undergoes, one or combination of elements that appear else-
needs to separate the intentional from the unin- where in the same chapter; and
tentional as much as possible. Among the latter,
besides copyists’ errors, homeoteleuton recurs in
passages with repetitive formulations.37 Here I logically expected. This has been described by John
Dagenais in Medieval Spanish scribal practice; see
idem, The Ethics of Reading in Manuscript Culture:
37 Another is the phenomenon of repetitio, i.e., the reuse Glossing the Libro de buen amor (Princeton, NJ:
of a prior or following word where another word is Princeton University Press, 1994), 132–34.
20
table 1.4 Synoptic edition of unit 20, “Rat describes common problem” based on seven manuscripts of Kalīla wa-Dimna. The color key on the lower left shows
their interrelation. Grey shading marks passages identical in all MSS; shades of blue mark select correspondences between specific manuscripts as
indicated; magenta shading marks passages limited to one MS only. Particles have been disregarded, as they cannot be unequivocally assigned.
Gruendler
A Rat and Its Redactors 21
− Substantial addition of entire phrases or sen- What harmed you used to bring me joy, and I
tences in which the copyist most visibly acts as used to consider your confinement to be my
a silent co-author. freedom, but today … (A4095 and similarly
L8751/P3466/L4044)
ّة ن كا ن� ���س ّ �ن �م�ا ��س�ا ء ك �أ � �م�ا ض���ّ�� ق
5.1 Synonymous Paraphrase
�ك
�ي ��
لو ��
ع ��س� �ل ك�
� �ل
ي �ع�و ر ى ي �� ي ر ي
This type occurs in the rat’s opening statement. ي
... ا �ل�يو
م
Surrounded by foes, he seeks the help of the cat,
who is caught in the hunter’s net. The rat wants
to propose that the cat shelter him, while he him- In P3473, however, the same sentence is negated
self bites through the cat’s fetters. But first he must and slightly expanded to make it apply to the rat’s
establish trust and convince the cat of his sincer- present feelings toward the cat:
ity. The words he uses differ in the manuscripts,
though the sense remains essentially the same I am sad about what harmed you, and the
(portions unique to one manuscript are marked confinement I see you in gives me no free-
bold): dom or peace (P3473).
ة أ
These words of mine contain no lie or decep- إن ّي لحزين لم�ا ��س�ا ء ك و�م�ا � را ه �م��ن ا �ل�ض����ي����ق�� ع��لي��ك ليس
tion (P3465). ة
ّب���س�ع�� ولا راحة علي
ل خ��د ��ع��ة � ف� ذ
��� ب� وا ي
وكلامي هذا ل��ي��س ي���ه ك Through the shift, the rat’s new feelings (of sympa-
thy) are given more weight than his former ones.
I am sincere … you will know that my words
to you are neither false nor foolish (P5881). 5.3 Reinterpreted or Adapted Consonantal
ت ز أ ّن �ق ت ت ّ أن خ
� �ا م
Skeleton (rasm)
و����س���ع��لم [�م�� �ا �ل�ي�] �ل�ك � ���ه�ا �ل��ي����س�� ب�� ور... ���ل���ص�ا This type of change is specific to the Arabic
ولا ب�ا ط�ل script, which, as a reductive script (an alphabet
of the abgad type), does not note short vowels.38
I am speaking to you without lying (P3466). Furthermore, the script used in Arabic manu-
�ثً ل أ ذ ف ّ نّ أ
�إ �ي� � ح�د ث��ك ح�د �ي��ا ا � ك
scripts is reductive to a higher degree than that
��� ب� �ي���ه of modern print, because these manuscripts do
not consistently include diacritic dots on homo-
These edits show that copyists also deemed the graphs. In some cases, this reduces the number of
style of their Vorlage (and not just the content) graphemes from twenty-eight to as few as twenty-
worthy of improvement and felt free to reformu- two. Manuscripts may further dispense with sev-
late passages as they saw fit. eral symbols that have become typographical
standard. First, the glottal stop (hamza) is mostly
5.2 Reformulation with a Changed Sense omitted. This is a consonant in classical Arabic
The variety of ways in which this can occur is lim- and is only represented by a supralinear sign (not
itless, and the following example from this unit
serves merely to show one of many possibilities.
In four manuscripts, the rat describes his gloating 38 For the function and applications of this writing sys-
tem, see Beatrice Gruendler, “Stability and Change in
at the cat’s misfortune as a former attitude that he Arabic Script,” in The Shape of Script: How and Why
no longer holds (in P3466, the words are slightly Writing Systems Change, ed. Stephen D. Houston (Santa
reformulated): Fe, NM: School for Advanced Research, 2013), 93–118.
22 Gruendler
a letter) in the Arabic alphabet. In manuscripts, it were made in order to recontextualize readings
can be indicated by the madda sign, if following a that had altered the meaning.
long ā (alif mamdūda), or it can be replaced by a In this light, the fact that, in many literary genres
glide (w, y) if placed between two different vow- and scholarly disciplines, the texts of Arabic man-
els. Elsewhere it is completely left out. Thus, the uscripts remained relatively stable is remarkable;
letter alif can stand either for the glottal stop or this stability is owed to well-established conven-
the long vowel ā. Second, the doubling of conso- tions of teaching and transmission and to the schol-
nants (marked in modern print by the supralinear ars and professional copyists who upheld them.
shadda sign) is often left unmarked. The present work, however, being sui generis –
These aspects of reduction and polysemy work a manual of statecraft presented as a collection of
together in allowing varying readings of an Arabic fables – slipped from high adab to the popular reg-
manuscript, especially in view of the vast vocabu- ister over the centuries and so moved beyond the
lary of classical Arabic, which was a formal idiom reach of scholarly control.
of high prestige, mastered fully only by the edu- As a result, the contextual reading and comple-
cated. At the same time, the abjad system was tion which the compact Arabic script required
conceived for Semitic languages, with their char- from the copyists of Kalīla wa-Dimna encouraged
acteristic relationship between lexical root and them to rethink their Vorlagen unchecked by schol-
morphological pattern, in which the completion ars. In particular, the copyists’ different readings of
of unwritten short vowels and case endings was the consonantal skeleton (rasm) produced every-
governed by rules. thing from synonymous to diverging (or illogical)
Reading Arabic script is therefore an interac- interpretations of a written passage and even the
tive process in which one needs to supply what is introduction of completely new ideas. This type of
missing on the page, though this is easy to do (in change further proves that alterations happened
most cases) by someone familiar with the written though rewriting, as opposed to oral retelling.40
Arabic language. As a result, a premodern user of In some cases, such reinterpretation produces
Arabic manuscript codices would be accustomed a cognate meaning. In others, however, it gives a
to reading contextually and filling in the gaps left sentence a new focus, as in the alternative inter- أ
in the written form. Such a contributive kind of pretation of a word’s أrasm, such as baʾs ()ب�� ��س
reading differs fundamentally from the reading of “courage” or yaʾs (“ )ي�� ��سdespair,” depending on
a more fully phonetic alphabet, such as Spanish or the addition of a single dot underneath the first
German written in Latin letters.39 To a reader with letter. Within its context, this rereading affects
limited competence in classical Arabic, however, the motivation for cooperating with an enemy.
completing the meaning posed problems, and one What reads in P3465 (4) as “The intelligent man
may argue that it was precisely the incomplete uses … a strategy … which is intrepidity towards
grasp of classical Arabic by some of the later copy- his enemy or kindness towards his friend” (wa-dhū
ists that spurred the reinterpretation of the script l-raʾyi yuḥdithu … raʾyan … ammā min qibali
and concomitant alterations in the text, which l-ʿaduwwi fa-bi-l-baʾsi wa-ammā min qibali l-ṣadīqi
fa-bi-l-istiʾnās) reads in L4044 inversely as “…
when one desires something from the enemy and
39 To be clear, Arabic script has the inbuilt option to ren- despairs of something from one’s friend” (… ʿinda
der a text fully phonetically. Yet this was exercised in
texts to a varying degree, according to the text type,
its readership, or the commissioning patron. In manu- 40 So does the alternation between nearly verbatim or
scripts of Kalīla wa-Dimna, diacritics and vocalization closely related passages with others that are com-
likewise range from being both fully present (e.g., in pletely rephrased or newly added, as mentioned in sec-
P3465) to almost completely absent (e.g., in L4044). tion 4 above.
A Rat and Its Redactors 23
l-ṭamʿi fīmā qibala l-ʿaduwwi wa-l-yaʾsi bimā ʿinda and the active participle of I. stem of the root j-n-y
l-ṣadīq). With the addition of the dot, the word is “to harvest”):
changed from “intrepidity” to “despair” and reas-
signed from describing the attitude shown the foe I am keeping my promise to you, while pro-
to that shown the friend. Moreover, the alternative tecting myself from incurring that which
options of behavior toward friend vs. foe in P3465 happened to him who harvested fruit at the
become simultaneous in L4044. The process may wrong time. (P3473)
have begun with a misspelling, but the adjustment
�غ قت �ل ن ث ف أ
of the surrounding particles to make sense of the ��ير و����ه ���ا �ي� �ل�ل�����مر �ي
�م�ا � �ص�ا ب� ا ج
reread rasm shows the scribe’s intention. Such
multi-step developments could occur only in the The new meaning of “the harvester” (al-jānī) is
absence of the scholarly transmission of a text, followed by the unique addition of a phrase that
which would reign in alterations at every step in makes sense of this and leads to the creation of a
the copying or dictation. new unit, the “Ill-timed harvest.” Here the copy-
When a reinterpretation or adaptation of the ist has heeded the context well, for his addition
rasm causes a change of meaning, this may prompt fits with the following unit, namely the maxim:
further conscious and deliberate interventions by “Everything has its time.”
the copyist-redactor in order to fit the newly gen- In the two previous examples, the new reading
erated meaning into the context. One such case occurs without any changes to the rasm itself. In
occurs in the unit in which the rat explains his other instances, however, not only are diacritics
strategy of self-protection (46). Having been saved, and additional signs changed (or missing ones sup-
the rat is taking his time in gnawing through the plied differently), but letters are also mistaken for
cat’s net, and the cat complains, suspecting the others that look similar in the manuscript hand.
rat of defaulting on his promise. The rat defends Another creative instance of adapted rasm
himself by saying that since he is the weaker party, occurs at the moment when the rat finds himself
he needs to protect himself. Here one rare word, surrounded by foes and tries to retain his sang-
aljaʾanī (a past tense verb in the IV. stem of the froid (12). In a number of manuscripts (A4095/
ً
root l-j-ʾ “to seek refuge,” with suffix pronoun), is L8751/P3466/L4044), he decides that he will not
open to interpretation, because the letter alif lacks allow his “mind to go to pieces (( ”)� �ش� �ع�ا ع�اlā yadh-
the hamza sign that would mark its function as a haba ʿaqlī shuʿāʿan). In Berlin, Wetzstein II 672,
glottal stop (ʾ) as opposed to a long ā. The formula- dated 1830, the rat’s mental process is detailed
tion below of the rat’s statement occurs thus (or further, and the reinterpretation of the rasm gives
similarly) in the majority of the manuscripts: rise to a metaphor: “I will make the ray ([ )� �ش� �ع�ا ع�هof
my ratio] shine upon my heart” (ajʿalu shuʿāʿahū
I am keeping my word to you, while guarding [sc. ʿaqlī] ilā lubbī).
myself lest the very thing which made me A semantic change of even wider proportions
have recourse to a truce with you happen occurs in Ibn al-Muqaffaʿ’s preface, in the passage
to me … (A4094 and similarly L8751/P3466/ (124) describing the four goals of Kalīla wa-Dimna
L4044) as a book. The book’s second goal, as formulated
in P3465, is أto be a companion for kings (li-yakūna
ًن
����م�ا �هو ألجأني �إ لى �ص��ل
ح�ك unsan ( )� ���س�اli-qulūbi l-mulūki), and therefore it
must be made more attractive with illustrations.
In P3473 the same segment is reformulated and In later manuscripts, the relevant word, unsan
the word is analyzed as al-jānī (the definite article (thus in the accusative), is reread as nisāʾ, and the
24 Gruendler
contemplation and enjoyment of pictures is now yaslamu ṣarʿatahū). This and the previous case of
ن
attributed to kings’ consorts instead, as in P3466 slightly awkward readings look like a copyist’s best
(li-takūna qulūbu nisāʾi ( )���س�ا ءl-mulūki amyalu guesses at a less than clear Vorlage.
ilayhi); P3473 further adds kings’ pages to the Instances of such adaptation and reinterpre-
ن
avid beholders of pictures (li-takūna qulūbu nisāʾi tation of the rasm occur regularly in Kalīla wa-
( )���س�ا ءl-mulūki wa-ghilmānihim ilayhi amyalu).41 Dimna, but some manuscripts, such as P5881 and
This description of the audience of illustrated L8751, contain significantly more cases than oth-
books deserves further study and comparison with ers, which may be a consequence of those copyists
reading practices contemporary with these manu- having either less legible Vorlagen or a freer atti-
scripts (fifteenth and seventeenth centuries).42 tude toward them.
In some cases, however, a rereading seems to veil
an earlier meaning. For instance, in the same pas- 5.4 Proliferation of the Text through
sage, which contrasts the true friend and the stra- the Recombination of Elements and
tegic friend (43), the designations for the former Syntactic Dilation
min ahli al-ʿilmi wa-l-faṣāḥa) in which the ignora- early Arabic narrative, which focuses on outward
mus then embarrasses himself. Elsewhere in this action), the characters’ inner feelings are spelled
manuscript (when compared with the earlier out, such as the cornered rat’s confusion (wa-
A4095), words and phrases are inserted in a way baqiya mutaḥayyiran fī amrihī wāqifan lā yadrī
that affects the syntax and changes the causal kayfa yaṣnaʿu,46 11), the cat’s joy at the proposal
connections. The impression is that of a growing (fa-fariḥa bi-dhālika faraḥan ʿaẓīman, 28), and,
mosaic. at the tale’s end, the rat’s fear for his soul (wa-ana
akhāfu ʿalā rūḥī minka wa-min ʿadāwati mā baynī
wa-baynaka, 71).
6 Substantial Additions and Trends The dialog is also enlivened with added changes
of Rewriting of speaker, and the characters’ direct speech is
expanded. In the rat’s monologue, he expresses
The last type of change, substantial addition, a hope that the cat might cooperate (laʿallahū …
is best shown throughout an entire chapter in yusallimu bi-dhālika li-ṣulḥī, 16) and decides
selected manuscripts, since it is the sum of edito- right from the start to be frank with his foe (qāla
rial interventions that need to be viewed in con- l-juradhu wa-llāhi lā akdhibuhū shayʾan mimmā
text. In the following discussion, the versions of the fī ḍamīrī, 19). In the fourth dialog, the rat is more
“Cat and Rat” chapter in those three manuscripts forceful and immediately interrupts the cat’s invi-
that contain the largest amount of rewriting (both tation with a refusal (qāla lahū l-juradhu lā sabīla
within shared units and through the addition of lī ʿalā l-dunuwwi minka, 60), before he explains
new units) are analyzed in detail. While the basic at length later on why he can no longer remain in
storyline remains intact and some formulation is the cat’s company. The cat similarly receives more
shared, the interventions identified develop the expressive lines; when he swears not to harm the
tale in each case in a different direction, and they rat, he invokes God as his witness (wa-ashhada ʿalā
do so consistently throughout this chapter. This nafsihī llāha subḥānahū wa-taʿālā ʿalā dhālika,
shows clearly that most of the changes were not the 63). This version accentuates the opposition
random products of a careless copyist, but rather between friend and foe. The friendship is recalled
deliberate acts of redaction.45 Even within this merely as a thing of the past (wa-hiya [al-ḥājatu]
short chapter, the visible interventions (whether llatī aḥdathat … hādhihī l-ṣadāqata wa-qad maḍā
cumulative or individual) can be shown to engage l-amr, 71) and then placed in an antithesis with
with the tale in an authorial manner. However, it enmity (taḥawwala ṣafāʾuhū ʿadāwatan, 69). The
cannot be ascertained whether this was the doing rat ends apodictically: never will he trust the cat
of the very copyist to whom we owe a particular again (fa-lā aṭmaʾinnu ilayka abadan, 73).
version, his Vorlage, or a combination of the two The rewriting of the maxims likewise brings
(i.e., individual or cumulative redaction) until all out the contrast between friend and foe: a virtu-
extant manuscripts have been compared. ous person has the moral duty to help if he is able
to do so (wa-laysa yajmalu dhālika bi-l-karīmi
6.1 P5881 – Emotion and Drama
The text portions that are unique to P5881 46 Bold words within a quotation are those unique to one
expand the chapter’s emotional and dramatur- manuscript (with the caveat expressed in note 35).
gical aspects. Unlike other versions (and unlike Words set in Roman style also appear in other manu-
scripts and are reused in P5881, even though at times
in a changed context. The numbers refer to the units in
45 For the argument that specific passages were selected which this occurs; bold numbers indicate entire units
for rewriting, whereas others remained more or less unique to the manuscript under discussion. For a list of
intact, see also sections 4 and 5.3 above. units, see tab. 1.1.
26 Gruendler
wa-lā yalīqu bihī dhālika wa-lā yanbaghī lahū 6.2 P3466 – Ethics and Power
an yuqaṣṣira fī ḥājati ṣāḥibihī in kāna muqtadiran This version focuses on ethics, relations of power,
ʿalayhi, 37). In another maxim, it is not the action and how the characters negotiate these in their
of treachery that is punishable (as elsewhere) interactions. The gravity of the situation is empha-
but the person of the enemy (wa- aʿjalu l-ʿuqūbati sized by the envisioning of failure.
ʿuqūbatu l-ʿaduwwi, 4047). Conversely, a person In his monologue, the rat dwells on his
who cannot keep a friend will be rejected by all his position of strength (lā yastaṭīʿu aḥadun an
friends (rafaḍūhu ahlu mawaddatihī, 62).48 The yukhalliṣahū ghayrī, 16) and sees the potential
distinction is clearest in the maxim bearing upon benefit of the cat as a means of increasing his
the diverse treatment of friend and foe in other own power (wa-laʿallahū … yaṭmaʿu fī maʿūnatī …
manuscripts. P5881 drops the part about the friend wa-yuṣāliḥunī li-nafʿi nafsihī, 16). This idea is then
and advises the manipulation of the foe alone broached directly to the cat: the rat alone has the
(wa-idhā khāfa l-ʿaduwwa abdā lahū l-ʿadāwa, 66), power to set him free (laysa aḥadun aqdara ʿalā
implying that what makes a friend is precisely that maʿūnatika wa-takhlīṣika min hādhihī l-warṭati
one need not calculate one’s behavior toward him. minnī, 20). In return, the cat must make a pact
The version concludes with a unique unit that in which he agrees not to kill the rat ( fa-in anta
returns to the frame dialog between the king and ammantani wa-ʿāhadtanī allā taqtulanī, 20). The
the philosopher; the latter comments that man rat stresses that only mutual help can save both
should heed the example of a weak, tiny rat who of them (kadhālika ayḍan bi-l-taʿāwuni minnī
foresaw the consequences of his actions, and saved wa-minka narjū l-khalāṣa min hādhihī l-baliyya,
himself by initiating and honoring an agreement 26), and thereafter an added sentence informs the
with his enemy. The philosopher ends with a direct reader of the pact’s conclusion (fa-taʿāqadā ʿalā
exhortation to the reader to apply this lesson (fa- dhālika, 28).
kayfa bi-l-insāni lawi qtaḍā bi-dhālika wa-ʿamila When the cat later demands the rat’s fulfillment
fīhi bi-lladhī yalzamuhū min ḥusni l-naẓari … of the pact by adducing a maxim, he implies that
yaʾkhudhu bi-awfā naṣībin minhu fī ʿājili l-umūri he has fulfilled his own part and that the rat has
wa-ājilihā fa-fham dhālika wa-dabbirhu turshad obtained his share of the deal ( fa-laysa hādhā
in shāʾa llāh, 83).49 l-karīmu50 bi-khalīqin an yatawānā fī ḥājati ṣāḥibihī
On the whole, the redaction of P5881 makes the idhā huwa qaḍā ḥājata nafsihī, 37). The cat fur-
fable more entertaining and applicable for regular ther applies a parallelism to his own and the rat’s
people, not only rulers or courtiers. The characters’ situation, reiterating the rat’s owed reciprocation
speech is livelier and their emotions expressed in a (kamā asraʿtu ana fī khalāṣika wa-maʿūnatika
way that lets the reader identify with them. It is a fa-kadhālika anta ḥaqīqun bi-l-ijtihādi fī khalāṣī
version that speaks to the situation of any human min hādhā l-withāq, 41). When the rat then
being and is not geared to the courtly realm or the explains that he must protect himself, he recalls
political arena. his situation of power, reminding the cat that he
never had a choice (mā huwa aljaʾanī ilā ṣulḥika
wa-aljāʾaka ilā qubūli dhālika minnī, 46).
Conversely, the weakness of the cat’s stance
47 Adapted rasm. is adumbrated with added narrative detail: at
48 Preposed verb agrees with plural subject, a feature of the hunter’s approach, he senses his imminent
Middle Arabic. death ( fa-lammā raʾāhu [sc. al-ṣayyāda] istashʿara
49 A similar final commentary appears at the end of the
chapter on the “King of the Mice” (Km), which appears
l-halakata, 53), and the oath he swears not to harm
in the older Syriac version but only in a few Arabic
manuscripts. 50 Misspelling of li-l-karīmi.
A Rat and Its Redactors 27
the rat in the future contains a protestation of sin- death. The message thus stays close to the book’s
cerity (wa-jtahada ʿalā ṣidqihī fīmā qāla, 63). original concerns of a Fürstenspiegel, but makes its
Cat and rat impersonate weakness and strength, teachings apply to ordinary people as well.
respectively. But in the maxims, the redaction gives
particular attention to the weaker party by fore- 6.3 L4044 – Character vs. Fate
grounding the ethical aspects. Thus, in a maxim In L4044, the fable is turned into a conflict
that condemns betrayal and the denial of mercy, a between fate and inborn character traits; here
second part is further specified: if the supplication the function of the animals as placeholders for
comes from a sufferer or sinner, and the addressee humans becomes particularly real. It is narrated
has the power to forgive but refuses to do so, he from the vantage point of protagonists who find
thereby violates a code, namely, the “ethics and themselves thrown into a struggle between des-
deeds of the righteous” (wa-man idhā taḍarraʿa tiny and their own characters. Furthermore, this
ilayhi l-malhūfu aw … al-mudhnibu … lam yaʿfu wa- version acquires a psychological dimension by
lam yaghfir wa-huwa yajidu ilā dhālika sabīlan creating tension between the protagonists’ inner
wa-innahū51 laysa bi-ḥāfiẓin akhlāqa l-ṣāliḥīna feelings and outward actions.
man lā yaʿmalu aʿmālahum, 40). This is an ethics In the monologue, the rat expresses his own
that supports the weak: it accords them rights and wishes and wonders how the cat will react to
makes the powerful responsible for upholding and his proposal, imagining the cat’s emotions and
protecting these. thoughts: “He will realize the benefit of what I
In addition to the behavior of an individual describe and ask him to do” (qāla fī nafsihī …
toward different types of friends, this version adds laʿallahū law qad samiʿa mā urīdu an ukallimahū
the sense of safety enjoyed with a true fried – in bihī … wa-lā mimmā kariha wa-lā taqrīʿan …
contrast to the constant fear created by a strategic wa-yaʿrifu manfaʿata mā aṣifu lahū wa-adʿūhu53
friend ( fa-ammā l-ṭāʾ iʿu … yuʾman ʿalā kulli ḥālin fa-laʿallahū yaṭmaʿu fī maʿūnatī iyyāhu fī khalāṣihī
wa-ammā l-muḍṭarru fa-inna lahū … ḥālāni52 fīmā yajnī, 16). Of the two characters, the rat is
yustarsalu ilayhi fīhā wa-ḥālāni yuttaqā fīhā, 43). given the greater emotional complexity. He asks
Here the concern is safety from other people, the cat to consider the situation and requests not
as opposed to any relationships among them, only a guarantee (as elsewhere) but also a solemn
as in P5881. In the maxim about losing a friend, oath to save the rat’s life in the process of saving his
friendship is highlighted by hendiadys (al-aṣdiqāʾ own ( fa-in anta ammantanī wa-ḥalafta wa-jaʿalta
wa-l-ikhāʾ, al-ikhwān wa-l-aṣdiqāʾ, 62), but friend- lī an tamnaʿanī54 mimmā tamnaʿa bihī nafsaka …
ship is made coextensive with usefulness in the qaṭaʿtu ḥabāʾilaka, 20).
subsequent parallelism (idhā zālati l-ḥājatu llatī The cat remains unidimensional. He believes
ḥamalathu ʿalā dhālika zālat ṣadāqatuhū, 69). the rat right away (qāla li-l-juradhi innaka ʿalā
The ideas foregrounded in this version are qawlika fawran lā ashukku fī ṣidqika ʿindī fīhi,
power and ethical behavior toward the weak, and 28) and, upon the rat’s explanation of the proce-
how both are negotiated. This redaction turns dure, readily agrees to the pact (qāla l-sinnawru
the fable into a tale of moral vs. political strategy afʿalu dhālika wa-niʿma, 31). When explaining his
in which mutual commitments are paramount; strategy of delay, the rat explains this by reference
ignoring them comes at the risk of censure or to his past and present fear of the cat (ana wāfin
laka … wa-muḥtarisun minka maʿa dhālika min
51 Pleonastic conjunction wa- before the protasis, a fea-
ture of Middle Arabic.
52 Pseudo-correction for the accusative singular ḥālan, a 53 The object ilayhi is missing.
feature of Middle Arabic. 54 Emendation of tamtaʿanna in L4044.
28 Gruendler
ḥaythu akhāfuka55 an yuṣībani minka mithlumā maqādīru bimā huwa ātin, 27). Here the oppos-
takhawwaftuhū, 46). This is a very different rat ing forces of fate and nature are weighed against
from the one flaunting his power in P3466. each other, and the rat must navigate between
As the hunter approaches, the cat expects to them. Prudence dictates that one must as quickly
die ( fa-lammā raʾāhu istaslama li-l-mawt, 53), as possible give up an unnatural friendship which
but the rat promptly announces and executes the one has been compelled by fate to make. Although
fulfillment of his part of the deal (wa-qāla hādhā this maxim appears earlier (75), the rat restates it
l-waqtu lladhī akhkhartu lahū qaṭʿa l-silki, 55; in L4044 in an unique added unit endowed with
wa-akhadha l-juradhu fī qaṭʿi l-silki, 56), show- scholarly authority:
ing himself to be reliable at the moment of the
cat’s despair. When the cat later calls the rat back, The learned say: The intelligent person
as the danger is past, he swears eternal gratitude must make peace with his powerful foe and
and specifically rules out any betrayal on his part bear humiliating himself before the foe for
for all future time ( fa-ʿinnaka mā baqītu fa56-lā a good reason but shun him whenever he
takhāfanna minnī ghadran, 63). However, this is can dispense with him and avoid trusting
exactly the kind of promise that a natural enemy him. (77)
cannot keep, according to the logic of this version.
The rat lays out the dangers of such duplicity in The rat, who is aware of this, distrusts the cat’s
a long, unique passage within the fourth dialog: “fancy talk” and expects nothing but evil from him
outward protestations that hide opposite feelings (wa-lā aẓunnu bika illā sharran fīmā tadʿūnī bihī
constitute enmity of the worst kind. Here the rat ilā nafsika min laṭīfi l-manṭiq, 73). Safety can only
introduces the concept of inborn enmity: be ensured by mutual distance:
king gives the gist of the situation for which he paratexts (incipit, table of contents, colophon,
wishes an exemplum (2).58 With its substantial corrections, marginalia, readers’ notes, owner-
additions, some of which have been cited above, ship marks, and any unrelated notes added at a
it is the most sophisticated version of the three: manuscript’s beginning or end); and other works
it introduces the psychological conflict between assembled with Kalīla wa-Dimna into a single
a person’s feelings and actions, and situates this binding (such as in L8751 and L4044). Illustrations
within the larger conflict between God’s decrees in particular take a life of their own when illumi-
and human nature, which the fable translates into nators add characters that do not appear in the
animal behavior. This rewriting changes the fable text, such as the bare-breasted women in bed with
from a tool for teaching political strategy into a the protagonist in the substory of the pauper win-
study of character in extraordinary circumstances, ning clothes from a thief (P5881, fol. 11v, chapt.
when the divine and natural orders collide. Im, unit 62). It is interesting to note that, in some
The microanalysis and particularly the survey manuscripts, spaces for illustrations have not been
of substantial additions have answered the second filled in (P3466 and P3478), or illustration legends
initial question regarding the unique and distinc- are recorded in the text but with no space left for
tive features of each manuscript. It is indeed fas- the illustrations themselves (P3473). The former
cinating in how many ways the copyist-redactors might be a copyist’s way of targeting both wealthy
manage to reinterpret the character of the rat in and less wealthy buyers – he would complete the
only three versions of this chapter – as emotional illustrations only if he was paid to do so. The latter
(P5881), powerful (P3466), or conflicted (L4044). may be due to a copyist’s omission of illustrations
It remains to be seen whether the emphases on not desired by a commissioning customer, while
drama, strategy, and psychology visible in this allowing their potential reinsertion in a subse-
chapter extend throughout these manuscripts in quent copy of the same manuscript; alternatively,
their entirety, and whether they give each version they may have been retained to act as side titles
of the book a consistent trend. Such a detailed to quickly locate a particular passage. A majority
study of the full manuscripts remains to be of images illustrate substories, which are thus easy
undertaken. to find. The fact that accessibility was a factor in
the manuscripts of Kalīla wa-Dimna is also evi-
dent in the thorough subdivision and highlighting
7 Linguistic and Codicological Aspects of the text in many of the manuscripts. Marked
are usually substories, wisdom sayings and their
The results of the preceding (partial) narratologi- elements, and the inquit formula (qāla) indicat-
cal analysis need to be placed in the context of ing a change of speaker or a return to the narra-
the whole manuscripts, both in their textual and tive frame. Hereafter, I will limit myself to a few
material aspects as well as in their combination remarks about the incipits and colophons of the
with other works in multiple text manuscripts. three manuscripts analyzed.
This comprehensive view can only be sketched out
here; it is also limited by the incomplete state of 7.1 P5881
preservation of some of the specimens. P5881 is written in narrow-spaced, small naskh
Indications about a version’s production and with a double frame around the text and color-
reception can be gathered from its manuscript’s ful illustrations. The chapter of “The Cat and the
physical characteristics: visual arrangement (page Rat” contains two illustrations, one in the first dia-
layout, sequence and topics of illustrations); log (between 18 and 19) and another in the fourth
and last dialog (between 59 and 60), so that both
58 For this internal repetition, see also section 5.4 above. images are placed immediately before a unit that
30 Gruendler
is unique to this version. The joint emphasis of needs to protect oneself. No unified moral stance
redactor and illustrator on these redacted pas- emerges; instead, the precepts are contrastive
sages points to their collaboration or perhaps to and complementary. The framing structure of the
their being the same person. book is pointed out as constituted by “340 interca-
The manuscript also contains an incipit before lated stories (uḥdūthatan mutadākhilatan baʿḍuhā
the first preface of the book (Lv). After the bas- fī baʿḍ).” The list of qualities reappears in the sec-
mala, subdivided by paragraph markers in the ond and longer table of contents that follows. Each
shape of red dots, two versions of a table of con- synopsis gives a chapter’s gist, general message,
tents follow. The first is a short enumeration for and particular addressees. Kings are targeted in
quick reference with chapter titles (given as six- only four of the sixteen chapters. Elsewhere, those
teen in total, preceded by two epistles59), which who should follow the lessons are “the intelligent”
are numbered serially and written in red ink. The (al-arīb, al-ʿāqil, dhawū l-ʿuqūl, dhawū l-albāb),
second is an expanded list of chapter titles with “the knowledgeable” (dhū l-maʿrifa, al-ʿārifūn), and
intercalated synopses of their content.60 Between “people of all trades” (ahlu kulli ṣanʿa). The chapter
both tables appears an enumeration of the book’s synopsis of “Cat and Rat” specifies no recipient at
benefits, defined as “330 types of wisdom (bāb min all and runs as follows:
al-ḥikma),” covering everything from correct and
prudent behavior toward kings to general human This is the parable (mathal) of a man with
virtues, including good deeds, present and future, many enemies who surrounded him from all
the proper management of wealth, the renuncia- sides so that he verged on perdition, and he
tion of worldly goods, and vices from which one sought rescue by befriending and conclud-
ing a truce with one of these enemies, and
59 The first epistle is Lv (fol. 3v), the second is not described he [thereby] extricated himself from what
in P5881 (nor in Cairo, al-Maktaba l-Markaziyya 1169, he had feared. He who forms a pact with
a nearly verbatim copy), but Istanbul, Ayasofya 4214,
which is another nearly verbatim copy of P5881, adds
his enemy must use caution and place his
a description of the second epistle as praise of King trust [carefully]. This chapter indicates [the
Anūshirvān by Buzurjmihr (fol. 1v,6–7). Such a praise of use of] strategy (iḥtiyāl) when catastrophes
a king by an advisor forms the conclusion of the book’s occur and proper planning, from the begin-
frame dialog after the last chapter (Ag) but it occurs
ning to the consequences.
between the Indian king Dabshalīm and his vizier
Baydabāʾ, and is not marked by a heading (fol. 119v,2).
Im and Bu are each counted as chapters, so that Lo, the The synopsis of the chapter of “The King and His
first parable, becomes the third chapter. Other manu- Dreams” (Kd) contains, in addition to the above
scripts do not count the prefaces (nor include them in points, a literary commentary on its wisdom say-
their tables of contents), listing and numbering only
the actual parables.
ings (taʾaddubun balīghun fī l-dunyā wa-l-dīn).
60 Other than the mentioned MSS similar to P5881, MS A one-line colophon records the date (1092 AH
Oxford, Bodleian Library, Pococke 400, dated to the [1681]) but no copyist’s name or place.
fourteenth century, contains a table of contents with The language displays moderate Middle Arabic
intercalated synopses and is so far the earliest example.
features (twelve instances within the chapter).
In the indirect transmission, such a table with synopses
is given by al-Yaʿqūbī (d. 286/897) in a chapter sequence These mostly represent cases of “relaxed gram-
resembling D, though some chapters are missing; mar,” such as morpho-syntactical errors, dropped
see al-Yaʿqūbī, Taʾrīkh al-Yaʿqūbī qui dicitur al-Jaʿqūbī particles, tautologies, and two pseudo-corrections,
Historiae, ed. M.T. Houtsma (Leiden: Brill, 1883), 98–99; but there is no admixture of dialect.61
and al-Yaʿqūbī, The Works of Ibn Wāḍiḥ al-Yaʿqūbī:
An English Translation, ed. Matthew S. Gordon,
Chase F. Robinson, Everett K. Rowson, and Michael 61 For a list of Middle Arabic features in popular Arabic
Fishbein (Leiden: Brill, 2018), vol. 2, 351–52. literature, see Bruce Fudge’s introduction to A Hundred
A Rat and Its Redactors 31
This version, besides being lavishly produced, is chapter of “The Lion and the Ox” (corresponding
accessible and its contents broadly applicable; it to Lo), which is the beginning of the actual book.
targets a wide readership and highlights the book’s Thereafter follows the preface written by ʿAlī b.
literary qualities. The existence of two nearly ver- Shāh (As).
batim copies62 is noteworthy and infrequent The table of contents is not given within the
among manuscripts of Kalīla wa-Dimna. incipit (as it is in P5881) but inserted into As at the
place where Baydabāʾ’s composition of the origi-
7.2 P3466 nal Indian version is narrated. Fourteen headings
This manuscript may have been produced with are announced and listed, though the manuscript
aspirations for an elite market, but its illustrations actually contains fifteen, because Km (located
were never filled in (only the legends are written between Kd and Mc) is omitted from the table of
vertically in the margin next to the blanks); per- contents. Baydabāʾ’s procedure in the book’s cre-
haps the eventual buyer was not willing or able to ation is described as ambitious and perfectionist:
pay the added expense. Unfilled illustrations occur
in a number of manuscripts and may be an indica- Then he began to compose the book (thumma
tion that a more limited upscale market had been btadaʾa fī naẓmi al-kitābi wa-taṣnīfihī) with-
widened to include a less well-to-do clientele. out interruption. He dictated and his disciple
The basmala and subsequent text, written in wrote, and he altered and revised until the
a wide-spaced, artless, but legible naskh, is sub- book turned out utterly perfect, contain-
divided with paragraph markers in the form of ing fourteen chapters, each of them self-
red dots, circles, or three inverted apostrophes, contained, and he included himself and
and inquit formulae occasionally receive a red the king and the answer[s] given to him as
overstrike. part of the book, so that the king would be
The incipit is rather brief; it motivates the book’s directly addressed. (pp. 25,ult.–26,4)
composition and then enumerates the reasons for
adding each of the prefaces. First Baydabāʾ com- The parable form and its animal characters are
posed the book overtly as a collection of animal explained twice, before and after the list of chap-
tales in order to reserve its hidden wisdom for ters. Before the list, the book’s double-layered
philosophers, who are able to decode it. Then meaning, overt and hidden (ẓāhir/bāṭin), and its
Anūshirwān sent Burzoy to bring the book from dual function of giving pleasure and conveying
India, which he did together with the large chess wisdom (lahw/ḥikma) are motivated as being a
game (with 10 × 10 fields; this corresponds to Lv). code for the elite (khāṣṣa) and the wise (ḥukamāʾ)
Thereafter Buzurjmihr added a preface on the and a convention of wisdom books (ka-rasmi
book’s benefits63 and another on Burzoy himself sāʾiri l-kutubi llatī rusima li-l-ḥikam), and its con-
(corresponding to Bu), placing it before the tent described as everything needed for the here
and the hereafter, for serving kings and counseling
and One Nights, ed. and trans. Bruce Fudge, (New York: them, and for avoiding what should be avoided.
New York University Press, 2016), xxxvi–xxxvii; and
The table of contents itself (pp. 26,10–28,1) is
Jérôme Lentin, “Middle Arabic,” in Encyclopedia of
Arabic Language and Linguistics, ed. Kees Versteegh highlighted in a larger red script and the chapter
(Leiden: Brill, 2006–9), vol. 3, 215–24. headings are formatted as a centered column with
62 See note 16. intercalated synopses supplying the gist of each
63 This is does not correspond to a separately titled chapter.
entity in the manuscript but rather describes Ibn
al-Muqaffaʿ’s preface (Im). A similar reattribution
occurs in P5881, where the initial part of Im is moved scene in which Buzurmihr is commissioned to com-
to the “Long Voyage of Burzoy” (Lv) following the final pose the “Life of Burzoy.”
32 Gruendler
The language contains moderate Middle Arabic fits with the idiosyncratic nature of the redac-
features (sixteen instances within the chap- tion and the spacious mise en page, with its large
ter), of the same type as in P5881 but with six pages and wide margins. Some aspects show great
pseudo-corrections. care. Introductory formulae are often highlighted
The book shows ample traces of use (and in a larger pen, and substories are given their own
reuse) at the end (pp. 341–44). These include read- titles in larger script within the running text. Select
ing marks, one them containing the date 854 AH words are highlighted with a large pen, such as the
[1450 CE], which is a post quem, and unrelated imperative “be aware” (iʿlam, 76), which precedes
texts, such as two recipes for falūniya,64 a panacea, one of the redaction’s unique passages (77), as if to
a notice about the birth of a son, and writing exer- point out what the copyist held to be particularly
cises of Qurʾān verses. This version, too, obviously important. This presentation indeed suggests an
circulated among a non-elite readership. educated client of some social status.
The large page format with its lavish images,
7.3 L4044 some on double pages, confirms the appearance
The case of L4044 is a puzzle because the manu- of a luxury copy. However, most of the images are
script’s different aspects seem to contradict each placed in the margins or the empty parts of title
other. The beginning, with its potential incipit, and pages and might potentially be later additions.66
the end are missing. The preserved part begins in However, the writing, a horizontally stretched,
the middle of the long version of “Burzoy’s Voyage” wide-spaced, and artless naskh, falls between two
(Lv), with another lacuna of several pages in the extremes: prose passages omit most diacritical
same section, and the text of Kalīla wa-Dimna dots and all vocalization, whereas verse passages
ends without a colophon, since it is followed by (in the following work, the Sulwān al-muṭāʿ), are
Sulwān al-muṭāʿ by Ibn Ẓafar al-Ṣiqillī (d. 565/1169 fully vocalized. Did the scribe target an educated
or 568/1172), which in turn lacks its last pages. This reader who could dispense with such added
twelfth-century work is an emulation of Kalīla detail? The text’s language contains the same
wa-Dimna, composed for a Sicilian military offi- moderate amount of Middle Arabic features as the
cial, and the combination of both works speaks two other manuscripts (twelve instances within
to the specific interest of the copyist or his client the “Cat and Rat” chapter, including six pseudo-
in the genre.65 A personal commission further corrections) and is not conclusive.67 In terms of
64 From Greek philomena, named after its inventor Philon Paris BnF Arabe 2789, the popular tale of Kalʿād and
of Tarsus, a contemporary of Augustus, and preserved Shimās in Riyadh 2407, and Alf layla wa-layla in Paris
in versified form by Galen; see Dozy, Supplément aux BnF Arabe 3612. Merged with Kalīla and Dimna is, for
dictionnaires arabes, s.v. f-l-y, and Der Kleine Pauly: instance, al-Tanūkhī’s al-Faraj baʿda l-shidda in Beirut
Lexikon der Antike in fünf Bänden (Munich: Deutscher USJ 00022.
Taschenbuch Verlag, 1979 (Munich, 1979), vol. 4, 776b, 66 I am grateful for this suggestion to Annie Vernay-Nouri.
s.v. Philon. 67 This moderate degree characterizes the majority of
65 The combination of Kalīla and Dimna with other works, manuscripts of Kalīla wa-Dimna so far inspected and
either subsequent in a multiple text manuscript or must be taken as its own idiom. Thus, it is comparable
merged into one, requires a separate investigation. The to the lughat al-ḥikāya or lugha thālitha, defined by
diversity of such texts is remarkable: combined with Muhsin S. Mahdi based on the earliest manuscript of
it is, e.g., an emulation in verse, al-Ṣādiḥ wa-l-bāghim, Alf layla wa-layla; see The Thousand and One Nights
by Ibn al-Habbāriyya in L8571, an emulation in prose, (Alf Layla wa-layla): The Classic Edition (1984–1994),
Sulwān al-muṭāʿ, by Muḥammad b. ʿAbdallāh b. Ẓafar ed. M. S. Mahdi (Leiden; Boston: Brill, 2014), vol. 1, 45
al-Ṣiqillī in L4044, Pseudo-Ṣafadī’s Lawʿat al-shākī and 47. It remains to be seen whether the fact that this
in London BL Add. 23466, Ibn ʿAbd al-ʿĪd al-Mālikī’s book belonged to the fable genre made a less stringent
Ḥilyat al-kubarāʾ wa-bahjat al-nudamāʾ in Paris BnF adherence to the ʿarabiyya acceptable. An example of
Arabe 3476, al-Damīrī’s, Ḥayāt al-ḥayawān al-kubrā in the opposite case, a manuscript in perfect ʿarabiyya,
A Rat and Its Redactors 33
content, the version displays a careful reworking (less frequent).72 From this textual flux, individual
not only of the psychological aspects in this chap- voices of anonymous copyist-redactors can be
ter, as shown above, but also of the epistemology isolated by comparison with other manuscripts.
of reading in Ibn al-Muqaffaʿ’s preface (Im). Other These distinguish themselves by the appearance
luxurious and illustrated versions usually mark of consistent trends in their instances of rewriting
diacritics and vowels, whereas it is the simpler and (reformulation, cuts, and additions), as shown in
less carefully executed versions with substandard the three examples above. This answers in part the
Arabic features that lack vocalizations and skip two preliminary questions posed at the beginning.
some diacritics. The careless script of L4044 belies But there is little more we can know about
the quality of the redaction. Possibly the present the redactors of the three diverging manuscripts
manuscript is a reproduction of an earlier, more of Kalīla wa-Dimna examined in more detail.
carefully executed Vorlage or an autograph by an It appears that they were produced in different
educated writer. social contexts, L4044 in a more educated one
than P5881 and P3466. But P5881 and L4044 were
costly illustrated exemplars, whereas P3466 circu-
8 Envoi lated in its cheaper, unfinished state and shows
ample signs of use with numerous manuscript
In summary, Kalīla wa-Dimna reveals itself to notes. Nonetheless, in all three cases, the copyists
form a fluctuating manuscript tradition in which leave their individual imprint on a chapter, fore-
no groups can be detected; rather, manuscripts grounding, respectively, drama; strategy and fair-
resemble one another (in the form of shared plot ness; and fate and psychology. The copyists treated
segments, as shown in tab. 1.1 and 1.3, and partial the work as an opportunity to act as redactors,
and non-contiguous places of overlapping formu- and they gave their versions an individual bent.
lation, as in tab. 1.4), forming continua (between But why did they not care, or dare, to declare this
which some manuscripts shift); the use of more explicitly? This is the place to return to the second
than one Vorlage can be ascertained in at least set of broader questions posed above, which we
three cases.68 Among the manuscripts, some are may now refine.
strongly inclusive,69 assembling passages that Did the copyists not wish their interference to
appear scattered across different manuscripts, be known and rather preferred to pass off their
whereas others selectively abridge70 (as shown in opinions as those of Ibn al-Muqaffaʿ and his
tab. 1.2). In both types, rewriting occurs. In terms sources? The phenomenon of anonymity as a type
of the resemblance across the manuscripts, this of authorship and textual authority still remains to
ranges from drastic difference (occasional) via sim- be investigated for premodern literature in general
ilarity of the macrostructure with changes within and Arabic literature in particular. Arabic pseude-
units (most frequent),71 to near verbatim identity pigraphica are frequently attributed to Ancient
Greek or legendary authors, especially in hermetic
is MS BRR 3655. To assign this work to a genre is dif- and magical texts. The anonymity of specimens of
ficult, and premodern authors taxonomized it in vari- Kalīla wa-Dimna falls into several categories. From
ous ways; see Mathew L. Keegan in Gruendler et al.,
truncated manuscripts, where the beginning and
“An Interim Report on the Editorial and Analytical
Work of the AnonymClassic Project.” Medieval Worlds end are missing, nothing may be concluded. Those
11 (2020), 241–79, esp. 269–70 and in more detail, idem, with colophons either lack the name of the copy-
“Its Meaning Lies Elsewhere: The Vagaries of Kalila and ist, or if mentioned, only acknowledge the mere
Dimna.” Poetica 52 (2021): 13–40.
68 See note 32.
69 E.g., P3466, L4044, and Wetzstein II 672.
70 E.g., P3465 and BRR 3566. 72 E.g., between P5881, Ayasofya 4214, and Cairo, al-
71 E.g., P3473 and Wetzstein II 672. Maktaba al-Markaziyya 1169. For a fuller list, see note 16.
34 Gruendler
act of copying.73 Keeping silent about one’s inter- in which the phases cannot so far be assigned to
ventions may have translated into a higher market any individual agents.74
value, since the book was a popular classic. Even Or again, had the interference in such a text,
in later times, when the linguistic register and which conveyed wisdom in the fictional guise
material execution of the manuscripts indicate a of fables and thus drew close to popular tales,
readership below the elite, this group continued to become so common that such a free treatment
consider reading Kalīla wa-Dimna an act of acquir- of the wording was self-evident and required no
ing wisdom, ethics, and “philosophy.” (This last comment? Was the authorship of such texts sim-
term increases in frequency with the progression ply of no great relevance to their readers and
of time and, paradoxically, is concomitant with (overt) originality out of place? That said, Kalīla
the relaxation of the grammatical norms and the wa-Dimna is sui generis in its double nature as
increase in pseudo-corrections and dialectal influ- a book of wisdom and a collection of fables. It
ence.) Did changes in the text not matter or, rather, had no fixed place in the generic grid of classical
was it in the interest of those less educated readers Arabic literature. At its inception the component
to pass over the instability of the work in silence, of “applied culture” (adab) dominated, and, dur-
because it allowed them to share in an ideal of ing the book’s “dark” phase up to the thirteenth
education they could not otherwise reach? Did century and the emergence of full-text manu-
Kalīla wa-Dimna serve as a token book of wisdom scripts, quotations from Kalīla wa-Dimna in works
for amateurs who were keen to uphold a learned of adab mainly concern ḥikam, short analogical
status for it (one that it had de facto long lost) and images (amthāl), and sections of dialog bear-
in the process claim the same for themselves as ing upon ethics and political strategy. The fable
its users? aspect functioned as a cover and a literary device;
Or did they purposely veil what they altered out al-Yamanī, in his counter-text to Ibn al-Muqaffaʿ,
of respect for the work? Was it perhaps a stance denigrated this aspect as “stuffing” and cut out all
of modesty on the part of craftsmen who did not narrative passages. From the time of the book’s
consider it fitting to call themselves (co-)authors, fully documented state in the thirteenth century,
or did not dare to declare themselves as such? the fictional character dominated, as is visible in
The medieval translators of Kalīla wa-Dimna had the frequency and choice of illustrations, which
no such qualms, nor did the later versifiers of the mostly depict enframed tales. Its language, with a
Arabic version, both of whom might even substi- more or less pronounced presence of substandard
tute their names for Ibn al-Muqaffaʿ’s, and one features (Middle Arabic), brings it closer to popu-
might argue that the interventions of all three lar literature.75 As time progressed, an opposite
groups varied only in degree, not in kind; the trend surfaced in some manuscripts, beginning in
Arabic version underwent a change in the linguis- the fifteenth century, namely a refocusing on the
tic register, too, since the Arabic copyist-redactors
performed an intra-lingual translation, even 74 An intriguing case is the copyist of London BL 3900,
though this is visible only as a cumulative process who claims authorship through ingenious cross-
copying. He merges his incipit with the preface of Ibn
al-Muqaffaʿ, whose title and author he suppresses,
so that the first-person voice of the latter becomes
73 An exception is MS Beirut USJ 00022, which merges his own. To achieve this, he switches between two
passages of al-Tanūkhī’s al-Faraj baʿda l-shidda into the Vorlagen, taking the incipit from Riyadh 2536, itself
text, and the copyist declared in the colophon: “Here close to Istanbul, TSMK, Ahmet III 3015, an augmented
ends what we intended to assemble (aradnā jamʿahū) retranslation from the Persian version by Munshī, and
in this book in terms of important stories and transmit- the Im chapter from Oxford, Pococke 400.
ted tales, so that it may let us attain important stations” 75 A few manuscripts, however, remain in classical Arabic,
(fol. 191r). such as P3465 and BRR 3655.
A Rat and Its Redactors 35
figure 1.1 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 5881, fol. 2v. Incipit with short table of contents
36 Gruendler
figure 1.2 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 5881, fol. 79v. Beginning of the chapter
of “The Cat and the Rat”
A Rat and Its Redactors
figure 1.3 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3466, pp. 26–27. Beginning of the table of contents including the synopsis of the sixth chapter,
37
“The Cat and the Rat,” within the preface of ʿAlī b. Shāh
38
figure 1.4 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3466, pp. 273–74. Beginning of the chapter of “The Cat and the Rat” with unfilled blank for
an illustration
Gruendler
A Rat and Its Redactors 39
figure 1.5 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, London, British Library, Or. 4044, fol. 97v. Beginning of the
chapter of “The Cat and the Rat”
40 Gruendler
book’s ethical and “philosophical” purport.76 This Paris, Bibliothèque nationale de France, Arabe 2789.
Janus-headed book was flipped back to its original Paris, Bibliothèque nationale de France, Arabe 3465.
focus, yet the social context and reading public Paris, Bibliothèque nationale de France, Arabe 3466.
had expanded into less educated strata and these Paris, Bibliothèque nationale de France, Arabe 3473.
aspects are still in need of further investigation. Paris, Bibliothèque nationale de France, Arabe 3475.
This initial foray can only raise such and other Paris, Bibliothèque nationale de France, Arabe 3476.
questions. Full answers need to await a fuller sur- Paris, Bibliothèque nationale de France, Arabe 3478.
vey of the ever-surprising textual history of Kalīla Paris, Bibliothèque nationale de France, Arabe 3593.
wa-Dimna, including the conditions of its produc- Paris, Bibliothèque nationale de France, Arabe 3612.
tion and reception and the work’s place in the Paris, Bibliothèque nationale de France, Arabe 5881.
shifting landscape of classical and popular Arabic Rabat, al-Maktaba al-Malikiyya (BRR) 3655.
literature (whose boundaries it muddles) between Riyadh, King Fayṣal Center 2407.
the thirteenth and nineteenth centuries. What Riyadh, King Fayṣal Center 2536.
is clear, however, is that “classical” and “popular” Tunis, Bibliothèque nationale de Tunisie, 2281.
should be seen as endpoints of a spectrum with
many shades in between, which Kalīla wa-Dimna, Primary Sources
as an unruly classic and a text on the move, helps Anonymous. A Hundred and One Nights. Edited and
to throw light upon. translated by Bruce Fudge. New York: New York
University Press, 2016.
Anonymous. The Thousand and One Nights (Alf layla
Bibliography wa-layla): The Classic Edition (1984–1994). Edited by
Muhsin S. Mahdi. Leiden: Brill, 2014.
Manuscripts Cheikho, Louis. La version arabe de Kalīla et Dimnah ou
Berlin, Staatsbibliothek, Wetzstein II 672. Les fables de Bidpai. Beirut: Imprimerie catholique,
Beirut, Université Saint Joseph (USJ), 0022(2). 1905; reprint Amsterdam: Academic Publishers
Cairo, al-Maktaba al-Markaziyya li-l-Makhṭūtāt Associated – Philo Press, 1981.
al-Islāmiyya 1169. Gaulmin, Gilbert. Le livre des lumières ou la conduite
Chicago, Oriental Institute, A 11991. des rois, composé par le sage Pilpay indien, traduit
Istanbul, Süleymaniye Kütüphanesi Ayasofya 4095. en français par David Sahib d’Isfahan. Paris: Simeon
Istanbul, Süleymaniye Kütüphanesi Ayasofya 4213. Piget, 1644.
Istanbul, Süleymaniye Kütüphanesi Ayasofya 4214. La Fontaine, Jean de. Fables, préface de Jean-Charles
Istanbul, Topkapı Sarayı Müzesi Kütüphanesi, Ahmet Darmon, dossier et notes par Jean-Charles Darmon
III 3015. et Sabine Gruffat. Paris: Librairie générale française,
London, British Library, Or. 8571. 2002.
London, British Library, Or. 4044. Ibn al-Muqaffaʿ. Kalila wa-Dimna. Edited by Ṭ. Ḥusayn
London, British Library, Or. 3900. and ʿA. ʿAzzām. Cairo: Dar al-Maʿārif, 1941.
Montreal, McGill University Library 117953 (Mc Lennan Ibn al-Muqaffaʿ. Le livre de Kalila et Dimna : Fables
no. 94). de Bidpaï. Translated by André Miquel. Paris:
Munich, Bayerische Staatsbibliothek, Cod. arab. 615. C. Klincksieck, 1957; reprinted 1980.
Munich, Bayerische Staatsbibliothek, Cod. arab. 616. Schulthess, Friedrich. Kalīla und Dimna. Die alt-
New York, MET 1981.373. syrische Version des indischen Fürstenspiegels
Oxford, Bodleian Library, Pococke 400. (Pantschatantra) oder Bidpai’s Fabeln, syrischer Text,
neu herausgegeben und übersetzt, mit einer Einleitung
76 E.g., in L4044 and Wetzstein II 672. und kritischen Anmerkungen und Varianten. Berlin,
A Rat and Its Redactors 41
1911; reprinted with Nöldeke’s recension (1911) and Cerquiglini, Bernard. L’éloge de la variante : histoire cri-
Burzoy’s preface (1912), Amsterdam: Academic tique de la philologie. Paris: Seuil, 1989.
Publishers Associated – Philo Press, 1982; reprinted Cooperson, Michael. “Ibn al-Muqaffaʿ.” In Arabic Literary
Berlin: De Gruyter, 2011. Culture, Dictionary of Literary Biography, vol. 311, edited
al-Yamanī, Abū ʿAbdallāh Muḥammad b. al-Ḥusayn by Michael Cooperson and Shawkat M. Toorawa,
b. ʿUmar. Muḍāhāt amthāl Kalīla wa-Dimna bimā 150–63. London: Thomson Gale, 2005.
ashbahahā min ashʿār al-ʿarab. Edited by M. Yūsuf Dagenais, John. The Ethics of Reading in Manuscript
Najm. Beirut: Dār al-Thaqāfa, 1961. Culture: Glossing the Libro de buen amor. Princeton,
al-Yaʿqūbī. Taʾrīkh al-Yaʿqūbī qui dicitur al-Jaʿqūbī N J : Princeton University Press, 1994.
Historiae. Edited by M.T. Houtsma. Leiden: Brill, Der Kleine Pauly: Lexikon der Antike in fünf Bänden, 5
1883. vols. Munich: Deutscher Taschenbuch Verlag, 1979.
al-Yaʿqūbī. The Works of Ibn Wāḍiḥ al-Yaʿqūbī: An Dozy, Reinhart. Supplément aux dictionnaires arabes.
English Translation. Edited by Matthew S. Gordon, Leiden: Brill, 1881; reprint Beirut: Librairie du Liban,
Chase F. Robinson, Everett K. Rowson, and Michael 1991.
Fishbein. Leiden: Brill, 2018. Gruendler, Beatrice. “Les versions de Kalīla wa-Dimna :
une transmission et une circulation mouvantes.”
In Énoncés sapientiels et littérature exemplaire :
Secondary Literature une intertextualité complexe, edited by Marie-Sol
Akel, Ibrahim. “Aḥmad ar-Rabbāṭ : sa bibliothèque Ortola and Marie-Christine Bornes-Varol, 385–416.
et son rôle dans la réception, diffusion et enrichis- Nancy: Presses Universitaires de Nancy – Éditions
sement des Mille et une nuits.” Doctoral thesis, Universitaires de Lorraine, 2013.
Université Sorbonne Paris IV, 2016. Gruendler, Beatrice. “Stability and Change in Arabic
Basset, René. “Notice sur un manuscrit des fables du Script.” In The Shape of Script: How and Why Writing
Kalilah et Dimna.” Journal asiatique 9, no. 16 (1900) : Systems Change, edited by Stephen D. Houston,
360–69. 93–118. Santa Fe, N M : School for Advanced
Bellino, Francesca. “Animal Fables in the Sulwān al-muṭāʿ Research, 2012.
by Ibn Ẓafar al-Ṣiqillī.” In Islamic Sicily: Philological Gruendler et al., “An Interim Report on the Editorial
and Literary Essays, edited by Mirella Cassarino, and Analytical Work of the AnonymClassic Project.”
Quaderni di Studi Arabi, no. 10 (2015): 103–22. Medieval Worlds 11 (2020), 241–79.
Biardeau, Madeleine. Le Mahabharata. Paris: Seuil, Hanna, Nelly. In Praise of Books: A Cultural History of
2002. Cairo’s Middle Class, Sixteenth to Eighteenth Century.
Bloch, R. Howard, Alison Calhoun, Jacqueline Cerquiglini- Syracuse, N Y : Syracuse University Press, 2003.
Toulet, Joachim Küpper, and Jeanette Patterson, eds. Hirschler, Konrad. The Written Word in the Medieval
Rethinking the New Medievalism. Baltimore, M D : Arabic Lands: A Social and Cultural History of
Johns Hopkins University Press, 2014. Reading Practice. Edinburgh: Edinburgh University
de Blois, François. Burzōy’s Voyage to India and the Press, 2012.
Origin of the Book of Kalīlah Wa Dimnah. London: Irwin, Robert. “The Arabic Beast Fable.” Journal of the
The Royal Asiatic Society, 1990. Warburg and Courtauld Institutes 55 (1992): 36–50.
Breasted, James H. “The Oriental Institute in Chicago – Keegan, Matthew L., “Its Meaning Lies Elsewhere: The
a Beginning and a Program.” American Journal of Vagaries of Kalila and Dimna.” Poetica 52 (2021):
Semitic Languages and Literatures 38, no. 4 (1922): 13-40.
233–328. Kristó Nagy, István T. La pensée d’Ibn al-Muqaffaʿ : un
Brockelmann, Carl. “Kalīla wa-Dimna.” In Encyclopaedia “agent double” dans le monde persan et arabe. Paris:
of Islam, second edition, edited by P. Bearman et al., Éditions de Paris, 2013.
vol. 4, 503–6. Leiden: Brill, 1960–2007.
42 Gruendler
Lentin, Jérôme. “Middle Arabic.” In Encyclopedia of trans.].” Zeitschrift der Deutschen Morgenländischen
Arabic Language and Linguistics, edited by Kees Gesellschaft 65 (1911): 578–88; reprinted in Schulthess
Versteegh, vol. 3, 215–24. Leiden: Brill, 2006–9. 1982.
Liebrenz, Boris. “The Library of Aḥmad al-Rabbāṭ: Nöldeke, Theodor. Burzôes Einleitung zu dem Buche
Books and Their Audiences in 12th to 13th/18th to Kalīla wa Dimna, übersetzt und erläutert. Strasbourg:
19th Century Syria.” In Marginal Perspectives on Early Karl J. Trübner, 1912; reprinted in Schulthess, 1982.
Modern Ottoman Culture: Missionaries, Travelers, O’Kane, Bernard. Early Persian Painting: Kalila and
Booksellers, edited by Ralf Elger and Ute Pietruschka, Dimna Manuscripts of the Late Fourteenth Century.
17–59. Halle: Zentrum für Interdisziplinäre Regional- London: I.B. Tauris, 2003.
studien Vorderer Orient, Afrika, Asien der Martin- Ruymbeke, Christine van. Kashefi’s Anvar-e Sohaili:
Luther-Univ. Halle-Wittenberg, 2013. Rewriting Kalila and Dimna in Timurid Herat. Leiden:
Marzolph, Ulrich, and Heinz and Sophia Grotzfeld. Brill, 2016.
“Kalīla und Dimna.” Enzyklopädie des Märchens, vol. Sacks, Jeffrey. “The Philological Present: Reading
7 (1993): 888–95. the Arabic Nineteenth Century. Review Essay of
Nichols, Stephen, ed. “The New Philology.” Speculum Four Books.” Journal of Arabic Literature 47 (2016):
special issue, 65, no. 1 (1990). 169–207.
Nöldeke, Theodor. “Die Erzählung vom Mäusekönig Sajdi, Dana. The Barber of Damascus: Nouveau Literacy
und seinen Ministern. Ein Abschnitt der Pehlewî- in the Eighteenth-Century Ottoman Levant. Stanford,
Bearbeitung des altindischen Fürstenspiegels.” C A : Stanford University Press, 2013.
Abhandlungen der Deutschen Morgenländischen Sprengling, Martin. “Kalīla Studies.” American Journal of
Gesellschaft 30 (1876): 752–72. Semitic Languages and Literatures 40 (1924): 81–97.
Nöldeke, Theodor. “Zu Kalīla wa Dimna.” Zeitschrift der Winternitz, Moriz. Geschichte der indischen Literatur.
Deutschen Morgenländischen Gesellschaft 59 (1905): Leipzig: Amelangs Verlag, 1909.
794–806; reprinted in Cheikho, 1981. Zumthor, Paul. Essai de poétique médiévale. Paris:
Nöldeke, Theodor. “Kalīla und Dimna, syrisch und Éditions du Seuil, 2000.
deutsch: Eine Rezension [of Schulthess’s ed. and
2
The Donkey’s Ears and Heart: Reshaping, Rewriting, Adapting,
and Enlightening the Kalīla wa-Dimna Text in Two Medieval
Persian Versions
Christine van Ruymbeke
The Book of Kalīla wa-Dimna (henceforth, KD), a with the KD text are nevertheless clear. This is so
pollinating bestseller with a rich history of rewrit- from the moment the text really begins to stand
ing, translation, and adaptation, is related in sev- out in history, in the form of the translation into
eral fundamental ways to the medieval Persian Arabic of a mysterious Pahlavi text, produced by
literary and political scenes.1 Though mostly over- Ibn al-Muqaffaʿ (d. c. AD 760), a Persian member of
shadowed by the scholarly focus on its textual the Umayyad, and later the ʿAbbasid, administra-
origin, which is traditionally believed to be Indian tion.3 The ties between the KD rewritings and the
(see below: this is not as straightforward as is usu- Persian literary world continue through the medi-
ally accepted),2 and by the undeniable importance eval and early-modern periods: the remarkable
of the Arabic version as the hypotext for all sub- proliferation of Persian versions through the cen-
sequent rewritings and translations, Persian ties turies attests to the text’s recurring attraction for a
courtly, Persian-speaking audience. It also shows
the text’s fascination for what appears to have
1 The present study is based on, and continues, the findings been a Persianized, research-active, courtly intel-
I have published in several research articles and in the
monograph centered on Kāshifī’s Anvār-i Suhaylī. I men-
ligentsia.4 Apart from these full versions, the KD
tion here in condensed form several crucial findings that
are explained at length in the most recent publications: 3 It is a questionable practice to base a literary analysis on
Christine van Ruymbeke, Kāshefi’s Anvār-e Sohayli: Rewrit- the author’s biographical data, especially when the avail-
ing Kalila and Dimna in Timurid Herat (Leiden: Brill, 2016); able data is so scant and likely to have been doctored, or at
Christine van Ruymbeke, “Authorship, Ownership and the very least amplified, over the centuries. However, some
Rewriting: Vāʿiz Kāshifī and Abū’l-Fazl b. Mubārak within elements that appear reliable can also be relevant: in the
the Hereditary Line of Kalīla wa-Dimna authors,” Jerusa- present case, I make much of the claim that the author is
lem Studies in Arabic and Islam 45 (2018): 181–210; Chris- of Iranian origin. The relevance of this claim goes beyond
tine van Ruymbeke, “Kalila and Dimna as a Case-Study: bilingualism: it affirms a cultural identity that spills over
Ibn al-Muqaffaʿ’s and Nasrullah Munshi’s Translations,” in into the influential works which Ibn al-Muqaffaʿ produced
The Routledge Handbook of Arabic Translation, ed. Sameh for his cultured Arabic-speaking audience.
Hanna, Hanem El-Farahaty, and Abdel-Wahab Khalifa 4 The Persian KD field cries out for extensive analysis. We
(London: Routledge, 2020), 253–268. are aware of the following Persian verse, prose, or prosi-
2 The bibliography around the origins of the KD is enormous metric KD versions: a lost Sāmānid prose text (written by
but often repetitive. I cannot examine it, even cursorily, or for the vizier Abū al-Faḍl Balʿamī [d. c. 936–40], father
within the boundaries of this research article. Rather than of the historian); a lost contemporary Sāmānid masnavī by
providing a critical review of existing scholarship, I hope the poet Rūdakī (only a few stray verses of this masnavī
here to suggest new attitudes towards the KD based on the have been discovered in Abū Manṣūr Aḥmad ibn ʿAlī Asadī
textual evidence provided by its versions. Van Ruymbeke, Ṭūsī, Asadī’s neupersisches Wörterbuch Lughat-i Furs: nach
Kāshefi’s Anvār, xi and 25–31, suggests reasons to revisit der einzigen vaticanischen Handschrift, ed. Paul Horn
this traditional heredity. I also argue that once we are lib- (Berlin: Weidmannsche Buchhandlung, 1897); the twelfth-
erated from the focus on a possible Indian and Sanskrit century bilingual Arab–Persian prosimetrum by Naṣrallāh
origin, clues relate the text’s roots to the Persian world and Munshī (used in the present essay), on which the Anvār-i
link it with Classical Antiquity. Suhaylī is based; a fifteenth-century prosimetrum by Vāʿiz
text also had an impact on a medley of other writ- hundred years later, in the last years of the fif-
ings, literary, philosophical, and mystical, in both teenth century, another famous Persian prosimet-
Arabic and Persian. The present chapter focuses ric version was produced in Herat at the culturally
on the Arabic text by Ibn al-Muqaffaʿ (henceforth, vibrant Timurid court of Sulṭān-Ḥusayn Bayqarā
IAM KD) and on two complete Persian versions. (r. 1469–1506). The author, Vāʿiẓ Kāshifī (d. 1504 or
Both are believed to be part of a direct hereditary 1505), is known to have been a preacher and reli-
line sprouting from this IAM KD. gious figure with a specialized interest in scientific
During the first half of the twelfth century, five topics. The filiation of this version, entitled Anvār-i
centuries after Ibn al-Muqaffaʿ’s canonical but Suhaylī (henceforth, AS),6 is straightforward: it
provokingly elusive ʿAbbasid version was pro- claims to be a rewriting of Naṣrallāh Munshī’s
duced, a Persian administrator at the Ghaznavid twelfth-century text. And so it is, cannibalizing full
court, Abū al-Maʿālī Naṣrallāh Munshī (d. second sentences and complete passages from its hypo-
half of the twelfth century) wrote a Persian prosi- text. Traditionally, academia has been virulently
metric version of KD, the Kitāb-i Kalīla va-Dimna critical of Kāshifī’s version. I argue for the need to
(henceforth, NM KD).5 As we shall see, this ver- revisit this subjective and damaging attitude: my
sion’s filiation is not absolutely clear. About three analysis of his text shows, inter alia, the rewriter’s
intelligent and successful opening up of the sto-
Kāshifī (also used in the present essay), itself the origin for ries, his use of the cognitive effects of style, and his
the ʿAyār-i Dānish; a sixteenth-century abbreviation of the scholarly attitude toward the challenges offered by
latter by the Mughal statesman Abū al-Fażl; and a twelfth- the text’s history.7
century prose version by Bukharī, who is an exact contem-
porary of the twelfth-century Persian author Naṣrallāh
Munshī, and who wrote at the court of Mosul. His work
appears to have survived in a single manuscript and has
been edited; see Mohammad ibn ʿAbdullah Al-Bokhari, 6 I have used the Tehran Amir Kabir 1362 H. sh. edition of
Dastan-Ha-Ye Bidpay, ed. Parviz Natel Khanlari and the complete Anvār-i Suhaylī text (hereafter, AS ed.). It
Mohammed Roshan (Tehran, 1361). There is a thirteenth- is based on a Berlin 1301H.sh/1922 facsimile of an Indian
century masnavī by Aḥmad b. Maḥmud al-Ṭusi Qāni’ī that 1270 H.q. Mumbai edition. I found it relatively reliable
also appears to have survived in a single manuscript but in that it is obviously close to the unidentified edition
is as yet unedited: Qāni’ī, Kalīla va Dimna, British Library, used for Eastwick’s full translation: Ḥusayn Vāʿiẓ Kāshīfī,
Persian manuscript Add. 7766, in Charles Rieu, Supplement The Anvár-i Suhailí: Or, The Lights of Canopus, trans.
to the Catalogue of the Persian Manuscripts in the British Edward B. Eastwick (Hertford: Stephen Austin, 1854),
Museum (London: British Museum, 1895), vol. 2, 582. unabridged facsimile repr. Marston Gate, 2005 (hence-
5 I have used the Tehran 1977 edition by Mojtaba Minovī, forth, AS tr.) and to the few manuscripts (of Indian origin)
which is based on twelve different manuscripts (including that I was able to consult as well. See the special issue of
the 1156 one): Abu-l-Maʿali Nasrallāh Munshī, Tarjome-ye the Journal of the International Society for Iranian Studies,
Kalila va Dimna, ed. Mojtaba Minovi (Tehran: Dānišgah-i Iranian Studies 36/4 (December 2003), which is devoted
Tihrān, 1964), (henceforth, NM KD ed.). All translated to aspects of Kāshifī’s oeuvre. See also Maria E. Subtelny,
passages cited in this article are mine. I also give the ref- “Ḥusayn Vāʿiẓ Kāshifī,” in Encyclopaedia of Islam, third edi-
erences to the complete German translation, which is tion, ed. Kate Fleet, Gudrun Krämer, Denis Matringe, John
based on the Minovi edition: Nasr Allāh Munshī, Kalila Nawas, and Everett Rowson (Leiden: Brill, 2014), accessed
und Dimna: Fabeln aus dem klassischen Persien, trans. April 7, 2018; Maria E. Subtelny, “Kāshefi,” in Encyclopaedia
Sayf al-Dīn Najmā́bādī and Siegfried Weber (Munich: Iranica Online, 6:658–61, accessed November 2015;
C.H. Beck, 1996) (henceforth, NM KD Gtr.); as well as the Georgina Hermann, “Biographisches zu Ḥusayn Wāʿiẓ
references to the recent English translation, also based on Kāšifī,” in Corolla Iranica: Papers in Honour of Prof. Dr.
the Minovi edition: Nasrullah Munshi, Kalila and Dimna, David Neil MacKenzie on the Occasion of His 65th Birthday
trans. Wheeler Thackston (Indianapolis, IN: Hackett on April 8th, 1991, ed. Ronald Eric Emmerick and Dieter
Publishing Company, Inc., 2019) (henceforth, NM KD Etr). Weber (Frankfurt am Main; New York: Peter Lang, 1991),
See M. Omidsalar, “Kalila Wa Demna ii. The Translation by 90–100, provides details on Kāshifī’s career as preacher,
Abu’l-Maʿali Nasr Allah Monshi,” in Encyclopedia Iranica qādī, and leader of a Sufi convent.
Online, 2015, accessed November 2015. 7 Van Ruymbeke, Kāshefi’s Anvār.
The Donkey ’ s Ears and Heart 45
Both Persian authors provide interesting details move one step further up we see that, astonish-
in their authorial prefaces about their hypotexts. ingly, the author makes no reference to himself
The passage below is taken from the AS. Kāshifī or his Arab patron in the surviving Arabic IAM
traces – perhaps too neatly – the heredity of his versions.9 We may thus wonder whether Kāshifī
text back to the IAM KD and its putative Pahlavi had access to conclusive information that he is
ancestor: echoing, or whether his literary needs are satisfied
with a simplification (of, for example, the prevail-
The second Abbasid Caliph, Abū Jaʿfar ing ideas that are already present in such reli-
al-Manṣūr b. Muḥammad b. ʿAlī b. ʿAbdallāh able sources as Ibn al-Nadīm’s Fihrist10). Another
b. ʿAbbās (May God approve of them!), heard aspect to this heredity question is the value added
about the book. Very eager to obtain it, he by such a chain to the rewriters’ works. The most
used clever devices in order to get posses- evident effect is to stamp the text with the wisdom
sion of the Pahlavi copy. He ordered Imam of older generations and to advertise its present-
Abū al-Ḥasan ʿAbdallāh b. Muqanna, who day use and universal scope through these refer-
was the chief of the learned men of that age, ences to, respectively, its success and its patrons
to translate the whole of it from Pahlavi into across the centuries. Academia has not yet fully
Arabic…. And then again, Abū al-Ḥasan Naṣr clarified the function of Ibn al-Muqaffaʿ’s striking
b. Aḥmad Sāmānī commanded one of the and unusual introductory chapters, which contain
learned men of the age to translate that man- the legendary links to the origins of the text and
uscript from the Arabic language into the form such an integral part of the work. I will return
Persian idiom, and the poet Rūdakī, on the to some of these as-yet-unstudied aspects below.
order of the Sultan, shaped it as a string of Written more than three and a half centuries
verses. Again, Abū al-Muẓaffar Bahrām Shāh before Kāshifī’s work and the passage cited above,
b. Sulṭān Masʿūd, one of the descendants of the NM KD preface elicits more questions than
Sulṭān Maḥmūd the Ghaznavid Ghazi, the answers on the heredity question. Meanwhile,
patron of Ḥakīm Sanāʾī, issued the command it also contains several interesting references to
that the most eloquent of rhetoricians and the state of the KD field in the twelfth century. In
most rhetorically gifted of eloquent men, the introduction to his work, Naṣrallāh Munshī
Abū al-Maʿālī Naṣrallāh b. Muḥammad b. explains that several copies of the KD text were
al-Ḥamīd … , translate it from a copy of Ibn-i in circulation in Ghaznavid Iran. But we are not
Muqanna and this book which is now famous told whether all these versions were attributed to
as Kalīla and Dimna is the translation of the Ibn al-Muqaffaʿ, or whether they were the work
aforesaid learned man.8 of different authors. The text’s success among
the intellectual elite is suggested en passant by a
The neat piling up of textual heredity resembles name-dropping mention of an intellectual who
an isnād and makes much of the exalted personali- donated a copy of the book to Naṣrallāh Munshī:
ties of the patrons, who are introduced each time “At that time, the jurist ʿAlī Ibrāhīm Ismāʿīl … gave
as the intellectual instigators and financial back-
ers of the new versions. But how far can we rely 9 In the AS edition which I have used, the Arabic trans-
upon the crystal clarity of this textual genealogy, lator is known as Ibn al Muqannaʿ. This is mostly dis-
presented in fifteenth-century Herat? The previ- regarded as nothing more than a misspelling of Ibn
ous link in this chain, expressed in the preface of al-Muqaffaʿ.
10 Muḥammad ibn Ishāq Ibn al-Nadīm, The Fihrist of
the NM KD, is provokingly unclear. And when we al-Nadim: A Tenth-Century Survey of Muslim Culture,
ed. and trans. Bayard Dodge (New York: Columbia
8 AS, authorial preface, ed. 6; tr. 7–8. University Press, 1970), 715.
46 van Ruymbeke
me a copy of the book KD. I already had in my pos- advice lay forgotten and even obliterated. It
session several copies of this book, and I saw this came to my mind to produce a translation
as a good omen.”11 of the book, giving space to the words, and
By mentioning these different copies, he hints exposing its advice. I also thought to colour
that his Persian text will be a compilation of dif- it with Qurʾanic passages, tales, verses and
ferent (IAM KD?) versions. Next, he explains that proverbs, in order to give this book, which
several translations of the KD text existed at the proposes the quintessence of thousands of
time. Had all these been translated from Pahlavi years, a new interest and so that its use would
into Arabic? As this mention of them follows one not be kept from the people.14
of the Persian poet Rūdakī, they might also be
translations from Arabic into Persian: “After Ibn This last part refers to the prosimetric form of his
al-Muqaffaʿ had translated the book and Rūdakī Persian KD. The author mentions how mingling
had versified it, it was translated several times.”12 prose with verse and adding Qurʾanic sayings,
Naṣrallāh Munshī then moves on to another, (Arabic and Persian) verses, and tales and prov-
equally interesting topic: his understanding of the erbs whip up the audience’s interest. This remark
act of rewriting, together with a critique of his pre- forms part of a debate around the original form of
decessors. The rewriter’s aim is to give a chance for the IAM KD; the available versions are all in prose.
the meaning of the stories to resurface. It is inter- Naṣrallāh Munshī certainly implies here that the
esting to see him expose the contrast between pos- prosimetrum was his innovation and thus that the
itive theory (each successful rewriter moves a little versions he had at his disposal, his hypotext (IAM
closer to the ideal rendering of the content, giving KD and perhaps others as well), were in prose,
it a new form and clothing it in words that sharpen devoid of Qurʾanic and poetical inclusions. We
the reader’s understanding of its meaning) and may also note how he explicitly indicates that the
negative reality (in fact, the rewriters increasingly book is not meant to be kept secret, which con-
lose interest in the ideal meaning, and focus on the trasts with what is stated in the traditional descrip-
text’s entertainment value): tions of the prehistory of the text.
At all events, each of the two Persian writers
Every [author] has progressed according claims a descent from the mid-eighth-century
to his own skill, in the field of clear expres- Arabic text as part of his own text’s identity. The
sion. But it is evident that their intention vagaries of manuscript survival have ensured that
has been to narrate the tales and write the the oldest surviving NM KD manuscript is older
stories, not teach wisdom and advice. What than the oldest surviving IAM KD manuscript.15
beautiful words have they mutilated, in addi-
tion to abbreviating the stories! In short, as religious passages. The narrative and the dialogues are
people had given up the wish to study Arabic in Persian. The recent 2019 translation by Thackston
books,13 all [the KD’s] wisdom and good (NM KD Etr.) leaves out the Arabic passages.
14 NM KD, authorial preface, ed. 25, Gtr. 383, Etr. xxvi.
15 The oldest known NM KD manuscript dates from AD
11 NM KD, ed. 18, Gtr. 376, Etr. xxiii. 1156, thus during or very close to Nasrallāh Munshī’s life-
12 NM KD, ed. 25, Gtr. 383, Etr xxvi. time. It was used by Minovi for his 1977 Tehran edition
13 When we read Kāshifī’s assessment of Naṣrallāh of the text (see above, footnote 5). The earliest known
Munshī’s version (AS, authorial preface, tr. 7–9), the IAM KD manuscript dates from AD 1221, more than
same topos emerges in relation to the contemporary four and a half centuries after IAM’s death. The second
audience’s boredom with difficult Arabic passages. oldest dates from AD 1338. See François De Blois, “The
Despite the author’s explanations quoted above, the Pancatantra: From India to the West – and Back”, in A
NM KD version can almost be described as bilingual Mirror for Princes from India: Illustrated Versions of the
due to its inclusion of numerous Arabic verses and Kalilah Wa Dimnah, Anvar-i Suhayli, Iyar-i Danish, and
The Donkey ’ s Ears and Heart 47
Globally, it is fair to say that, through its greater 1 “The Sick Lion, the Fox, and the
length and sophistication, the Persian NM KD ver- Donkey’s Ears and Heart”
sion offers clearer access to the KD’s pedagogy of
universal political philosophy and human psychol- In what follows I compare this brief sub-story, in
ogy than do the surviving versions of the IAM KD. the fifth chapter of the KD, as it is presented in
Basing his own work on this version, Kāshifī has several editions of “the” IAM KD version(s) and in
taken further steps to broaden our understanding the Persian NM KD and AS versions. The discussion
of the KD text. In a very helpful exegetical spirit, will open up the story’s different levels and mean-
he has expanded upon the existing stories and while also demonstrate aspects that are typical of
has added sub-stories in order to comment upon the KD’s structure and content in general. I will
or clarify thorny textual points. He has also thor- also expose the relevance of textual fluctuations
oughly revisited the evidence around the origins within the primary sources on which the KD field
of the text and, as I mention below, provides us is based, pointing out key elements of the rewrit-
with fascinating conclusions that reflect thorough ings. Meanwhile, I will highlight how the Persian
research into the question. I thus argue here that versions develop the story’s depth and interest.
aspects of the Persian versions might legitimately The KD’s fifth chapter is built around the compli-
and usefully illuminate the field of Arabic KD stud- cated main story of “The Monkey and the Tortoise,”
ies. But it is also evident that there is an indepen- which I summarize below as it gives the embed-
dent “Persian” KD field: internal textual evidence ding context for our sub-story.17 The fifth chapter
shows how the Persian intelligentsia nurtured a is brief in comparison to the four earlier ones and
fascination for the KD’s literary history and mean- contains only one sub-story. The KD story is well
ing, and initiated integrated compilations of KD known and I will not retell it here, but I highlight
versions. The two Persian versions also show a full the elements that are relevant for my discussion.
critical engagement with the multiple meanings of The fifth chapter’s main story is announced as
the text. The analysis of a sub-story which has not an illustration of the necessity to hold on to that
been comprehensively analyzed to date, despite which has been achieved through great effort.18 As
its intriguing content, will illustrate the challenges
the KD text throws our way.16 17 In view of the difficulty of classifying the varied con-
tents of the KD as fables, I prefer to adopt the follow-
ing designations: the chapters are introduced by a
frame-story in the form of a dialogue between King
Dabshalim and the philosopher Bidpay; Bidpay then
narrates the main stories, in which are embedded one
Humayun Nameh, ed. Ernst J. Grube (Mumbai: Marg or two levels of sub-stories.
Publications, 1991), 10–15; Béatrice Gruendler, “Les ver- 18 This is emphasized in the table of contents of the IAM
sions arabes de Kalīla wa-Dimna : une transmission KD translation by André Miquel, who systematically
et une circulation mouvantes,” in Énoncés sapientiels adopts these indications and adds them to the usual
et littérature exemplaire: une intertextualité complexe, character-based titles: “Le singe et la tortue, ou : il ne
ed. Marie-Sol Ortola and Marie-Christine Bornes-Varol suffit pas d’acquérir, il faut encore conserver”; see ʿAbd
(Nancy: Presses Universitaires de Nancy – Éditions Allāh Ibn al-Muqaffaʿ, Le livre de Kalila et Dimna, trans.
Universitaires de Lorraine, 2013), 387–418. André Miquel (Paris: C. Klincksieck, 1957). Another
16 I chose to analyze this particular sub-story mainly misleading element in Miquel’s table of contents is his
because it does not feature in Van Ruymbeke, Kāshefī’s listing of imagery within the text on the same level as
Anvār. A typical KD story, it provides an excellent the sub-stories. In the fifth chapter, he mentions “Le
example of the necessity for layered analysis and illus- veau et sa mère” and “L’homme tombé à terre” next to
trates several KD key features. The difference in charac- “L’âne, le lion et le chacal,” while the first two items do
ters between the Arabic and Persian versions, which is not refer to full stories, but are simply alluded to by the
specific to this sub-story, also offers interesting insights characters. In his version, Kāshifī, typically, has fleshed
into the rewriters’ engagement with their text. out this lean chapter with an additional sub-story, that
48 van Ruymbeke
is most often the case, it is with a counter-example plan to murder the monkey as just a step further in
that the KD story illustrates its wise maxim. This the practical uses of this “friend.”19
maxim, the superficial meaning of the counter- The monkey’s mistake is to accept to cross the
example, and the “moral” that caps the story do not water in total reliance on this “friend” whom he
circumscribe the chapter’s full scope, but only its barely knows. Having lost his power as a ruler, he
most superficial level. The different ways in which is seduced into thinking that nothing he possesses
the adventure of “The Monkey and the Tortoise” could kindle anyone’s interest and he thus trusts
impacts each of its characters hold important that the tortoise’s friendship is as materially self-
information. The value of the fifth chapter nestles less as his own. He does not reflect that his very
in the exposition of merciless ambition, as well body could be considered valuable. The tortoise’s
as in the setting out of the techniques of trickery, mistake is to consider that the monkey will not
and in the explanation of the pitfalls and strengths mind the switch from generously sharing superflu-
of manipulation. At its core, it denounces human ous figs to sharing something as vital as his heart.
reliance upon friendship and trust, a denunciation As he believes that the monkey is unable to escape
which is the true central thread woven through the
book. The fifth main story and its sub-story shine
19 The importance of this philosophical topic and the care
a harsh light on the loneliness of the ruler, even with which KD analyses it are underscored by evident
when he has grown old and has retired. He must similarities between situations portrayed within the
learn to view with doubtful eyes the lure of friend- stories and disquisitions such as those found in Books 8
ship proposed by apparently selfless but secretly and 9 of Aristotle’s Nichomachean Ethics; Aristotle, The
Nicomachean Ethics, ed. Lesley Brown, trans. David Ross
ambitious second parties. (Oxford: Oxford University Press, 1980). These were
Challenged by a younger monkey, the old mon- later integrated, discussed, and expanded in Cicero’s
key king has been unable to hold on to his rule treatise Laelius de Amicitia. Here is Aristotle recogniz-
and, now an outcast, has retired to a fig tree, in ing the luring power of friendship: “For without friends,
no one would choose to live, though he had all other
which he spends dull, lazy, solitary days. Used to goods; even rich men and those in possession of office
easy munificence, he idly throws figs into the river. and of dominating power are thought to need friends
He is unaware that a tortoise in the water is enjoy- most of all; for what is the use of such prosperity with-
ing the delicious figs as a magnificent present and out the opportunity of beneficence, which is exercised
chiefly and in its most laudable form towards friends?
consequently wishes to attach himself to this gen-
Or how can prosperity be guarded and preserved with-
erous monkey. Two different but equally selfish out friends? The greater it is, the more exposed it is to
interests are active in this unnatural friendship: risk.” Aristotle, 142. Here is Cicero mentioning exactly
the monkey craves company and conversation the sort of bondage that prevails between the monkey
with an intellectually attuned friend, while the tor- and the tortoise, and he rejects labelling it as friend-
ship: “Well, then, it has very often occurred to me when
toise views their association as a profitable friend- thinking about friendship, that the chief point to be
ship bringing him material advantages. He uses considered was this: is it weakness and want of means
the monkey as a benevolent, undemanding meal that make friendship desired? I mean, is its object an
ticket. It is thus unproblematic for him to yield interchange of good offices, so that each may give that
in which he is strong, and receive that in which he is
to his jealous wife’s demands that he should kill
weak? … [A]s to material advantages, it often happens
the monkey and use his heart to cure her illness. that those are obtained even by men who are courted
Despite some inner qualms, the tortoise views the by a mere show of friendship and treated with respect
from interested motives. But friendship by its nature
admits of no feigning, no pretense: as far as it goes it is
both genuine and spontaneous.” Cicero, De Senectute;
De Amicitia; De Divinatione, trans. William A. Falconer
of the king of Kashmir and his monkey-sentinel (AS tr. (London; Cambridge MA: W. Heinemann Ltd; Harvard
376–79). University Press, 1959), sec. 8.
The Donkey ’ s Ears and Heart 49
(as he cannot swim), he rashly unveils his murder- I deconstruct this sub-story as follows, using the
ous plan to his victim. When the monkey realizes version proposed by A. Miquel as a starting point.21
the tortoise’s intentions, he uses the best of his psy- It is worth remembering here that the existing
chological skills within the register of the friendly Arabic IAM KD manuscripts propose versions that
relationship they have established. He pretends he differ in more than simple details. As Miquel chose
agrees to donate his heart, but as he left his heart several manuscripts on which to base his French
behind, he first needs to be brought back to the translation, his text should in fact be considered
tree. The tortoise is thus tricked into taking him a new version, a compilation of the elements that
back. As soon as the monkey is safe in the tree, he are common to all his manuscripts, thus, mostly,
explains to the tortoise that he will not trust him erasing their specificities. There are clear differ-
a second time. He tells the story of the donkey’s ences among the three Arabic translations I have
ears and heart, which appears to illustrate just used here, and also between these and the two
that mistake of falling twice into the same trap. Persian rewritings. An obvious and intriguing dif-
The chapter closes on the tortoise’s final expres- ference is that in the two Persian versions, the role
sion of regret. of the jackal is filled by a fox (rūbāh). I examine
The KD sub-stories carefully weave in formal below the probable significance of this difference
themes and imagery that cross-reference with in animal characters. In the case of the differences
those of their main stories. Nevertheless, these between the two Persian versions, we have enough
sub-stories are not indispensable to an under- evidence to understand the rationale that guided
standing of the main stories. In fact, the analysis Kāshifī. I argue for considering and working with
of the main stories mostly benefits from being the differences within the textual tradition: they
sealed off from the intrusion of sub-stories. And are situated at the kinks in the storyline and
inversely, as discussed above concerning the intro- show how each rewriter, translator, or scribe has
ductory announcements of the chapters’ main engaged with these difficulties, offering interpre-
stories, the sub-stories’ scope similarly stretches tations and addressing apparent incoherencies.
well beyond the narrow straightjacket of the pro- There is an important caveat concerning the dis-
and epimythia that introduce the “official” way cussion that follows: for brevity’s sake, I consider
of understanding their multiple facets.20 In fact, only the differences that are relevant to the limited
these stories puzzle us because they do not illus- aspects I wish to develop within the boundaries of
trate their announced meaning. We cannot make this article.22
sense of all the elements in this sub-story if we
take them as a narrow illustration of the monkey’s 21 Ibn al-Muqaffaʿ, Le livre de Kalila et Dimna, 192–95.
lucky escape. As we shall see, if the inclusion of Regrettably, Miquel’s IAM KD does not provide a schol-
arly apparatus and we are left guessing at the differ-
this story within the main story of “The Monkey ences that his version has smoothed out. This is also
and the Tortoise” really functioned only as an illus- the case with each of the three other IAM KD transla-
tration of the pro- and epimythia (to show that tions used here; it is also the state of the recommended
the monkey is not as stupid as the donkey and will editions of the Arabic text. The edition of the Persian
NM KD proposes manuscript variations. The AS edition
not be caught twice in the same obvious mortal
does not.
trap), then the second part of the sub-story (after 22 To palliate somewhat the unstable manuscript tradi-
the killing of the donkey), which contains the tion, I have used the following four translations of the
sub-story’s elucidation and raison d’être, would be IAM KD text, which are based on different manuscripts
supererogatory. and show important variations: Ibn al-Muqaffaʿ, Le
livre de Kalila et Dimna (based on the Cairo 1941 edition
by Azzām, itself based on a 1221 manuscript from Aya
Sofia); ʿAbd Allāh Ibn al-Muqaffaʿ, Kalila and Dimna,
20 See Van Ruymbeke, Kāshefi’s Anvār, 75–76, 116–18. or, The Fables of Bidpaï, trans. Wyndham Knatchbull
50 van Ruymbeke
1. A lion, weakened by a terrible rash, cannot 7. The jackal says that he will now need to use
hunt any longer. His worried subordinate, the serious trickery to bring back the donkey,
jackal, enquires about this change of habit. who is upset. He explains to the donkey that
The lion explains his illness. There is one the attack was in reality the fiery she-donkey
cure: to eat the ears and heart of a donkey. jumping on its back. If the donkey had stayed
2. The jackal knows that every day a donkey is on a little, then she would have submitted.
brought to the neighboring meadow. He will 8. The excited donkey hurries back to the lion’s
lure it to the lion’s meadow. meadow. The lion kills it.
3. The jackal manipulates the donkey: pitying 9. Now, the lion explains that he should first
his exhaustion, he advises it to run away from make his ablutions, and only eat the ears
his cruel master. He adds that, hidden in a and the heart and donate the rest as alms.
neighboring meadow full of tempting grass, He orders the jackal to guard the body till he
there is a stunning she-donkey in search of a returns.
male. 10. Remaining on his own, the jackal then eats
4. The jackal warns the lion of the approaching the ears and the heart. He knows that the
donkey. But the lion is too weak and cannot lion will see this as a bad omen and will not
kill it. The prey runs off. touch the rest.
5. The jackal is upset. If the lion let the donkey 11. As the lion returns, the jackal explains that
slip away on purpose, then why ask for it to this donkey had neither ears nor heart, for
be brought to him? Or was he too weak to otherwise it would never have returned after
catch it? This is the end of everything: the the first lucky escape.23
king is unable to prevent a donkey from slip- The KD ploy identified above is also used in this
ping away! sub-story. The superficial point of the story is
6. The lion understands the danger: if he says announced by its context in the main story (i.e.,
he let the donkey escape, then it will appear the monkey’s escape and awareness of the tor-
as if he has treated the jackal like an idiot; toise’s murderous intentions; see above) as cen-
if he admits he is too weak, the jackal will tered on the stupidity of the donkey, who falls
despise him. He proposes a deal: if the jackal into the same trap a second time. Our first reac-
is able to bring the donkey back, then he will tion when attempting to decode the story is to
answer the question. conclude that the donkey would have been well-
advised to question the jackal/fox’s motive in pro-
posing to guide it to a lush meadow complete with
(Oxford: Printed by W. Baxter for J. Parler, 1819) (based an attractive she-donkey, and that his stupidity in
on Silvestre de Sacy’s 1816 collation of different manu-
scripts in the Bibliothèque nationale of Paris); ʿAbd
returning there a second time is quite extraordi-
Allāh Ibn al-Muqaffaʿ, Kalila und Dimna: die Fabeln des nary, and rather unbelievable. The Arabic authors
Bidpai, trans. Philipp Wolff (Zurich: Manesse Verlag, grappled with this problem and resolved it by
1995) (based on Silvestre de Sacy’s 1816 collation of introducing a farcical register. By the inclusion of
different manuscripts of the Bibliothèque nationale
a gross “sexual” element that irresistibly lures the
of Paris); ʿAbd Allāh Ibn al-Muqaffaʿ, Le Pouvoir et les
Intellectuels ou les Aventures de Kalila et Dimna, trans. donkey to the meadow, the necessity for prudence
René R. Khawam (Paris: G.-P. Maisonneuve et Larose,
1986) (based on a personal collation of all of the
above, plus the Beirut 1904 edition by P.L. Cheikho 23 IAM KD, Miquel’s translation, 192–95. The passage is
[based on a 1338 manuscript], and on Kalīla wa-Dimna found in IAM KD Khawam’s translation: 263–65, IAM
extracts present in a tenth/eleventh-century text by KD Knatchbull, 264–67, and IAM KD Wolff, 190–93).
Muḥammad al-Yamānī [Mudhāhāt amsāl kitāb Kalīla NM KD, ed. 253–257, tr. 205–09; AS, ed. 353–359, tr.
wa Dimna bimā ashbahahā min ashʿār al-ʿArab]). 393–400.
The Donkey ’ s Ears and Heart 51
and logic is evaded. The two Persian rewriters gloss admitted that he is ready to do anything in order
over the sexual references. (NM speaks of “another to get the prey. This marks out the relationship
donkey.” AS does not retain this second donkey.) between the jackal/fox and the lion as central to
But they keep the farcical atmosphere around the story, and helps us recognize that the donkey,
the donkey by stressing the victim’s truly asinine rather than being the central victim, is merely
blindness: in one case, it is unable to tell a lion a passive object, a bone of contention between
from a she-donkey; in the other version, it believes the two protagonists. Of these, the lion loses to
that the lion is a statue. The variations show that the jackal/fox’s proactive, winning manipulation.
the authors all struggled with the implausibility of Nevertheless, as we shall see, it is the lion who is
this element within the plot. This weakness warns the main character in the story; the central lesson
us that the story’s relevance should necessarily be relates to his mistakes, not the jackal/fox’s tricks.
found in another analytical dimension. The lion’s final mistake, carelessly leaving the
The jackal’s successful manipulative technique jackal/fox to guard over the dead prey, is again a
introduces a level of interest but its pedagogy is contrived element. One of the Arabic versions
limited because of its reliance on ad hoc elements includes a pious reference to pre-prandial ablu-
such as the specific character of the donkey. We tions: rather than insisting on purifying himself
learn how to capture the future victim’s trust by before his meal, here the lion will not eat the ears
pitying its status and proposing an attractive solu- and heart unless they are cleansed and purified,
tion, based in this case on the lure of sloth, food, and leaves this to his subordinate’s care.24 The
and sexual satisfaction. Extreme sexual excite- function of this mistake is to illustrate one of the
ment might blind a certain type of victim to real- themes of the fifth chapter (how both the monkey
ity. This is recognized and used in the four Arabic and the tortoise are punished for their negligence)
versions, while the Persian authors choose rather and to show what tragic consequences this negli-
to denounce the friendship argument that equally gence may have.
powerfully manipulates the donkey. AS expands The final trick the jackal/fox plays on the lion
the dialog between the jackal/fox and the donkey is unexpected and unexplained. It is independent
into an exercise in comprehensive manipulation. of the trick played on the donkey and appears
Both Persian versions, however, also emphasize to lead to a non-conclusion. On the surface, its
the manipulation of the lion by the jackal/fox. ultimate function might almost be to provide an
The story’s interest skyrockets when understood opportunity for the jackal/fox’s final bon mot. He
from this perspective. The story does indeed high- gleefully puns on the ears and heart as the seats of
light the dangers of repeating a mistake. However, hearing and reason, with their direct relationship
beyond the donkey’s blindness and repeatedly to intelligence. But I read this piece of pseudo-
misplaced trust, the lion also makes a mistake in wisdom as an unsatisfying narrative pirouette dis-
using the puny subordinate’s help a second time. guised as an aphorism. The stated morals point in
The jackal/fox has now understood the lion’s the wrong direction and sit uncomfortably with
weakness and has gauged the extent of his depen- the meanings of the stories they are supposed to
dence on help. The subordinate’s ambition is thus introduce or conclude; this is a stock technique of
no longer checked by his previous awe for the lion’s the KD. Taken literally, the jackal’s final decision
might and physical power. The lion is helpless to is problematic: it robs the lion of his only cure,
escape his subordinate’s manipulation and is thus but this also boomerangs on the jackal/fox’s life.
persuaded to act against his nature and dignity. He A weak lion incapable of hunting also means that
becomes a passive player in ensuring the success the jackal/fox can no longer live on the leftovers.
of the jackal/fox’s second attempt. The subordi-
nate jackal/fox now dominates the lion, who has 24 Knatchbull’s Arabic IAM KD version, 266.
52 van Ruymbeke
We are not given an explanation for the surprise (allegorized by the ears and the heart of the don-
decision to humiliate the lion. Is it greed for the key) between an old, weakened ruler (the lion)
whole carcass? Several authors mention the jack- and a puny mercenary or close associate (the
al’s confidence that, rather improbably, the lion jackal/fox). The story opens at a moment when an
would not eat the rest of the donkey if he could old ruler rashly trusts a mercenary with the secret
not have the ears and heart. The difficulty has not of the ultimate fragility of his position. But the
escaped Kāshifī’s notice either. As is his wont, he story’s poignancy and interest lie in our realization
emphasizes the misleading passage by adding to that the ruler has no choice other than to rely on
its incoherence: tongue in cheek, he explains that another’s help. This is allegorized by the fact that
the fox eats the ears and heart as these are the the lion’s illness prevents him from hunting and
choicest parts of the donkey.25 condemns him and his courtiers to starvation.
As this literal analysis does not provide us with There is more than pure pride in a ruler avoid-
a meaningful explanation for the jackal/fox’s final ing the confession of weakness. To do so would
trick, we are challenged to read the story differ- be tantamount to inviting a change of regime. To
ently, leaving the superficial level behind and div- give his subordinates the information that he is no
ing into another dimension in which the jackal/ longer able to dominate his external and internal
fox’s final action is expected and fully rational. enemies is a capital mistake for one whose posi-
In fact, the KD episodes blossom into their full tion is dependent on his aura of strength. Mindful
meaning when considered as political metaphors. that he should avoid making a strong contender
Universal political psychology and philosophy even stronger, our lion correctly chooses to ask
form the core of this text, which is known to have for help from the jackal/fox, an irrelevant charac-
fascinated the ruling elite through the centuries ter (who is neither a member of the lion family,
and across cultures and which presents itself as nor one of the courtiers). His explanation mini-
a mirror for princes.26 Not surprisingly, it is in mizes his loss of strength (he is no longer hunting
this light that we may hope to grasp some of its because he is weakened by a rash) and he presents
redoubtable relevance. his needs as very modest (to eat the ears and heart
In the story we are examining, we may dis- of a donkey).27 At this stage, the jackal/fox is glad
cern the model of a desperate struggle for power to offer his help, hoping for normal gratification.
This is, in a way, a confidence boost for the lion,
25 This is one of the recurring techniques used by Kāshifī who believes the win-win contract will satisfy his
to attract the reader’s attention and to suggest other subordinate. Interestingly, in the two Persian ver-
levels of meaning. sions, the jackal/fox appears to understand that
26 I relate KD’s structure to the way in which the politi- the lion’s misery is simply caused by his loss of
cal theorist Hannah Arendt (d. 1975) has described
the political use of storytelling. Storytelling is the only
beauty. He is asked to help cure aesthetic worries,
possible medium in order to express the complex- not the loss of hunting powers. This is a meaning-
ity of philosophical or other conundrums. See Van ful nuance and shows a greater political awareness
Ruymbeke, Kāshefi’s Anvār, p. 127 (“Storytelling as a in the Persian lions, who remain wary of disclosing
Methodology of Political Theory”): “The tragedies
physical weakness.
enacted within the [KD] stories present situations in
which the choice is amongst values that cannot be The lion’s inability to catch the donkey is the
calibrated against a common standard of measure…. point of no return in this tragedy and each of the
The point of the stories is intellectual, not practical,
and this point is found behind the action depicted, not
within in.” Decoding the stories yields a meaning that 27 This is an intriguing detail that probably holds inter-
is of universal use, because every new reader will relate textual references that I cannot place, though there is
it to his/her own personal case. The reader’s efforts to also an obvious echo of the monkey’s heart mentioned
decode the stories contain the pedagogy. in the main story.
The Donkey ’ s Ears and Heart 53
rewriters has engaged with the jackal/fox’s reaction a crucial political moment, or to offer a subordi-
to it. Unable to believe that the lion is so fickle as nate who is aware of one’s weakness an opening
to let the prey escape without reason, he suddenly for appropriating power. This also holds an echo
gauges the full extent of the lion’s weakness. Some of the tortoise’s catastrophic procrastination in
of the Arabic versions have the jackal exclaim, “Are allowing the old monkey to climb back in the fig
you as weak as all that?”28 Intriguingly, the Persian tree to fetch his heart. The jackal/fox immediately
foxes do not comment on this shattering discov- seizes the opportunity: he gobbles up the very
ery. The NM KD fox mutters that no intelligent and parts that their contract had reserved as the lion’s
courageous person would let go of nourishment if share and that he knows are vital for the recovery
there were no advantage to be had from doing this, of the ruler’s strength. By doing this, he effectively
or if it were not absolutely unavoidable. The AS fox destroys the lion’s hope of recovery. He triumphs
berates the lion for cruelty against the donkey. In and allows himself to further humiliate the former
both Persian texts, the mercenary is careful not to ruler with a mocking explanation. It is now the
show that he has understood the lion’s weakness. jackal/fox who holds the power and can manip-
Meanwhile, the jackal/fox’s position has ulate the ruler as he wishes. We note that by his
become one of strength. He now knows the ruler’s decision to go for help to the subaltern jackal/fox
secret and what is needed to cure him and help rather than to his own family, the lion has in effect
him restore his rule. Too late, the lion realizes the enabled the latter’s political takeover.
danger implicit in the subordinate’s expression The jackal/fox is the secondary character in the
of surprise. The different versions also grapple story’s demonstration. His actions are simply logi-
with this moment of sickening realization. At the cal responses to the lion’s mistakes. This is another
jackal/fox’s query, “Are you as weak as all that?,” central technique of the KD text: the core inter-
the lion either balances the possible responses, or est of the stories always lies in both the mistakes
instinctively does not answer but orders that the and the apparently correct decisions of the loser.
donkey be brought to him a second time.29 The Rather than offering lessons or techniques for
Persian lions hold on to the appearance of regal attaining success, the KD sets up problematic situ-
dignity: the NM KD lion says that not everyone is ations and illustrates the concatenations of inevi-
able to understand a ruler’s decisions. The AS lion table catastrophes following upon (re)actions
speaks harshly about the fox’s lack of respect in based on common sense. The lion’s apparently
questioning his actions. At this point in the story, wise decision to ask for help from a small jackal/
it is likely that the jackal/fox has already hatched fox, his religious scruples, and his trust in his sub-
his full plan, but he still needs the lion to kill the ordinate’s awed respect all relate to common sense
prey, offering him the opportunity to appropriate and all push him closer toward catastrophe. The
the carcass and, in allegorical terms, seize power. text’s technique is to play out, not spell out, this
Once the victim is killed, the lion makes his lesson. In contrast, the “moral lessons” that intro-
last mistake, a truly catastrophic one. He does not duce or cap the stories and any expressed autho-
capitalize on his effort and victory, but rather fol- rial indication of how we could interpret the
lows too scrupulously what seems to be a religious stories are all teasingly superficial, or downright
rule. The reason for the procrastination is in fact opposed to the stories’ meanings.30
irrelevant and does not interfere with the story’s
ultimate meaning: it is clearly unwise to delay at 30 As is the case with many pre-modern texts, the KD
does not provide information. It presents situations
designed to make the reader think. Decoding the sto-
28 Khawam and Wolff. Knatchbull (265) voices this ries yields a meaning that is of universal use, because
slightly differently. every reader anew will relate it to his own personal
29 Pace Khawam, Knatchbull, and Wolff. case. The reader’s effort to decode the stories contains
54 van Ruymbeke
Behind the conviction that fortune’s apparent with its numerous intratextual references between
benevolence, which might ultimately be revealed the main story and the sub-story. In the example
as malevolence, cannot be conquered by rulers, discussed here, the misadventure of the sick lion
however energetic and intelligent, the KD also relates to the opening of the fifth main story,
teaches the danger of friendship.31 Over and over where we are introduced to the monkey king, who
again, from many angles, couched in every type has lost his power because of old age. This main
of relationship enacted by animals or humans, story is presented as an illustration of the danger
this mirror for princes holds a grim warning: the of negligence and the difficulty of maintaining
ruler is emphatically alone and can trust no one one’s grasp on what one has acquired. Conquest
within the dangerous jungle of courtiers, viziers, often happens through fortune and may be dis-
politicians, and nobles, who surround him like armingly easy; it is the administration of the con-
sharks. Some stories illustrate the other side of quest, its preservation, that is the tricky part and
the coin, which is equally dangerous: a subordi- that demands a lot of cunning and effort. The main
nate who accepts friendship from a powerful ruler story shows this twice over: due to his age, the old
also courts danger. Humanity’s natural inclination monkey is unable to preserve his rule over the
to rely upon either political or private friendship tribe; and later in the story, the tortoise is unable
relations is denounced by each of the stories in to keep hold of his prey, the monkey, this time
the book. through relying on what he believes is common
friendship. He should never have told the monkey
of the plan to kill him and should never have fallen
2 A Tight Structure into the psychological trap set by the desperate
monkey. The same questions are explored in the
Moving from the topical analysis to other aspects sub-story: the weakened state of the lion means an
of the story, I would like to note an important irretrievable loss of power, whatever he may do to
structural point: the story’s careful construction, palliate it; and the lion is careless to leave his re-
conquered power, allegorized by the dead prey’s
the pedagogy. I have analyzed the workings of the KD ears and heart, with the emboldened subordinate.
stories in Christine van Ruymbeke, “Dimna’s Trial and
Furthermore, the donkey, having miraculously
Apologia in Kashifī’s Anvār-i Suhaylī. Morality’s Place
in the Corrupt Trial of a Rhetorical and Dialectical escaped with his life a first time, should not have
Genius,” Journal of the Royal Asiatic Society 26, no. 4 carelessly neglected the warning and walked into
(2016): 549–83. The introduction and the conclusion the same trap again. Finally, the lion should never
of the story relate to the punishment of the manipula- have believed that it was possible to pull off the
tor, while the chapter deals throughout with the unfair
charging of a “black sheep,” with the ruler influencing
trick a second time, in different circumstances, as
a corrupt judicial power, and with a stunning demon- the jackal/fox was now aware of his weakness.
stration of the power of forensic rhetoric.
31 The works of Niccolò Machiavelli (d. 1527), such as
The Prince and the Discourses on Livy, share many of
3 Fox or Jackal?
KD’s universal themes, and I have found I could benefit
from aspects highlighted in research on his ideas. But
Machiavelli illustrates these themes not with “animal I have highlighted several instances illustrating the
fables” but with examples from Classical Antiquity and Arabic and Persian rewriters’ engagement with the
from the recent Italian past. And the Florentine author text. The nuances and changes that we note across
spells out the lessons he intends his readers to learn
from his chosen exempla, thus reducing the possible
the versions mostly occur at significant moments
scope of the stories to ad hoc theories, which are often when the authors/rewriters/translators stumble
problematic. over a lack of verisimilitude in the synopsis and
The Donkey ’ s Ears and Heart 55
engage with them. In the case of the Persian ver- individual psychology that makes him choose a
sions, the two authors have adopted a technique specific attitude.
that exacerbates rather than erases the unlike- In the case of the jackal vs. the fox, I have shown
lihood of these situations. In the story above, elsewhere that there is a clear difference between
Naṣrallāh Munshī tackles the donkey’s inability the characteristics of these two animals in the
to recognize that the meadow is inhabited by a KD texts. It is helpful to clarify immediately that
lion by exaggerating the donkey’s stupidity: it is neither of the two match the label of a paradig-
made to believe that the lion is the she-donkey of matic trickster that we automatically attach to a
its dreams so it strokes the lion. Similarly, Kāshifī fox in European lore. This latter role is rather held
makes his donkey unable to distinguish between a by the hare characters in the KD stories. The KD
living lion and a statue, and has his lion masquer- fox represents a not-too-clever actor whose con-
ade as a statue for a long time, until the donkey stant devouring hunger leads him into problem-
falls asleep. atic, sometimes dangerous situations.32 A typical
Among the changes introduced by the differ- KD fox features in the sub-story of “The Fox and
ent versions, I would like to look more closely now the Drum,” which occurs early on in the first main
at what might appear to be simply an intriguing story of “The Lion and the Bull.” The fox is hun-
nuance between the Arabic version(s) mention- gry and decides to attack a drum rather than a fat
ing a jackal, and the two Persian versions featur- fowl: he has never seen a drum before but trusts
ing a fox. In any case, this difference between the that it will reward his pains as it makes such a
Arabic and the Persian versions reveals a constant huge noise. He is disappointed when he discov-
effort toward thoughtful exegesis in each new ers that the drum is empty and worth nothing as
KD version. prey. The other character, the KD jackal, is not a
Not all the KD stories feature animals. Some trickster either. He is a sophisticated actor, char-
feature human characters, some portray interac- acterized by his parresia, true and wise language
tions between humans and animals, and some “as spoken to power”,33 and usually features as a
take place in a human-like world peopled by ani- trusted but unimportant legal advisor, expressing
mal actors. The latter contain a further level of the long view, listing bullet points of legal rules,
psychological complexity because the animals are
individuals who might choose not to act according 32 Van Ruymbeke, Kāshefi’s Anvār, 91–99. Kāshifī has
to the typical traits of their species; additionally, contrasted the fox and the jackal characters in an addi-
these animals often hold an office or function that tional sub-story in the second chapter: a hungry fox
stupidly lets go of an uninteresting prey in the hope
is also expected to come with certain character- of catching a fat fowl. A passing jackal wisely advises
istic attitudes and ideas. These may, or may not, against this course and does not attempt to trick the
fit with the characteristics of the animal. Thus, a fox. See also further similar characterizations of the fox
human hermit, for example, is expected to act as in: “The Fox and the Drum” (Chapter I); “The Fox and
the Two Rams” (I); “The Hare, the Wolf and the Fox”
a saintly person. If he acts as a fiend, he becomes
(I); “The Hunter and the Fox” (I); “The Sick Lion, the
a far more interesting character because he for- Fox and the Donkey’s Ears and Heart” (Chapter V);
sakes his expected nature. A cat-hermit is a more “The Lioness, the Hunter and her Cubs” (Chapter X);
complex character: as a cat, he will act cruelly and “The Lynx and the Hierarchy of Power” (X). Kāshifī
toward mice, but as a hermit he should act in a has also added a sub-story illustrating the difference
between the trickster hare and the hungry fox.
saintly manner. The cat-hermit character embod- 33 Michel Foucault, Discourse and Truth and Parrēsia,
ies an intrinsic psychological clash. There is no edited by Henri-Paul Fruchaud and Daniele Lorenzini
way to know in advance how he will act: it is his (Chicago: 2019) 8–11.
56 van Ruymbeke
and defending the status quo. This wise jackal is come up with the usual legal musings we expect
typified by Kalīla.34 The fascination we feel for the from the KD jackals. Meanwhile, the reasons for
character of the jackal Dimna comes from the fact featuring a jackal character relate to evident intra-
that he is a black sheep among the advisor jackals. textual ties with at least two other KD episodes fea-
He has personal ambition and has exchanged par- turing this animal. Shanzaba the bull is introduced
resia for rhetoric. Pretending to give advice, he in to the lion by the jackal Dimna, which leads to the
fact manipulates in order to achieve his ambition: bull’s eventual murder by the lion. This is echoed
to climb from legal advisor to the exalted rung in our sub-story by the way in which the jackal/
of private royal advisor. This is unexpected for a fox lures the donkey to the lion’s meadow. There
jackal character and explains why both the lion is also the role played by the jackal in the story of
and the bull trust his advice so blindly. “The Lion, the Camel, the Jackal, the Crow, and the
There are several other complicated jackal char- Wolf,” which occurs as a sub-story within the first
acters in the KD. The vegetarian hermit jackal who main story. A jackal is part of the plan to lure the
is the central character in the main story of “The camel to serve as the lion’s meal; he reminds his
Lion and the Jackal” has a very complex persona. comrades of their legal duties toward the lion. The
Instead of playing his part in society as a legal advi- jackal character as the paradigmatic smaller and
sor, he has decided to cut himself off and live as a weaker advisor of the lion is also well adapted to
hermit. It is not as a typical legal advisor, but in the his role in the story analyzed here.
much higher position of a hermit-sage that he is Contrary to the accepted view, I suggest here
invited to the court. Though as a jackal he should that the Arabic versions of the story that feature
be a carnivore, here he has become a vegetarian a jackal might derive from a later revision of the
(stressing the disconnectedness between jackals text. Though it is impossible to establish whether
and food). His enemies attempt to frame him for the jackal or the fox is original, I think that the
stealing meat but the revelation of his vegetarian- available IAM KD versions assimilated the begin-
ism saves him. ning of our sub-story to the Dimna story and con-
In the story of “The Donkey’s Ears and Heart,” sidered that a jackal character was better adapted
the servant hovers between the persona of the to the role, when in fact the character’s similarities
jackal (as a petty advisor) and that of the fox (as to the emblematic KD fox character are the stron-
the story revolves around hunger and greed). ger. Whatever the case might have been, this detail
There are grounds for choosing either of the per- shows how the authors/translators engaged with
sonalities as the original one, though my guess is their texts, and also illustrates how these detailed
that the fox personality came first: food is key to innovations blur the evidence toward a recon-
the action and the trick that is used on the don- struction of the hereditary line of rewritings and
key is elementary while also relating to food. The translations of the KD versions.
fox manipulates but does not advise, nor does he We have seen how several passages have been
fleshed out and made more amusing or interesting
by the Persian rewriters. There are also moments
34 It is certainly not by chance that his character is
in the Persian KD versions where the psychologi-
adopted by the Ikhwān al-Ṣafāʾ in the twenty-second
Epistle, where the animals elect Kalīla for his wisdom cal or philosophical depth they propose opens up
and rhetorical proficiency as their spokesman to pres- the stories’ full meaning, which is absent from the
ent their case against the humans; Ikhwān al-Safa, The Arabic IAM KD versions that have come down to
Case of the Animals versus Man before the King of the us. Again, this suggests that the Persian authors
Jinn: A Translation from the Epistles of the Brethren
of Purity, trans. Richard J.A. McGregor and Lenn Evan
culled this sophistication from more complete ver-
Goodman (Oxford: Oxford University Press, 2009), sions, now lost. The unlikely alternative would be
86–90, 159–65. to credit them with a tremendous exegetical and
The Donkey ’ s Ears and Heart 57
literary effort that allowed them to inject philo- jackals in the Sanskrit text.35 This achieves in the
sophical meaning into innocuous stories. Sanskrit text a blurring of the two different animal
characters and seems to indicate a late transla-
tion rather than textual anteriority. The KD stories
4 Concluding Remarks and their themes also point toward roots situated
within the Persian rather than the Indian literary
The analysis here has addressed two methodologi- and historical past, and abound with literary and
cal issues that I identify as essential within the philosophical references to Western Antiquity,
field of KD studies. The first is the importance of with similarities to passages of Aristotle and
close textual analysis, which allows us to recog- Cicero.36
nize the various obfuscating tools used to mislead The literary legend of the Indian origin of the
the hurried reader, while alerting the thought- text is expressed in the story of Burzūya, found
ful one to the necessity of analytical decoding. It in the preface to the IAM KD, which pollinated
seems that the Persian versions, in their additional every mention of the KD text. However, it is pos-
details, propose a more sophisticated approach sible to view Burzūya’s adventure in India and his
to the stories’ breadth and depth of meaning. The literary theft of a secret text hidden in an Indian
example used in the present article may represent royal treasure as an allegory of Ibn al-Muqaffaʿ’s
only flimsy evidence, but this is supported by simi- translation into Arabic of a well-guarded and trea-
lar and more robust findings in my recent mono- sured Pahlavi text. Such an allegory would explain
graph on AS. why the Arabic KD authorial introduction lacks a
It is also true that Kāshifī in particular provides mention of the names of the actual author and his
clues on the stories’ scope by adding seemingly
innocuous comments or additional stories, and
also by stressing the ridiculousness of a situation
or a reaction beyond the boundaries of reality. 35 Van Ruymbeke, Kāshefi’s Anvār, 91–97. See for example,
These latter moments were misunderstood by ear- Patrick Olivelle, trans., Pañcatantra: The Book of India’s
lier commentators, who read them at face value Folk Wisdom (New York: Oxford University Press, 1997).
and derived the impression that the AS rewriting Inversely, it is remarkable that in the European versions
all the jackals are translated into foxes. Note how the
was worthless. I rather see these changes as mark-
animal characterizations lose their clarity with each
ers provided by an author who has adopted the translation. The jackals Kalīla and Dimna thus become
core technique of the KD: the reader must learn asses in some Western versions; Thomas North, The
to rise above the misleading comments and obfus- Morall Philosophie of Doni Drawne out of the Auncient
cating remarks. The nimbleness at decoding that Writers (London: Henry Denham, 1570).
36 I point interested readers to the following passages
he will acquire through this exercise is an impor- in my recent monograph on AS: Chapter 1 (1.2.2.3
tant part of the text’s practical pedagogy. “Zopyros, the Wandermotiv of the mutilated hero”) and
This leads me full circle to one of the introduc- (1.5.2 “Exit Borzuyeh”); Chapter 2 (2.3.2 “Never Mistake
tory remarks: the KD text itself provides evidence a Jackal for a Fox”); Chapter 3 (3.1.5 “Pigeonholing the
Professional Vizier and the Occasional Hermit”) and
that the available Sanskrit Pañchatantra versions
(3.3.1.3 “The Horrendous Zahhak-Theme”); Chapter 5
cannot be considered authoritative. The analysis (5.1.2.3 “Wrapping up Each Frame”; and 5.2.2 “KD/AS’s
above touches upon just one of these clues: the Elusive Internal Architecture”). Chapter 7 in toto con-
Arabic and Persian KD versions contain both jackal tains misgivings on the way the hereditary line of the
and fox characters with clearly differentiated psy- KD text has been hitched onto the Sanskrit texts. See
also Louise Marlow, “Among Kings and Sages: Greek
chological characteristics. The Pañchatantra does and Indian Wisdom in an Arabic Mirror for Princes,”
not recognize these differences: all the foxes that Arabica: Journal of Arabic and Islamic Studies 60, no.
appear in the Arabic and Persian versions are 1–2 (2013): 1–57.
58 van Ruymbeke
patron. It is also remarkable that Ibn al-Muqaffaʿ’s by Ernst J. Grube, 10–15. Mombai: Marg Publications,
Persian name, Rūzba, is an anagram of Burzūya.37 1991.
One of the most interesting points we may Foucault, Michel. Discourse and Truth and Parrēsia.
derive from this final remark is the awareness that Edited by Henri-Paul Fruchaud and Daniele
the KD text has challenged and intrigued its audi- Lorenzini. Chicago and London: The University of
ence through the centuries. The Timurid court in Chicago Press, 2019.
particular appears to have examined with much Gruendler, Béatrice. “Les versions arabes de Kalīla wa-
care and attention the received knowledge around Dimna : une transmission et une circulation mou-
the puzzling origins of the text, coupling this with vantes.” In Enoncés sapientiels et littérature exem-
an in-depth study of its content which resulted in plaire : une intertextualité complexe, edited by
the AS, Kāshifī’s masterful rewriting. With its revo- Marie-Sol Ortola and Marie-Christine Bornes-Varol,
lutionary revisiting of key aspects of the text, and 387–418. Nancy: Presses Universitaires de Nancy –
with its helpful exegetical additions, it revivifies Éditions Universitaires de Lorraine, 2013.
the KD field and suggests fascinating new avenues Hermann, Georgina. “Biographisches zu Ḥusayn Wāʿiẓ
of exploration. Kāšifī.” In Corolla Iranica: Papers in Honour of Prof.
Dr. David Neil MacKenzie on the Occasion of His 65th
Birthday on April 8th, 1991, edited by Ronald Eric
Bibliography Emmerick and Dieter Weber, 90–100. Frankfurt am
Main; New York: Peter Lang, 1991.
Al-Bokhari, Mohammad ibn ʿAbdullah. Dastan-Ha-Ye Ibn al-Muqaffaʿ, ʿAbd Allāh. Kalila and Dimna, or, The
Bidpay. Edited by Parviz Natel Khanlari and Fables of Bidpaï. Translated by Wyndham Knatchbull.
Mohammed Roshan. Tehran, 1361. Oxford: Printed by W. Baxter for J. Parler, 1819.
Aristotle. The Nicomachean Ethics. Edited by Lesley Ibn al-Muqaffaʿ, ʿAbd Allāh. Kalila und Dimna: die
Brown. Translated by David Ross. Oxford: Oxford Fabeln des Bidpai. Translated by Philipp Wolff.
University Press, 1980. Zurich: Manesse Verlag, 1995.
Asadī Ṭūsī, Abū Manṣūr Aḥmad ibn ʿAlī. Asadī’s neu- Ibn al-Muqaffaʿ, ʿAbd Allāh. Le livre de Kalila et Dimna.
persisches Wörterbuch Lughat-i Furs: nach der einzi- Translated by André Miquel. Paris: C. Klincksieck,
gen vaticanischen Handschrift. Edited by Paul Horn. 1957.
Berlin: Weidmannsche Buchhandlung, 1897. Ibn al-Muqaffaʿ, ʿAbd Allāh. Le Pouvoir et les intellectuels
Cicero. De Senectute; De Amicitia; De Divinatione. Trans- ou les Aventures de Kalila et Dimna. Translated by
lated by William A. Falconer. London; Cambridge René R. Khawam. Paris: G.-P. Maisonneuve et Larose,
MA: W. Heinemann Ltd; Harvard University Press, 1986.
1959. Ibn al-Nadīm, Muḥammad ibn Ishāq. The Fihrist of al-
De Blois, François. “The Pancatantra: From India to the Nadim: A Tenth-Century Survey of Muslim Culture.
West – and Back.” In A Mirror for Princes from India: Edited and translated by Bayard Dodge. New York:
Illustrated Versions of the Kalilah Wa Dimnah, Anvar-i Columbia University Press, 1970.
Suhayli, Iyar-i Danish, and Humayun Nameh, edited Ikhwān al-Safa. The Case of the Animals versus Man
before the King of the Jinn: A Translation from the
37 Van Ruymbeke, “Kalila and Dimna as a Case-Study”; Epistles of the Brethren of Purity. Translated by
also Van Ruymbeke, Kāshefi’s Anvār, 338: each of the Richard J.A. McGregor and Lenn Evan Goodman.
letters of RUZBH are used to form BRZUH. In the lat- Oxford: Oxford University Press, 2009.
ter form, Burzūya, the final -H is constructed as a Kashīfī, Ḥusayn Vāʿiẓ. The Anvár-i Suhailí: Or, The Lights
vocalic -eh/a, with an inserted -y- (similar to the vari-
of Canopus. Translated by Edward B. Eastwick.
ant of the ezafe after a vowel). Ibn al-Muqaffaʿ might
have borrowed the name of a historical character who Hertford: Stephen Austin, 1854, unabridged facsimile
is mentioned in historical references. repr. Marston Gate, 2005.
The Donkey ’ s Ears and Heart 59
Kashīfī, Ḥusayn Vāʿiẓ. Anvar-e Sohayli, ya, Kalila wa Ruymbeke, Christine van. “Dimna’s Trial and Apologia
Dimna-ye Kashefi, Tehran: Amir Kabir, 1362 H. sh. in Kashifī’s Anvār-i Suhaylī. Morality’s Place in the
Marlow, Louise. “Among Kings and Sages: Greek and Corrupt Trial of a Rhetorical and Dialectical Genius.”
Indian Wisdom in an Arabic Mirror for Princes.” Journal of the Royal Asiatic Society 26, no. 4 (2016):
Arabica: Journal of Arabic and Islamic Studies 60, 549–83.
no. 1–2 (2013): 1–57. Ruymbeke, Christine van. “Kalila and Dimna as a Case-
Nasrallāh Munshī, Abu-l-Maʿali. Kalila und Dimna: Study: Ibn al-Muqaffaʿ’s and Nasrullah Munshi’s
Fabeln aus dem klassischen Persien. Translated by Translations.” In The Routledge Handbook of Arabic
Sayf al-Dīn Najmā́bādī and Siegfried Weber. Munich: Translation, edited by Sameh Hanna, Hanem El-
C.H. Beck, 1996. Farahaty, and Abdel-Wahab Khalifa, 253–68. Lon-
Nasrallāh Munshī, Abu-l-Maʿali. Kalila and Dimna. don: Routledge, 2020.
Translated by Wheeler Thackston. Indianapolis, IN: Ruymbeke, Christine van. “Authorship, Ownership and
Hackett Publishing Company, Inc., 2019. Rewriting: Vaʿiz Kashifi and Abu’l-Fazl B. Mubarak
Nasrallāh Munshī, Abu-l-Maʿali. Tarjome-ye Kalila within the Hereditary Line of Kalila wa-Dimna
va Dimna. Edited by Mojtaba Minovi. Tehran: Authors.” Jerusalem Studies in Arabic and Islam 45
Dānišgah-i Tihrān, 1964. (2018): 181–210.
North, Thomas. The Morall Philosophie of Doni Drawne Ruymbeke, Christine van. Kāshefi’s Anvār-e Sohayli:
out of the Auncient Writers. London: Henry Denham, Rewriting Kalila and Dimna in Timurid Herat. Leiden:
1570. Brill, 2016.
Olivelle, Patrick, trans. Pañcatantra: The Book of India’s Subtelny, Maria E. “Ḥusayn Vāʿiẓ Kāshifī.” In Encyclo-
Folk Wisdom. New York: Oxford University Press, paedia of Islam, third edition, edited by Kate Fleet,
1997. Gudrun Krämer, Denis Matringe, John Nawas, and
Omidsalar, M. “Kalila Wa Demna ii. The Translation by Everett Rowson. Leiden: Brill, 2014.
Abu’l-Maʿali Nasr Allah Monshi.” In Encyclopedia Subtelny, Maria E. “Kāshefi.” In Encyclopaedia Iranica
Iranica Online, 2015. Online, 6:658–61.
Qāniʿi, Aḥmad b. Maḥmud al-Ṭusi. Kalīla va Dimna. Venuti, Lawrence. “The Translator’s Invisibility.”
Add. 7766. British Library, London. Criticism 28, no. 2 (1986): 179–212.
Rieu, Charles. Supplement to the Catalogue of the Persian
Manuscripts in the British Museum. London: British
Museum, 1895.
3
Kalīla, Dimna et la géomancie
Jean-Charles Coulon
Le manuscrit illustré de Kalīla wa-Dimna de la leur importance et d’ouvrir des pistes d’interpréta-
Bibliothèque nationale de France Arabe 3467 com- tion. Nous nous appuierons alors sur des exemples
porte de nombreuses figures géomantiques, tant concrets à travers les illustrations du manuscrit.
dans les images qu’en marge du texte1. En effet,
les « figures géomantiques » sont un ensemble
de seize figures de quatre rangées comportant 1 Sources et principes fondamentaux
chacune un ou deux points. Elles servent de base de la géomancie arabe
à une technique divinatoire arabe appelée « géo-
mancie » (ḍarb al-raml ou khaṭṭ al-raml en tant La géomancie (ʿilm al-raml) est, littéralement,
que praxis, ʿilm al-raml en tant que science). Cette la science de la divination par le sable. Il s’agit, à
branche de la divination a connu un immense l’origine, d’un rite cléromantique consistant en
succès dans le monde islamique et une impor- un jet de cailloux (ṭarq ou ḍarb) – ou de grains de
tante production écrite en témoigne. Ces figures céréales ou de noyaux. Le résultat de ce jet devait
ont donc des significations particulières, que être interprété par un géomancien (rammāl)2.
le géomancien doit pouvoir interpréter dans le La géomancie acquit un statut particulier parmi
cadre d’un tirage. La présence de ces figures en les diverses formes de divination pratiquées dans
marge d’un manuscrit dont le contenu n’a aucun le monde islamique au Moyen Âge. En effet, cette
rapport avec la géomancie a donc de quoi sur- science était réputée avoir été révélée par l’ar-
prendre. Pourquoi des « figures géomantiques » change Gabriel au prophète Idrīs (Hénoch), qu’il
se retrouvent-elles ainsi perdues dans un recueil aurait lui-même transmise au mystérieux Ṭumṭum
de fables animalières illustré ? Nous proposerons al-Hindī, devenu une autorité importante en la
donc des pistes d’interprétation sur le sens à don-
ner à ces figures mystérieusement placées dans ce 2 Pour une introduction générale à la géomancie, voir
manuscrit alors que Kalīla wa-Dimna ne traite pas Toufiq Fahd, La Divination arabe : études religieuses, socio-
de géomancie. Tout d’abord, il convient de présen- logiques et folkloriques sur le milieu natif d’Islam (Paris :
ter la géomancie et le rôle de ses figures dans un Sindbad, 1987), 196‑204. Voir également les travaux d’Anne
Regourd, « La géomancie comme voie d’accès à un savoir
tirage afin de comprendre le sens qu’elles peuvent ésotérique : un cas isolé au nord du Yémen ? », in Scienza
revêtir d’une manière générale. Un dénombre- e islam : atti della giornata di studio, Venezia, 30 gennaio
ment et une description de la place de ces figures 1999, éd. Giovanni Canova (Rome : Herder, 1999), 5-16 et
dans ce manuscrit permettent ensuite d’apprécier Regourd, « Pratiques de géomancie au Yémen », in Religion
et pratiques de puissance, éd. Alain de Surgy (Paris :
L’Harmattan, 1997), 105-127 ; ainsi que les articles récem-
1 Note de l’auteur : nous tenons à remercier Éloïse Brac de ment parus de Matthew Melvin-Koushki, « In Defense
la Perrière, Annie Vernay-Nouri et Aïda El Khiari de nous of Geomancy : Šaraf al-Dīn Yazdī Rebuts Ibn Ḫaldūn’s
avoir donné cette occasion de se pencher sur ce manuscrit, Critique of the Occult Sciences », Arabica 64, no 3‑4 (2017) :
ainsi que pour leurs remarques et commentaires qui nous 346‑403 et Melvin-Koushki, « Persianate Geomancy from
furent précieuses. Nous remercions également Abdallah Ṭūsī to the Millennium : A Preliminary Survey », in The
Cheikh-Moussa, qui nous a généreusement transmis des Occult Sciences in Pre-modern Islamic Cultures, éd. Nader
copies des éditions de Kalīla wa-Dimna, ainsi que le rap- El-Bizri et Eva Orthmann (Beyrouth : Orient-Institut,
porteur anonyme pour ses précieuses remarques. 2018), 151‑200.
matière3. Un hadith fait même dire à Muḥammad : largement circulé6. Cependant, maintes fois réé-
« Parmi les prophètes, il y en avait un qui prati- dité, chaque version comporte des variantes, sans
quait le khaṭṭ ; quiconque réussira à le faire confor- que le manuscrit d’origine soit toujours clairement
mément à son procédé, saura ce que ce prophète identifié7. Dans ces conditions, il demeure fort
savait4. » Ainsi, loin d’être perçue comme une difficile de faire un tableau cohérent et unifié de
science sulfureuse ou suspecte, la géomancie cette tradition textuelle. Tout reste encore à faire
jouissait d’une aura qui explique l’abondante pro- en matière d’édition critique sur le sujet. Nous
duction écrite à son sujet. puiserons donc notre matière sur la géomancie
Au regard de cette abondante production écrite des manuscrits à notre disposition, étant entendu
arabe médiévale sur la géomancie, les études sur le qu’une étude approfondie des textes de géoman-
sujet n’en sont qu’à leurs débuts. En effet, la quasi cie dépasserait nettement les limites imposées par
absence d’édition critique de textes géomantiques notre étude des figures de géomancie du manus-
et les nombreuses éditions défectueuses et contra- crit illustré de Kalīla wa-Dimna.
dictoires des textes fondamentaux de la discipline La séance divinatoire consiste en l’établisse-
ne permettent pas de faire une présentation rigou- ment et l’interprétation d’un tableau géomantique
reuse et précise de tous les aspects du système de seize figures (shakl) disposées dans seize mai-
géomantique5. À titre d’exemple, la plus impor- sons (bayt). Les quatre premières maisons (mai-
tante autorité en la matière est al-Zanātī (date sons I à IV), appelées « les mères » (al-ummuhāt),
de mort inconnue, floruit avant 629H/1230), dont sont établies en fonction de traits tracés dans le
un opuscule édité au Caire en 1280H/1864 sous sable ou la terre ou de séries de points par séries
le titre al-Faṣl fī ʿilm al-raml qui lui est attribué a de quatre lignes sur une feuille de papier. Chacune
de ces lignes est réduite à un point si le nombre
de traits ou de points est impair, et à deux points
ou un trait si le nombre de traits ou de points est
3 Emilie Savage-Smith et Marion B. Smith, Islamic Geomancy
pair. Il en résulte seize figures possibles, que l’on
and a Thirteenth-Century Divinatory Device (Malibu : appelle « figures géomantiques » ou « squilles »
Undena, 1980). Anton Hauber avait émis l’hypothèse que (de l’arabe shakl)8. Une fois ces quatre premières
le nom de Ṭumṭum al-Hindī dériverait de Dindymus, maisons établies, on en déduit les quatre suivantes
soit saint Thomas. Sur Ṭumṭum al-Hindī, voir Anton
(maisons V à VIII), appelées « les filles » (al-banāt).
Hauber, « Ṭomṭom (Ṭimṭim) = Δανδαμις = Dindymus ? »,
Zeitschriften der deutschen Morgenländischen Gesellschaft La figure de la première maison des filles (maison
63 (1909) : 457‑472 ; Felix Klein-Franke, « The Geomancy V) réunit les quatre rangées supérieures des figures
of Ahmad B. ‘Ali Zunbul a Study of the Arabic Corpus
Hermeticum », Ambix 20, no 1 (1973) : 26‑35 ; voir égale-
ment Jean-Charles Coulon, La Magie en terre d’islam au 6 Il s’agit d’une lithographie suivie d’une Risāla fī l-jafr
Moyen Âge (Paris : CTHS, 2017), 108. waqurʿa attribuée à Jaʿfar al-Ṣādiq. La lithographie a été
4 Traduit par Toufiq Fahd, « Khaṭṭ », in Encyclopaedia of faite au Caire par le cheikh ʿAbd al-Munʿim al-Ṣabrī ;
Islam, Second Edition, éd. Peri Bearman et al. (Leiden : Muḥammad al-Zanātī, al-Faṣl fī Uṣūl ʿIlm al-Raml (Le
Brill, 1960-2007). Le hadith se trouve dans les collections Caire : Maṭbaʿat al-jumal al-miṣriyya, 1927).
de Muslim (Masāǧid, 33 ; Salām, 121), Abū Dāwūd (Ṣalāt, 7 Voir Anne Regourd, « Au sujet des sources manuscrites de
167 ; Ṭibb, 23), al-Nasāʾī (Sahw, 20) et Ibn Ḥanbal (II, 394 ; l’ouvrage imprimé au Caire sous le titre d’Al-faṣl fī uṣūl ʿilm
V, 447) (références données ici d’après la Concordance de al-raml d’Al-Zanātī », Annales islamologiques 35 (2001) :
Wensinck). 393‑347.
5 Blanca Villuendas Sabaté travaille actuellement à l’édition 8 Gabriel Ferrand, « Un chapitre d’astrologie arabico-
d’un important traité d’Ibn Maḥfūf sur la géomancie. Elle a malgache », Journal Asiatique, 10e série, t. 6 (1905) : 195 ;
en outre publié des fragments en judéo-arabe de la Geniza repris par Fahd, « Khaṭṭ ». Le terme squille n’a toutefois
du Caire traitant de géomancie. Voir Blanca Villuendas pas eu de postérité en français et n’a jamais été reconnu
Sabaté, La geomancia en los manuscritos judeo-árabes de la dans les dictionnaires, c’est pourquoi nous utiliserons plu-
Gueniza de El Cairo (Cordoue : CNERU-CSIC, 2015). tôt l’expression « figure géomantique ».
62 Coulon
figure 3.1 Un texte de géomancie attribué à Abū Muḥammad ʿImrān al-Munajjim al-Jabsūyī al-Faylasūf (recueil, Paris
BULAC, ms. 579, f. 47v-48r)
des quatre maisons mères, la deuxième maison maisons XIII et XIV s’obtiennent selon le même
des filles (maison VI) réunit les quatre deuxièmes principe à partir des figures des maisons IX à XII :
rangées des quatre figures des maisons mères, la la figure de la maison XIII se déduit de la somme
troisième maison des filles (maison VII) réunit les des figures des maisons IX et X et la figure de la
quatre troisièmes rangées des figures des quatre maison XIV de la somme des figures des maisons
maisons mères, et la quatrième maison des filles XI et XII. Toujours selon le même principe, la
(maison VIII) réunit les quatre quatrièmes ran- figure de la maison XV est l’addition des figures des
gées des figures des quatre maisons mères. Leur maisons XIII et XIV. Enfin, la figure de la seizième
établissement suit le même principe que le tirage : et dernière maison (maison XVI) est le résultat de
si le nombre de points est impair, alors on ne la somme des figures des maisons I et XV. Pour
retient qu’un seul point ; s’il est pair, on en retient prendre un exemple simple, à partir d’un tirage
deux (ou un trait). Les quatre maisons suivantes donnant pour les maisons mères (I à IV) les figures
(maisons IX à XII) s’obtiennent par la réunion de suivantes : , , et , on obtient le tableau géoman-
deux maisons précédentes : ainsi, la figure de la tique suivant9.
maison IX est la somme des figures des maisons
I et II, la figure de la maison X est la somme des
figures des maisons III et IV, la figure de la maison
9 Nous reprenons ici l’exemple donné par Emilie Savage-
XI est la somme des figures des maisons V et VI, et Smith. Voir Savage-Smith et Smith, « Islamic Geomancy
la figure de la maison XII est la somme des figures and a Thirteenth Century Divinatory : Another Look », in
des maisons VII et VIII. Ensuite, les figures des Magic and Divination in Early Islam, 222.
Kalīla, Dimna et la géomancie 63
tableau 3.1 Exemple de tableau géomantique d’un tirage avec pour maisons mères (I à IV) les figures suivantes : , , et
Mères Filles
IX X XI XII
XIII XIV
XV XVI
Dans le cas d’une consultation géomantique, et en interprétant le texte de cette page. La pra-
chaque maison et chaque figure a un sens propre. tique est ancienne et attestée dans l’Antiquité, par
La conjonction des deux doit être interprétée par exemple avec les œuvres d’Homère et de Virgile
le géomancien. (on parle alors de rhapsodomancie)10. Cette tech-
Dans le cas qui nous occupe, le manuscrit Arabe nique divinatoire est attestée en terre d’islam dès
3467 ne contient que des figures géomantiques. Il l’époque omeyyade avec le Coran, mais égale-
n’y a pas d’organisation de ces figures en maisons ment à l’époque abbasside avec le célèbre recueil
et on ne peut pas en déduire un tel tableau. Il ne de hadiths Ṣaḥīḥ d’al-Bukhārī11 (m. 256H/870).
s’agit donc pas d’un cas ordinaire de tirage géo- Bien entendu, dans le cas du Coran ou du Ṣaḥīḥ,
mantique. Cependant, les figures en elles-mêmes le caractère sacré de ces textes concourt à leur
ont une signification, indépendamment de leur donner une légitimité dans cette pratique divi-
présence dans une maison ou non. La maison sert natoire. La bibliomancie trouva également un
à orienter le sens de la figure dans un cas précis important développement dans l’Iran safavide
soumis au devin, comme le contexte permet de avec des manuscrits spécifiquement dédiés à un
déterminer quel est le sens précis à donner à un usage divinatoire : les Fāl-Nāma. Si certains corans
mot, à partir de potentialités de significations. ont été copiés dans une optique divinatoire12,
c’est surtout le Dīwān du célèbre poète Ḥāfiẓ
(m. 792H/1390 ou 791H/1389), qui fit l’objet d’une
2 Le cas du manuscrit Arabe 3467 :
entre géomancie et bibliomancie ?
Nous pouvons toutefois émettre une hypothèse 10 Auguste Bouché-Leclercq, Histoire de la divination dans
l’Antiquité : divination hellénique et divination italique
sur le rôle ou l’utilisation de ces figures dans le
(Grenoble : J. Millon, 2003), 155.
manuscrit Arabe 3467 de la BnF. En effet, la pré- 11 Voir Fahd, La Divination arabe ; Fahd, « Khaṭṭ ».
sence de symboles divinatoires dans les marges 12 Voir par exemple Sabrina Alilouche et Ghazaleh
d’un manuscrit ne manque pas de rappeler les Esmailpour Qouchani, « Les gloses marginales et le
pratiques bibliomantiques. Comme son nom l’in- Fālnāma du coran de Gwalior, témoignages des usages
multiples du coran dans l’Inde des sultanats », in Le
dique, la bibliomancie est l’art de la divination par Coran de Gwalior : polysémie d’un manuscrit à peintures,
le livre. Plus concrètement, il s’agit de tirer des pré- éd. Éloïse Brac de la Perrière et Monique Burési (Paris :
sages en ouvrant un livre sur une page au hasard Éditions de Boccard, 2016), 85-110.
64 Coulon
formellement). Cette représentation fait donc pos- base du système symbolique utilisé dans les
siblement un lien entre ces deux formes de divina- « sciences occultes » islamiques. En effet, le rai-
tions que sont la géomancie et le fāl-nāma. sonnement analogique est un « mode d’identi-
Une autre spécificité du manuscrit Arabe 3467 fication qui fractionne l’ensemble des existants
dans la perspective d’un usage bibliomantique en une multiplicité d’essences, de formes et de
doit être mentionnée : d’une manière générale, la substances séparées par de faibles écarts, parfois
bibliomancie se pratique avec des manuscrits soit ordonnées dans une échelle graduée, de sorte qu’il
conçus dès l’origine comme des supports divina- devient possible de recomposer le système des
toires (par exemple, certains corans biblioman- contrastes initiaux en un dense réseau d’analo-
tiques), soit c’est le texte du manuscrit lui-même gies reliant les propriétés intrinsèques des entités
qui est un support reconnu de la divination distinguées19 ». Ainsi, dans ce système de pensée,
(Coran, Dīwān de Ḥāfiẓ, etc.). Dans le cas présent, chaque existant réunit les caractéristiques d’un
l’hypothèse d’un usage bibliomantique provient ensemble d’essences dont il est à la fois la somme
de figures géomantiques ajoutées postérieure- et une entité distincte. Par exemple, un objet aura
ment à la copie du texte lui-même. Il s’agirait donc à la fois les propriétés de sa couleur, de sa matière,
d’un cas unique dans notre connaissance actuelle. de l’astre sous lequel il a été fabriqué, etc. De là,
Il convient alors, afin d’examiner cette hypothèse, les figures géomantiques ont pu prendre des signi-
de tenter de donner un sens aux figures géoman- fications particulières en fonction des éléments
tiques qui ornent les pages de ce manuscrit. qui leur sont associés. Par exemple, dans le Kitāb
Il existe donc seize figures géomantiques grâce al-raml d’un auteur anonyme conservé dans le
auxquelles il est possible de tirer des présages. Leur manuscrit Arabe 2731 de la Bibliothèque nationale
nom peut parfois varier d’un manuscrit à l’autre, de France, les premiers folios sont consacrés aux
d’une région à l’autre, mais nous pouvons toutefois seize figures géomantiques20. À chaque figure est
observer des constantes. Nous pouvons réunir en attribué une maison, un nom, une ou deux lettres,
un tableau les noms des figures géomantiques18. un signe du Zodiaque, un astre, une couleur, une
Les géomanciens musulmans ont rapidement nature (chaud/froid, sec/humide, faste/néfaste,
inscrit les seize figures géomantiques et les seize masculin/féminin), un type de nourriture et une
maisons dans les correspondances analogiques, direction. Ces informations sont résumées dans le
tableau 3.3.
Les correspondances astrologiques ne sont
18 Afin d’établir ce tableau, nous avons compilé diverses
études sur la géomancie afin de réunir toutes les pas propres à ce traité, au contraire, elles se
variantes connues des noms des figures, ainsi que les retrouvent dans l’essentiel des traités de géo-
définitions et propositions de traduction de plusieurs mancie, à commencer par celui d’al-Zanātī21. Ces
dictionnaires. Voir Robert Jaulin, La Géomancie : ana-
lyse formelle (Paris, La Haye : Mouton, 1966) ; et Jaulin,
Géomancie et Islam (Paris : Christian Bourgois, 1991) ; 19 Philippe Descola, Par-delà nature et culture (Paris :
Aboubekr A. Ben Choaib, « La bonne aventure chez Gallimard, 2005), 280.
les musulmans du Moghrib », Revue africaine, n°50 20 Kitāb al-raml, Paris, BnF, Arabe 2731, ff.1v‑3.
(1906) : 62‑71 ; Jean-Claude Hébert, « Analyse structu- 21 Al-Zanātī, al-Faṣl fī Uṣūl ʿIlm al-Raml. Signalons à ce
rale des géomancies comoriennes malgaches et afri- propos un article substantiel disponible sur internet
caines », Journal des Africanistes 31, no 2 (1961) : 115‑208 ; bien qu’il n’ait jamais été publié à notre connaissance :
Villuendas Sabaté, La geomancia en los manuscritos Wim van Binsbergen, « The Astrological Origin of
judeo-árabes de la Gueniza de El Cairo. Pour les défini- Islamic Geomancy », communication présentée à la
tions que nous avons indiquées, voir aussi Jean-Baptiste 15e conférence annuelle de la SSIPS (Society for the
Belot, al-Farāʾid al-durriyya. ʿArabī – faransī (Beyrouth : Study of Islamic Philosophy and Science) et SAGP
Dār al-Mashriq, 1971) ; Albert de Kazimirski Biberstein, (Society of Ancient Greek Philosophy), Binghamton,
Dictionnaire arabe-français (Paris : Maisonneuve, 1860) ; octobre 1996. https://www.labirintoermetico.com/
Reinhart Dozy, Supplément aux dictionnaires arabes 07Geomanzia/Binsbergen_van_W_The_astrological
(Leyde, Paris : Brill, Maisonneuve, 1967). _origin_of_Islamic_Geomancy.pdf. Comme son titre
66 Coulon
Maisona Figure Nom arabe Lettres Signe Astre Couleur Nature Nourriture Direction
a Ici, il s’agit pour le texte d’indiquer quelle figure géomantique correspond le mieux à la symbolique de la maison dans un tirage. Cependant,
bien entendu, dans un tirage géomantique, n’importe quelle figure peut se retrouver dans n’importe quelle maison : les quatre premières
maisons sont déterminées directement par le tirage, les douze suivantes sont déduites à partir des quatre premières issues du tirage.
68 Coulon
Maisona Figure Nom arabe Lettres Signe Astre Couleur Nature Nourriture Direction
b Il y a une erreur dans le manuscrit car la figure représentée ici est celle d’al-ʿUqla/al-Iʿtikāf (6).
Kalīla, Dimna et la géomancie 69
données permettent déjà de voir le lien qui peut et hypocrite, al-Jawdala est immorale et sans
être fait avec Kalīla wa-Dimna. En effet, la symbo- honneur (mumāziq), Naqī al-khadd est pure
lique astrologique des fables de Kalīla wa-Dimna et mélangée, al-Jamāʿa est pure et mélangée,
ne pouvait échapper au lecteur arabe cultivé de al-Ṭarīq est immorale, al-Tiqāf24 est sincère
cette époque. L’histoire met en effet en scène et al-Ijtimāʿ est sincère et mélangée. » Ainsi, la
un lion et un taureau, deux signes du Zodiaque. signification des figures géomantiques peut égale-
Astrologiquement, le Lion22 est le signe fixe du ment revêtir une dimension morale. Or, Kalīla et
feu et le Taureau signe fixe de la terre, soient deux Dimna est un recueil de fables à vocation morale
parfaits antagonistes, le premier régnant sur le qui dénonce les agissements d’un menteur immo-
ciel et représentant la puissance céleste (l’astre ral et les dégâts qu’il peut causer auprès d’un sou-
associé au Lion est le Soleil) et le second étant verain désireux de faire régner la justice. Ainsi,
le signe de la terre et représentant la puissance ici encore nous pouvons tisser un lien entre les
chtonienne. Cet antagonisme se traduit et trouve significations potentielles des figures et le cadre
son paroxysme dans le combat où le taureau est du texte de Kalīla et Dimna : cette potentialité ne
finalement tué par le lion. Aussi, nous puiserons peut se réaliser que dans un contexte donné, qui
des informations dans les traités d’astrologie et de semble concorder dans le cas présent.
sciences occultes sur les représentations associées
aux astres et signes zodiacaux correspondant aux
figures géomantiques présents dans ce manuscrit. 3 Description quantitative
On trouve dans les autres traités de géomancie
des correspondances dans divers domaines. Un Le présent article vise à proposer des hypothèses
autre type de correspondances peut interpeler sur la présence de ces figures géomantiques dans
pour notre propos. Le manuscrit de la BnF Arabe le manuscrit Arabe 3467 de la BnF et sur leur rap-
2631 est un traité de géomancie attribué à Idrīs port au contenu du texte25. Cependant, avant d’en-
contenant une « section sur la description des trer dans une telle analyse, il convient de présenter
figures du point de vue de la sincérité, de l’immo- une description globale de ces figures dans le
ralité, du mensonge, de l’hypocrisie, du mélange manuscrit. Bien entendu, cette description quan-
et du pur » ( faṣl fī ṣifat ashkāl min al-ṣidq wa-l-fisq titative sera limitée par plusieurs facteurs : l’encre
wa-l-kadhb wa -l-nifāq wa-l-mumtazij wa-l-ṭāhir)23. est parfois effacée, empêchant de voir avec certi-
Cette courte section passe en revue les seize tude certaines figures, et le manuscrit lui-même
figures : « Al-Aḥyān est sincère et pure, al-Inkīs a été endommagé et réparé en plusieurs endroits,
est immorale et menteuse, al-Bayāḍ est sincère permettant de supposer que certaines figures ont
et pure, al-Ḥumra est menteuse et hypocrite, disparu.
al-Qabḍ al-dākhil est est sincère et pure, al-Qabḍ Le manuscrit en lui-même est acéphale et ses
al-khārij est immorale et suspecte (murtāb), al- feuillets sont en désordre. Cent dix-neuf feuil-
Nuṣra al-dākhila est sincère et pure, al-Nuṣra lets en ont été préservés, dont vingt sont des
al-khārija est pure, Rāyat al-faraj est sin-
cère et pure, al-ʿAtaba al-khārija est immorale
24 Il s’agit probablement d’une erreur : il faut lire al-
l’indique, l’auteur y analyse l’influence de l’astrologie Shaqqāf, un des noms d’al-ʿUqla ou al-Iʿtikāf.
et sa place fondamentale dans le développement de la 25 Une grande partie des illustrations de ce manuscrit
géomancie. sont reproduites dans une réédition récente de la
22 Par convention, nous utilisons la majuscule lorsqu’il traduction d’André Miquel publiée initialement par
s’agit du signe zodiacal. Klincksieck en 1957. Voir ʿAbd Allāh b. al-Muqaffaʿ, Le
23 Pseudo-Idrīs, Traité de géomancie, Paris, BnF, Arabe Livre de Kalila et Dimna, trad. André Miquel (Paris :
2631, f.63. Klincksieck, 1957 ; réédition 2012).
70 Coulon
restaurations plus tardives26. Ses pages mesurent fois dont huit dans une illustration31, al-Qabḍ
30 cm sur 23 cm, chacune contenant une quin- al-dākhil ( ) vingt-huit fois dont six dans une
zaine de lignes (sauf les pages ornées d’une illustration32, al-Ijtimāʿ [= al-Rajāʾ] ( ) vingt-cinq
illustration). Son ancienne cote était « Suppl. ar. fois dont sept dans une illustration33, al-Laḥyān
no 1802 ». Un ex-libris nous informe qu’il s’agit [= al-Ḍāḥika] ( ) vingt-quatre fois dont huit
d’un ouvrage que possédait le prélat Henri-Charles dans une illustration34, al-ʿUqla [= al-Iʿtikāf]
du Camboust (1665-1732), duc de Coislin dès 1710 ( ) vingt-trois fois dont neuf dans une illustra-
et évêque de Metz de 1697 jusqu’à sa mort27. tion35, al-Ḥumra ( ) vingt fois dont six dans une
Comme l’indique un autre ex-libris28, ses manus- illustration36, al-Qabḍ al-khārij ( ) dix-sept fois
crits furent légués à la bibliothèque de l’abbaye de dont quatre dans une illustration37, al-Jamāʿa ( )
Saint-Germain-des-Prés à sa mort en 1732, et ont quatorze fois dont une dans une illustration38,
finalement intégré les fonds de la Bibliothèque al-Ṭarīq ( ) quatorze fois également dont quatre
nationale lors de la nationalisation des biens du dans une illustration39, al-Kawsaj [= al-Jawdala]
clergé à la Révolution française. ( ) dix fois dont une dans une illustration40,
Les figures géomantiques ont été exécutées à al-ʿAtaba al-khārija ( ) six fois dont deux dans une
l’encre rouge et sont, à de rares exceptions près29, illustration41, Naqī al-khadd ( ) six fois également
surmontées d’un trait rouge. À partir du f.3r, l’im- dont une dans une illustration42, al-Bayāḍ ( ) cinq
mense majorité des folios comporte au moins une
figure géomantique30. Il faut cependant noter que
l’encre de certaines figures s’est estompée avec le 31 Ff.3r, 12r, 15r, 18r, 22r, 23r, 27v, 31r, 32v, 33r, 34v, 37v, 41r,
41v, 43r, 43v, 46r, 46v, 47r, 50r, 53v, 55r, 55v, 58r, 59r, 62r,
temps et que le manuscrit lui-même a connu d’im- 63v, 65r, 65v, 67v, 69r, 71r, 74v, 79v, 81v, 83r, 86v, 87v, 89v,
portantes réparations. Ainsi, la plupart des folios 90v (deux fois), 100r, 104v, 106r, 111r, 112r, 112v, 114v (illus-
sans figure géomantique sont des folios endom- trations : ff.3r, 22r, 41v, 46r, 58r, 71r, 100r, 112v).
magés et dont une partie a été réparée a posteriori 32 Ff.12v, 13r, 14v, 19v, 23v, 36r, 38r, 39r, 40r, 42v, 51r, 58v, 62v,
66r, 68r, 70r, 73r, 89r, 90r, 91r, 91v, 93v, 97v, 98r, 98v, 106v,
avec des morceaux de papier. 108r, 110r (illustrations : ff.42v, 51r, 70r, 98v, 106v, 110r).
Sur les 119 folios du manuscrit, nous avons pu 33 Ff.4v, 10r, 14r, 20r, 21r, 26v, 28v, 35r, 49r, 54r, 54v (autre
repérer, par ordre décroissant du nombre d’oc- main ?), 56r, 57v, 70v, 71v, 73v, 82r, 84r, 85r, 88v, 93r, 100v,
currences, al-ʿAtaba al-dākhila ( ) quarante-huit 103r, 110r, 114r (illustrations : ff.21r, 26v, 28v, 56r, 85r,
93r, 110r).
34 Ff.3r, 3v, 5r, 10v, 11v, 13r, 21r, 26v, 31r, 43r, 53r, 60r, 63r, 78r,
82v, 83v, 85v, 95v, 102v, 103v, 105r, 106v, 113r, 118r (illustra-
26 Ff.1, 8, 12, 13, 18, 20, 27, 31, 35, 37, 38, 44, 64, 82-84, 113, 115, tions : ff.3r, 3v, 21r, 26v, 63r, 105r, 106v, 118r).
119. 35 Ff.3r, 20v, 36v, 51r, 57r, 59v, 63r, 64r, 67r, 72r, 76r, 80r, 80v,
27 Nous pouvons lire un bref portrait de ce personnage 81r, 84v, 85r, 89v, 90r, 99v, 101v, 105r, 115r, 117v (illustra-
dans l’ouvrage de René Kerviler, Le Chancelier Pierre tions : ff.3r, 51r, 63r, 76r, 80r, 85r, 99v, 105r, 117v).
Séguier, second protecteur de l’Académie française : 36 Ff.6r, 25v, 27r, 30r, 30v, 32r, 35v, 38v, 40v, 44v, 48r, 51v, 69r,
études sur sa vie privée, politique et littéraire, et sur le 73v, 76r, 78v, 86r, 107r, 113v, 116r (illustrations : ff.30r, 40v,
groupe académique de ses familiers et commensaux 76r, 78v, 86r, 107r).
(Paris : Didier, 1875), 634‑636. 37 Ff.22r, 45v, 52r, 55r, 57r, 59v, 66r, 68v, 72r, 78v, 80r, 80v,
28 Ex Bibliotheca MSS. Coisliniana, olim Segueriana, 92r, 105v, 110v, 115v, 117v (illustration : ff.22r, 78v, 80r,
quam Illust. Henricus Du Cambout, Dux De Coislin, 117v).
Par Franciæ, Episcopus Metensis, &c. Monasterio 38 Ff.6v, 11r, 16v, 22r, 37r, 55r, 67v, 68r, 75r, 77r, 87r, 96r, 102r,
S. Germani à Pratis legavit. An. M. DCC. XXXII. 104r (illustration : ff.22r).
29 Ff.59v, 67v, 80v, 90v, 98r. Il est à signaler qu’à de rares 39 Ff.13v (deux fois), 27v, 33r, 41v, 50v, 54v, 56r, 64v, 89v, 99v,
occasions deux figures se trouvent sous un même trait, 103v, 104v, 111v (illustrations : ff.41v, 56r, 64v, 99v).
comme au f.89v. 40 Ff.19r, 33v, 41v, 48v, 61v, 69v, 75v, 79r, 97r, 101r (illustra-
30 Les folios sans figure géomantique sont ff.4r, 5v, 7r-9v, tion : ff.41v).
17r-17v, 21v, 22v, 24r-25r, 29v, 39v, 42r, 47v, 60v, 61r, 66v, 41 Ff.15v, 34r, 97v, 98v, 99r, 108v (illustration : ff.15v et 98v).
88r, 94r, 94v, 95r, 96v, 109r, 109v, 116v, 117r, 118v. 42 Ff.16r, 45r, 54r, 72v, 76v, 92v (illustration : ff.16r).
Kalīla, Dimna et la géomancie 71
fois dont aucune dans une illustration43, Inlīs manuscrit. Les images sans figures géomantiques
[= al-Majūsī] ( ) quatre fois dont une dans une se trouvent généralement sur des feuillets partiel-
illustration44, al-Naṣra al-khārija ( ) une fois dans lement mutilés. Il est donc très probable que cer-
une illustration45 et enfin al-Naṣra al-dākhila taines de ces images aient été dotées d’une figure
( ) aucune fois. Cela fait deux cent quarante-cinq géomantique avant mutilation. Parmi les illustra-
figures identifiées dont cinquante-huit dans des tions du manuscrit, dix-sept possèdent une figure
illustrations. géomantique clairement identifiable52, dix-neuf
Ce total ne tient pas compte de quelques figures ont deux figures clairement identifiables53 et trois
mal exécutées ou partiellement illisibles, qu’il en ont trois54. Il est notable que les illustrations
n’est pas possible d’identifier clairement. C’est le comportent généralement plus de figures géoman-
cas de quelques figures : une figure au f.15v (Fig. tiques que les folios avec uniquement du texte, qui
3.11) dont seules deux des quatre rangées peuvent en contiennent le plus souvent une seule.
être identifiées donnant huit possibilités d’identi- Nous pouvons nous demander quand ces
fication46, une figure au f.26r comportant cinq ran- figures géomantiques ont été exécutées : ont-elles
gées donnant deux possibilités d’identification47, été faites dans la foulée de la copie du texte et de
une figure au f.28r qui a trois rangées avec quatre la composition des illustrations ? Ont-elles été
possibilités d’identification48, une figure illisible placées plus tard et, dans ce cas, à quel moment
au f.29r49, une figure au f.74r dont seule une ran- de l’histoire du manuscrit ? Des indices contradic-
gée n’est pas lisible permettant deux possibilités toires ne permettent pas d’apporter une réponse
d’identification50, une figure probablement effa- définitive. En effet, un certain nombre de répara-
cée aux ff.88r et 94v, trois figures aux ff.98r, 107v tions ont été faites après l’exécution des figures
et 110v comportant trois rangées avec deux possi- puisque les morceaux manquants nous empêchent
bilités d’identification51 et enfin une figure dans le de lire certaines rangées de points (par exemple
colophon au f.119r dont la première rangée a trois aux f.3v), ce qui suppose que la figure était entière
points. avant la réparation et que la mutilation est posté-
Des figures géomantiques peuvent être identi- rieure. Cependant, une figure (f.15r) a été exécutée
fiées dans quarante des cinquante illustrations du sur la réparation, ce qui suggérerait que cette figure
a été tracée plus tardivement, alors qu’elle est en
tous points similaire aux autres (et il semble bien
43 Ff.18v, 19r, 49v, 52v, 56v.
qu’il s’agisse de la même encre, du même calame
44 Ff.16r, 31v, 44r, 77v (illustration : ff.16r).
45 Ff.28v. et de la même main que les figures disparues suite
46 Les possibilités d’identification sont al-ʿAtaba al- à une mutilation). Des feuillets furent remplacés
dākhila ( ), al-Naṣra al-dākhila ( ), al-ʿUqla [= al-Iʿtikāf] à plusieurs périodes, et plusieurs feuillets refaits
( ), al-Ijtimāʿ [= al-Rajāʾ] ( ), Naqī l-khadd ( ), al-Bayāḍ qui remplacent les feuillets d’origine portent des
( ), Inlīs [= al-Majūsī] ( ) ou al-Qabḍ al-dākhil ( ).
47 Les possibilités d’identification sont al-Qabḍ al-khārij
figures géomantiques55. Il est donc probable que
( ) ou al-Ḥumra ( ). les figures aient été exécutées après la copie du
48 Les possibilités d’identification sont al-Bayāḍ ( ), texte, l’exécution des illustrations et le remplace-
al-Qabḍ al-khārij ( ), al-Qabḍ al-dākhil ( ) ou al-Ḥumra ment de certains feuillets, mais avant certaines
( ).
autres mutilations subies par le manuscrit. Elles
49 Dans ce cas, la présence d’une figure est hautement
probable car on trouve la ligne rouge surlignant quasi- ne faisaient donc sans doute pas partie du projet
ment toutes les figures du manuscrit.
50 Les possibilités d’identification sont al-Ṭarīq ( ) ou 52 Ff.30r, 40v, 42v, 46r, 58r, 64v, 67r, 70r, 71r, 74r, 86r, 93r,
al-Kawsaj [= al-Jawdala] ( ). 100r, 107r, 107v, 112v, 118r.
51 Les possibilités d’indentification sont, pour le f.98r, 53 Ff.3v (une seule visible), 16r, 21r, 26v, 28v, 51r, 56r, 63r,
al-Qabḍ al-dākhil ( ) ou al-Ḥumra ( ) ; pour le f.107v 76r, 78v, 80r, 85r, 98v, 99v, 105r, 106v, 110r, 110v, 117v.
al-Jamāʿa ( ) ou Inlīs [= al-Majūsī] ( ) ; et pour le f.110v 54 Ff.3r, 22r, 41v.
al-Qabḍ al-dākhil ( ) ou al-ʿAtaba al-dākhila ( ). 55 Par exemple ff.12, 12v, 13, 18v, 20, 21.
72 Coulon
initial du manuscrit mais, dans l’hypothèse d’une représentations si une légende ou le texte ne les
utilisation géomantique, elles auraient alors été accompagnaient pas. Au f.3v (Fig. 3.2), le roi et
ajoutées soit pour soutenir un usage divinatoire le philosophe sont représentés en train de discu-
déjà effectif, soit pour ouvrir la voie à un usage ter56. On voit la figure géomantique al-Laḥyān [=
divinatoire de ce manuscrit. al-Ḍāḥika] ( ). Une autre est partiellement cachée
par une réparation sur le feuillet, mais les trois pre-
mières lignes sont visibles : elle pourrait donc cor-
4 Analyse qualitative respondre à al-Ijtimāʿ [= al-Rajāʾ] ( ) ou al-ʿAtaba
l-dākhila ( ). Au f.106v, on retrouve le roi et le phi-
Dans la mesure où ce manuscrit de Kalīla losophe57, avec les figures al-Laḥyān [= al-Ḍāḥika]
wa-Dimna est un hapax pour ce qui est des figures ( ) et al-Qabḍ al-dākhil ( ). Au f.110r, la même scène
géomantiques présentes dans les marges et les se retrouve58, cette fois avec les figures al-Ijtimāʿ [=
images, il est difficile de tirer des conclusions al-Rajāʾ] ( ) et al-Qabḍ al-dākhil ( ).
définitives. Nous ne pouvons avancer que des Les figures géomantiques de la troisième illus-
hypothèses, notamment que les figures géoman- tration sont aisées à comprendre dans ce contexte.
tiques ne sont pas réparties arbitrairement dans En effet, al-Qabḍ al-dākhil est associée au signe
le manuscrit, mais peuvent être mises en rapport du Lion et à l’astre du Soleil. Ces deux éléments
avec son contenu, qu’il soit textuel ou figuratif. représentent traditionnellement en astrologie
Afin d’étayer cette hypothèse, nous concentre- le pouvoir et le roi. L’association des astres à des
rons notre propos sur les figures géomantiques fonctions au sein du pouvoir se retrouve dans
liées aux illustrations des manuscrits, soient les de nombreux écrits. L’astrologue Ibn al-Ḥātim
trente-neuf (sur cinquante) illustrations compor- (floruit première moitié du IVe/Xe siècle) faisait
tant une à trois figures géomantiques. Certaines du Soleil un émir (amīr), de Vénus un vizir (wazīr),
illustrations concernent les deux trames géné- de Mercure un messager, de la Lune un danseur,
rales de l’histoire de Kalīla wa-Dimna et mettent de Jupiter un juge, de Saturne un geôlier et de
en scène les principaux personnages de l’histoire : Mars un policier et un bourreau59. Plus tard, le
d’une part, le roi sassanide Chosroès et le méde- cosmographe al-Qazwīnī (m. 682H/1283) affir-
cin Burzūya qui lui raconte l’histoire de Kalīla et mait dans ses ʿAjāʾib al-makhlūqāt : « Les astrolo-
Dimna que le « prince des philosophes », Bidpaï, gues affirment que le Soleil est parmi les planètes
racontait lui-même au roi indien Dabshalīm ; comme le roi, et le reste des planètes est comme
d’autre part, l’intrigue de Kalīla et Dimna en tant les auxiliaires et les soldats. La Lune est comme le
que telle : Bankala, le roi lion, Shanzaba, le taureau vizir et l’héritier présomptif. Mercure est comme
ami du roi, Dimna, le chacal qui complote contre le secrétaire, Mars comme le chef de la police,
le taureau, Kalīla, l’ami de Dimna, qui désapprouve Jupiter comme le juge, Saturne comme le trésorier
cependant la cabale qu’organise son compère, la
panthère qui dévoile l’imposture de Dimna à la 56 Image consultable sur http://gallica.bnf.fr/ark:/12148/
mère du roi lion, etc. btv1b84152188/f18.item. Consulté le 20/12/2016.
57 Image consultable sur http://gallica.bnf.fr/ark:/12148/
btv1b84152188/f224.item. Consulté le 20/12/2016.
4.1 Le roi et le philosophe 58 Image consultable sur http://gallica.bnf.fr/ark:/12148/
Le motif du roi et de son conseiller est l’histoire btv1b84152188/f231.item. Consulté le 20/12/2016.
au premier plan diégétique. En effet, la trame de 59 Kristen Lippincott et David Pingree, « Ibn al-Hātim on
l’histoire est la narration par Burzūya au souve- the Talismans of the Lunar Mansions », Journal of the
Warburg and Courtauld Institutes 50 (1987) : 57‑81 sur-
rain Chosroès des histoires que Bidpaï contait tout p. 62 ; sur l’auteur voir aussi Kristen Lippincott,
à Dabshalīm. Ce faisant, ces deux niveaux de « More on Ibn Al-Hātim », Journal of the Warburg and
narration sembleraient identiques dans leurs Courtauld Institutes 51 (1988) : 188‑190.
Kalīla, Dimna et la géomancie 73
figure 3.2 Le roi et le philosophe, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 3v
pourrait donc tout à fait représenter le détenteur se trouve juste au-dessus de la représentation du
du pouvoir. philosophe conseillant le roi. On peut émettre
Quant à al-Ijtimāʿ, elle est représentée en astro- l’hypothèse que cette figure peut aussi être com-
logie par le signe du Cancer et la planète Mercure, prise comme une expression de l’archétype astro-
soit le « scribe (ou secrétaire) du roi62 ». Mercure logique du philosophe et du savant. Quant à la
correspond notamment à la philosophie parmi les deuxième figure géomantique dont une partie est
sciences63. On peut donc sans trop de risque avan- illisible, parmi les deux possibilités de lecture se
cer que les figures géomantiques de cette figure trouve al-Ijtimāʿ, représentant probablement le
correspondent aux archétypes astrologiques des philosophe, comme nous l’avons montré dans le
deux personnages représentés. cas de l’illustration du f.110r (Fig. 3.3).
Pour ce qui est de la première figure (f.3v), Enfin, la deuxième figure (f.106v) est également
elle est surmontée du titre du chapitre, en lettres surmontée d’un titre en lettres dorées, correspon-
dorées : « le chapitre de l’orfèvre et du mendiant » dant à l’histoire qui est racontée : « le chapitre du
(bāb al-ṣawwāgh wa-l-sāʾiḥ). La figure al-Laḥyān roi et de l’oiseau Fanza/Qubbira » (bāb al-malik
est associée au signe du Taureau et à l’astre Jupiter. wa al-ṭayr Fanza/Qubbira)66. Les deux figures
Or, parmi les disciplines correspondant à Jupiter géomantiques qui lui sont associées, al-Laḥyān
figurent la sagesse, la philosophie et l’herméneu- et al-Qabḍ al-dākhil, étaient toutes deux présentes
tique des rêves64. Jupiter est en outre selon l’abrégé dans une des deux figures analysées ci-dessus et
d’al-Madḫal (L’introduction) d’Abū Maʿšar al-Balḫī peuvent être en lien avec les figures du roi et du
(m. 272H/886) et selon ʿAbd al-Raḥmān al-Bisṭāmī philosophe.
(m. 858H/1454) l’astre des savants65. Cette figure
figure 3.3 Le roi et le philosophe, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 110r
Les cas des figures du roi et du conseiller nous ren- 4.2 La cour du roi lion
voient à la « royauté astrale ». Anna Caiozzo a mis Les dialogues entre les animaux de Kalīla
en évidence dans son travail l’importance de l’uti- wa-Dimna sont illustrés en plusieurs endroits.
lisation de l’astrologie dans la mise en scène du Ainsi, au f.3r (Fig. 3.4), nous avons une repré-
pouvoir67. La dimension astrologique des figures sentation du lion avec trois chacals68. Une ins-
géomantiques trouve donc ici une pleine réalisa- cription sur le côté gauche indique « Figure du
tion dans ce dialogue du roi et de son conseiller. lion, du chacal et […] » (ṣūrat al-sabʿ wa-ibn āwā
wa-al-[…]69). Trois figures géomantiques sont
représentées : al-Laḥyān [= al-Ḍāḥika] ( ), al-ʿUqla
[= al-Iʿtikāf] ( ) et al-ʿAtaba al-dākhila ( ). Au
67 Voir notamment Anna Caiozzo, Images du ciel d’Orient f.21r (Fig. 3.6), Kalīla et Dimna sont représentés
au Moyen Âge : une histoire du zodiaque et de ses repré-
sentations dans les manuscrits du Proche-Orient musul-
man (Paris : Presses de l’Université de Paris-Sorbonne,
2003) ; et Caiozzo, Réminiscences de la royauté cos- 68 Image consultable sur http://gallica.bnf.fr/ark:/12148/
mique dans les représentations de l’Orient médiéval (Le btv1b84152188/f17.item. Consulté le 20/12/2016.
Caire : Institut français d’archéologie orientale, 2011). 69 La fin n’est pas lisible.
76 Coulon
en train de discuter70. Les figures géomantiques illustration relevant du même groupe, avec toute-
correspondantes sont al-Laḥyān [= al-Ḍāḥika] fois une réserve sur ses interprétations possibles.
( ) et al-Ijtimāʿ [= al-Rajāʾ] ( ) . Au f.22r, Dimna est On remarque que certaines figures sont plus
représenté face au lion, à côté duquel se trouve abondamment représentées que d’autres. Ainsi,
un autre lion (ṣifat Dimna qad dakhala ʿalā l-asad al-ʿAtaba al-dākhila ( ) est représentée sur cinq des
waḥawla-hu asad ākhar)71. Les figures géoman- huit illustrations. Or, ces scènes montrent des dis-
tiques correspondantes sont al-Jamāʿa ( ), al-ʿA- cussions. Les trois images qui n’ont pas cette figure
taba al-dākhila ( ) et al-Qabḍ al-khārij ( ), qui n’est sont une discussion de Kalīla et Dimna, le com-
pas surmontée d’un trait rouge. Au f.40v (Fig. 3.8), bat du lion et du taureau, et le procès de Dimna.
le roi lion est représenté en train de tuer le tau- Le nom de cette figure géomantique signifie « le
reau, avec la figure géomantique al-Ḥumra ( )72. seuil/linteau intérieur ». Le terme de ʿataba peut
Au f.41v (Fig. 3.7), Kalīla et Dimna discutent, avec aussi désigner une difficulté, un obstacle, un vice
al-Kawsaj [= al-Jawdala] ( ), al-Ṭarīq ( ) et al-ʿAtaba ou un désagrément, ainsi qu’un cas ou une affaire.
al-dākhila ( )73. Au f.46r (Fig. 3.5), le lion écoute Quant au verbe ʿataba, il signifie « se fâcher, être
sa mère, et on peut observer al-ʿAtaba al-dākhila en colère ; réprimander, faire des reproches à
( )74. Au f.56r est représentée une scène du pro- quelqu’un ». On peut ainsi assimiler cette figure à
cès de Dimna, avec le juge (al-qāḍī), le panthère deux significations : d’une part l’affaire interne, qui
(al-nimr), le garde (al-jund) et Dimna75. Les figures se déroule à huis clos, derrière les portes (ici les
sont al-Ijtimāʿ [= al-Rajāʾ] ( ) et al-Ṭarīq ( ). Enfin, portes du pouvoir), qui est un motif très présent
au f.100r, un lion discute avec un chacal (ibn āwā), dans l’histoire ; d’autre part, la réprimande contre
avec al-ʿAtaba al-dākhila ( )76. Cependant, dans ce Dimna et son procès ou contre le lion qui retarde
dernier cas, l’interprétation est plus ambigüe : il la punition et le châtiment du chacal. Dans le pre-
s’agit d’un autre conte mettant en scène un lion et mier cas, le motif du « huis clos », qu’il s’agisse des
un chacal, mais où le chacal est pieux et conseille discussions à la dérobée de Dimna ou du procès
le roi avant d’être victime d’une cabale des cour- qui lui est intenté et auquel ne participent que
tisans, jaloux de la faveur qu’il obtint auprès du des personnages de la cour du roi. Une phrase
lion. Le chacal de ce conte est donc totalement à illustre bien ce motif dans la fable lorsque la pan-
l’opposé de Dimna (peut-être peut-on l’assimiler à thère surprend Dimna en train de dévoiler son
un conte alternatif où c’est Kalīla qui conseille le complot à Kalīla : « Comme elle arrivait à la porte,
lion). En revanche, d’un point de vue strictement elle entendit Kalīla qui s’en prenait à Dimna et lui
iconographique, rien ne permet de distinguer reprochait ce qu’il avait fait » (« fa-lammā intahā
cette illustration de celles de l’histoire de Kalīla ilā l-bāb samiʿa Kalīla yuʿātibu Dimna ʿalā mā kāna
et Dimna. On peut donc l’ajouter comme une min-hu77 […] »). Ainsi, il s’agit de la seule figure
présente dans l’illustration de la mère du lion qui
70 Image consultable sur http://gallica.bnf.fr/ark:/12148/
lui reproche de ne pas prendre immédiatement les
btv1b84152188/f53.item. Consulté le 20/12/2016. mesures contre Dimna, laissant planer le doute sur
71 Image consultable sur http://gallica.bnf.fr/ark:/12148/ sa capacité à rendre la justice. De même, il s’agit
btv1b84152188/f55.item. Consulté le 20/12/2016. de la seule figure accompagnant le lion conseillé
72 Image consultable sur http://gallica.bnf.fr/ark:/12148/
par l’honnête chacal à l’abri des courtisans, qui en
btv1b84152188/f92.item. Consulté le 20/12/2016.
73 Image consultable sur http://gallica.bnf.fr/ark:/12148/ conçoivent de la jalousie. Cette figure géomantique
btv1b84152188/f94.item. Consulté le 20/12/2016.
74 Image consultable sur http://gallica.bnf.fr/ark:/12148/
btv1b84152188/f103.item. Consulté le 20/12/2016. 77 Éd. Silvestre de Sacy, 136 ; éd. Cheikho, 102 ; éd. Būlāq,
75 Image consultable sur http://gallica.bnf.fr/ark:/12148/ 154 ; éd. al-Marṣafī, 246 ; éd. ʿAzzām, 133 ; trad. Miquel,
btv1b84152188/f123.item. Consulté le 20/12/2016. 101. On remarquera que la phrase contient à la fois le
76 Image consultable sur http://gallica.bnf.fr/ark:/12148/ nom « porte » (bāb) et le verbe « gronder, réprimander,
btv1b84152188/f211.item. Consulté le 20/12/2016. reprocher » (ʿātaba).
Kalīla, Dimna et la géomancie 77
figure 3.4 Le lion et trois chacals, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 3r
représente astrologiquement la Balance et Jupiter, Dimna, les deux figures, al-Laḥyān et al-Ijtimāʿ,
soient la constellation de la justice et l’astre repré- sont associées à Jupiter et Mercure. Jupiter peut
sentant le juge. représenter Kalīla : n’est-il pas l’astre de « la jus-
Les autres figures géomantiques de ce groupe tice, l’équilibre, la dignité, la chasteté, la véracité,
d’illustrations sur la cour du roi lion pose quelques la probité, ce qui est digne de confiance, la réserve,
difficultés d’interprétation. En effet, il est diffi- la dévotion, la piété78 »? Kalīla est en effet le
cile d’y trouver une unité d’interprétation, même
si plusieurs pistes peuvent être explorées parmi 78 Maslama al-Qurṭubī (pseudo-Majrīṭī), Das Ziel des
Weisen, 199 ; Chébiri, « Le Livre du but du sage : pré-
leurs potentialités. Tout d’abord, il y a une certaine
éminence qu’il convient d’accorder à l’un des deux
cohérence astrologique. Nous avons vu la préé- plus dignes résultats », 158. On trouve des données
minence d’al-ʿAtaba al-dākhila et donc de Jupiter. similaires sur les vertus associées à Jupiter dans Abū
Cette planète est en outre celle d’al-Laḥyān, pré- Maʿshar al-Balkhī, The Abbreviation of the Introduction
sente dans deux des figures. L’une d’elle (f.3r ; Fig. to Astrology, 60-63 ; Abū al-Saqr ʿAbd al-Aziz b. ʿUthmān
al-Qabīṣī, The Introduction to Astrology. Editions of the
3.4) représente le lion avec trois chacals, mais Arabic and Latin texts and an English translation, éd. et
l’autre une discussion de Kalīla et Dimna (f.21r ; trad. Charles Burnett, Keiji Yamamoto et Michio Yano
Fig. 3.6). Concernant cette discussion de Kalīla et (Londres : The Warburg Institute, 2004), 64-69.
78 Coulon
figure 3.5 Le lion et sa mère, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 46r
garde-fou de Dimna, qui le sermonne chaque fois l’influençabilité, à la générosité, au travail des dif-
que l’aigrefin lui fait part de ses projets machia- férents arts, à la dextérité dans toutes choses, au
véliques. En revanche, Dimna pourrait bien être désir de toute opération subtile, à l’ensemble des
assimilé aux aspects négatifs de Mercure. Ainsi, biens et leur corruption, à la patience, à l’aide et
dans la Ghāyat al-ḥakīm de Maslama al-Qurṭubī, à l’abstention de faire le mal79 ». Beaucoup de ces
ces aspects moins connus des astres sont abor- vices peuvent tout à fait caractériser Dimna. On
dés par exemple dans le chapitre 7 du livre III. retrouve la figure d’al-Ijtimāʿ dans la représenta-
Mercure se voit alors assimilé aux qualités propres tion du procès de Dimna (f.56r).
aux manipulateurs, aux escrocs, aux charlatans ou
aux voleurs : il est associé « à la ruse, à la trompe- 79 Maslama al-Qurṭubī (pseudo-Majrīṭī), Das Ziel des
rie, à la fraude, à la dissimulation, à la rancœur, au Weisen, 201 ; Chébiri, « Le Livre du but du sage : préé-
mensonge, à la profondeur de réflexion qui rend minence qu’il convient d’accorder à l’un des deux plus
dignes résultats », 160. Voir aussi sur les sciences et qua-
inconnaissable ce qu’il y a en son for intérieur, à lités attribuées à Mercure Abū Maʿshar al-Balkhī, The
l’emploi d’écrits falsifiés, à l’adversité, à la peur Abbreviation of the Introduction to Astrology, 64-67 ;
des ennemis, à la rapidité dans les opérations, à al-Qabīṣī, The Introduction to Astrology, 78-81.
Kalīla, Dimna et la géomancie 79
figure 3.6 Kalīla et Dimna discutent, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 21r
figure 3.7 Kalīla et Dimna discutent, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 41v
dignes résultats », 160‑161. Le mensonge (al-kadhb) fait 83 Coulon, « La magie islamique et le “corpus bunianum”
également partie des attributs de la Lune dans Abū au Moyen Âge », vol. 2, 199.
Maʿshar al-Balkhī, The Abbreviation of the Introduction 84 Image consultable sur http://gallica.bnf.fr/ark:/12148/
to Astrology, 66-67. btv1b84152188/f183.item. Consulté le 20/12/2016.
Kalīla, Dimna et la géomancie 81
figure 3.8 Le lion tue le taureau, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 40v
deux trames diégétiques principales. Qu’en est-il entreposent du sésame dans une même maison.
des fables de moindre taille ? L’un des deux essaie de dépouiller l’autre à son
propre profit avec l’aide d’un voleur. L’associé
4.3 Les fables métadiégétiques malhonnête promet donc au voleur une part du
Il serait trop long d’analyser ici une à une cha- sésame volé et lui indique qu’il doit dérober le tas
cune des illustrations des fables racontées par les de sésame sur lequel se trouvera un manteau qu’il
différents protagonistes de l’histoire de Kalīla et aura lui-même posé. L’associé honnête, voyant le
Dimna. Nous pouvons cependant en décrire une manteau, pense que son compère a voulu ainsi
sélection afin d’observer s’il est possible de déga- protéger son tas de sésame et estime que son hon-
ger une certaine cohérence entre les illustrations nêteté doit lui être rendue en mettant le manteau
et les figures géomantiques associées. sur les graines de son partenaire. Le collaborateur
La fable du vendeur de sésame et son asso- malhonnête s’aperçoit donc le lendemain que
cié est illustrée au f.26v (Fig. 3.9) avec les figures c’est son propre sésame qui a été dérobé et qu’il
al-Ijtimāʿ [= al-Rajāʾ] ( ) et al-Laḥyān [= al-Ḍāḥika] en aura ainsi perdu une partie. Au moins une des
( )85. L’histoire met en scène deux associés qui deux figures trouve pleinement sens dans l’his-
toire : al-Ijtimāʿ, soit « l’assemblée, la réunion »,
85 Image consultable sur http://gallica.bnf.fr/ark:/12148/ dans la mesure où il s’agit de l’histoire d’asso-
btv1b84152188/f64.item. Consulté le 20/12/2016. ciés ayant réuni leurs avoirs dans un même lieu.
82 Coulon
figure 3.9 L’associé malhonnête et le voleur, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 26v
Al-Ijtimāʿ correspond en outre à Mercure, astre du L’histoire du pauvre et du voleur est figurée au
commerce86. Quant à al-Laḥyān, cette figure cor- f.28v (Fig. 3.10) avec les figures al-Naṣra al-khārija
respond à Jupiter, soit le juge des astres, en charge ( ) et al-Ijtimāʿ [= al-Rajāʾ] ( )88. Un voleur entre
de la justice. Cela peut correspondre au sens de dans la demeure d’un homme pauvre qui avait
l’histoire, dans lequel la traîtrise se retrouve punie passé sa journée à mendier un manteau en vain.
par l’honnêteté sincère du partenaire loyal. Jupiter Alors qu’il dormait, le voleur entra dans sa demeure
est également en charge des « commerces exempts et tenta de voler le seul bien qui restait : une jarre
de fraudes »87. de blé. Le voleur déposa son manteau afin d’y
mettre le blé et de pouvoir partir avec. L’homme
86 Maslama al-Qurṭubī (pseudo-Majrīṭī), Das Ziel des
Weisen, 154‑155 ; Chébiri, « Le Livre du but du sage : plus dignes résultats », 125‑126 ; Coulon, « Fumigations
prééminence qu’il convient d’accorder à l’un des deux et rituels magiques. Le rôle des encens et fumiga-
plus dignes résultats », 128‑129 ; Coulon, « Fumigations tions dans la magie arabe médiévale », 210. Voir aussi
et rituels magiques. Le rôle des encens et fumigations sur les vertus de Jupiter Abū Maʿshar al-Balkhī, The
dans la magie arabe médiévale », 215. Abbreviation of the Introduction to Astrology, 60-63 ;
87 Maslama al-Qurṭubī (pseudo-Majrīṭī), Das Ziel des al-Qabīṣī, The Introduction to Astrology, 64-69.
Weisen, 151‑152 ; Chébiri, « Le Livre du but du sage : 88 Image consultable sur http://gallica.bnf.fr/ark:/12148/
prééminence qu’il convient d’accorder à l’un des deux btv1b84152188/f68.item. Consulté le 20/12/2016.
Kalīla, Dimna et la géomancie 83
figure 3.10 Le pauvre surprend le voleur, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 28v
pauvre, réveillé par le voleur, avait continué à faire ou la victoire extérieure. En effet, naṣra désigne
semblant de dormir et surprit le voleur alors qu’il l’aide, le secours, et plus particulièrement l’aide
remplissait le manteau. Le voleur s’enfuit et aban- divine qui mène à la victoire. Dans le cas présent,
donna ainsi son manteau89. Dans le cas présent, il la présence du voleur dans la maison est un évé-
est plus difficile d’interpréter ces figures géoman- nement inattendu pour obtenir ce manteau tant
tiques correspondant respectivement au Soleil et à attendu. Il est à noter que le Soleil représente
Mercure, ou au Taureau et au Cancer. La première aussi le roi ou le sultan des astres et représente la
interprétation que nous pourrions formuler pour Toute-puissance divine. Quant à al-Ijtimāʿ, nous
al-Naṣra al-khārija tient d’abord à son nom : l’aide avons vu plus haut ce que peuvent représenter
ses aspects négatifs, qui comprennent les quali-
89 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, éd. Silvestre de Sacy,
tés requises pour un escroc, un charlatan ou un
51-52 ; éd. Cheikho, 49 ; éd. al-Marṣafī, 134-135 ; éd. voleur : elle pourrait représenter le brigand ou à la
ʿAzzām, 42-43 ; trad. Miquel, 28‑31. scène du cambriolage nocturne, dans laquelle le
84 Coulon
voleur fait preuve de dextérité avant de fuir lâche- À la page suivante du manuscrit est représentée
ment quand sa victime se relève, mais aussi dans une autre anecdote du voyage du moine94. Cette
laquelle le pauvre homme dupe le voleur en lui anecdote fait immédiatement suite à l’arrivée du
faisant croire qu’il est endormi et en se levant au moine en ville : il est hébergé dans une maison
moment idoine. close, où une prostituée est éperdument amou-
Le récit du moine observant deux bouquetins reuse d’un jeune homme. La patronne souhaite
se battre est représenté au f.15v (Fig. 3.11) avec la donc assassiner l’amant. Elle s’arrange alors pour
figure al-ʿAtaba al-khārija ( ) et une autre mal exé- qu’il s’endorme après lui avoir fait consommer du
cutée, pouvant être interprétée de quatre façons vin pur, puis glisse du poison dans un roseau pour
différentes90. L’histoire est des plus courtes et se le lui souffler dans l’anus. Le jeune pète cependant,
situe sur le chemin d’un moine : « Il rencontra projetant le poison dans la gorge de la femme, qui
en chemin deux bouquetins qui se battaient à meurt sur le coup95. L’anecdote est représentée
coups de cornes et dont le sang coulait. Survint un avec les figures Naqī al-khadd ( ) et al-Inkīs [=
renard qui se mit à le laper ; les deux bouquetins al-Majūsī] ( ) (Fig. 3.12). Encore une fois, nous pou-
se retournèrent alors contre lui et le tuèrent. Le vons esquisser une correspondance astrologique
moine poursuivit son chemin et parvint le soir à la entre les figures géomantiques et l’illustration. En
ville91. » Al-ʿAtaba al-khārija correspond parmi les effet, Naqī al-khadd (littéralement « pur de joue »,
signes zodiacaux au Scorpion et parmi les astres à ce qui pourrait aussi évoquer les atours physiques
la queue du Scorpion (al-Dhanb). Or, le Scorpion espérés pour une marchande d’amour) corres-
est la constellation de Mars par excellence, l’astre pond à la planète Vénus, alors que la scène montre
du sang et de la guerre92. La couleur associée est bien le couple du jeune homme et de la prostituée
le rouge, soit la couleur du sang et de la robe du amoureuse allongés sur la couche. Al-Inkīs, à l’in-
renard lapant le sang des bouquetins. Les animaux verse, est en rapport avec Saturne, planète néfaste
associés à Mars sont également d’une manière et funeste, qui pourrait donc bien correspondre à
générale les animaux de couleur rouge, ainsi que l’empoisonneuse.
« toute bête sauvage qui a des griffes et des canines L’histoire du lièvre et du lion96 est représentée
et qui est agressive »93. On peut donc voir ici une au f.30r (Fig. 3.13) avec la figure al-Ḥumra ( )97.
cohérence dans la symbolique de la figure géo- Un lièvre élabore une ruse pour débarrasser une
mantique et la scène représentée. contrée d’un lion tyrannique : il fait croire au lion
qu’un autre lion se trouve dans les parages et l’amène
à un puits au fond duquel le lion voit son propre
reflet. Il se jette alors dessus pour tuer ce qu’il croit
être un adversaire et tombe au fond du puits. Comme
90 Image consultable sur http://gallica.bnf.fr/ark:/12148/
nous l’avions vu, al-Ḥumra est associée à Mars et
btv1b84152188/f42.item. Consulté le 20/12/2016. c’est la figure qui accompagnait la représentation
91 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, éd. Silvestre de
Sacy, 94-95 ; éd. Cheikho, 65 ; éd. ʿAzzām, 94-97 ; trad.
Miquel, 68‑70. 94 Image consultable sur http://gallica.bnf.fr/ark:/12148/
92 Voir par exemple Abū Maʿshar al-Balkhī, The Abbrevia- btv1b84152188/f43.item. Consulté le 20/12/2016.
tion of the Introduction to Astrology, 62-63 ; al-Qabīṣī, 95 Ibn al-Muqaffaʿ, Kalīla wa-Dimna, éd. Silvestre de Sacy,
The Introduction to Astrology, 68-71. 95-96 ; éd. Cheikho, 65-66 ; éd. ʿAzzām, 94-97 ; trad.
93 Maslama al-Qurṭubī (pseudo-Majrīṭī), Das Ziel des Miquel 70.
Weisen, 152 ; Chébiri, « Le Livre du but du sage : préé- 96 Ibn al-Muqaffaʿ, éd. de Sacy, 103-105 ; éd. Cheikho,
minence qu’il convient d’accorder à l’un des deux plus 72-73 ; éd. Būlāq, 116-119 ; éd. al-Marṣafī, 202-203 ; éd.
dignes résultats », 126 ; Coulon, « Fumigations et rituels ʿAzzām, 102 ; trad. Miquel 75‑76.
magiques. Le rôle des encen et fumigations dans la 97 Image consultable sur http://gallica.bnf.fr/ark:/12148/
magie arabe médiévale », 212. btv1b84152188/f71.item. Consulté le 20/12/2016.
Kalīla, Dimna et la géomancie 85
figure 3.11
Le moine et les deux
bouquetins tuant un
renard dans leur combat,
Ibn al-Muqaffaʿ, Kalīla
wa-Dimna, Paris, BnF,
Arabe 3467, f. 15v
figure 3.12
L’empoisonneuse à la
sarbacane, Ibn al-Muqaffaʿ,
Kalīla wa-Dimna, Paris, BnF,
Arabe 3467, f. 16r
86 Coulon
figure 3.13 Le lion, prêt à se jeter sur son propre reflet, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 30r
du lion qui se jette sur le taureau pour le tuer. Nous présentes sur cette image : la seconde n’est cepen-
avons là un exemple de concordance claire entre dant pas clairement identifiable, deux des quatre
deux scènes similaires aux protagonistes diffé- rangées de points n’étant pas clairement exécutées,
rents, mais avec une même figure géomantique ce qui laisse quatre possibilités d’interprétation.
en cohérence avec le contenu de l’illustration. L’histoire raconte que deux singes voyant un ver
C’est encore al-Ḥumra ( ) qui accompagne l’il- luisant crurent qu’ils pourraient allumer un feu de
lustration de l’histoire des singes et du ver luisant au bois avec ce qu’ils pensaient être une étincelle. Un
f.42v (Fig. 3.14) 98. Deux figures géomantiques sont oiseau voulut leur expliquer qu’il ne leur serait pas
possible d’allumer un feu puisqu’il ne s’agissait pas
98 Image consultable sur http://gallica.bnf.fr/ark:/12148/ de feu comme ils le croyaient. Cependant, malgré
btv1b84152188/f96.item. Consulté le 20/12/2016. l’avertissement d’un homme qui lui indiqua que
Kalīla, Dimna et la géomancie 87
figure 3.14 Un singe se saisit de l’oiseau pour le jeter à terre, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe
3467, f. 42v
leur expliquer leur tort était peine perdue, l’oiseau signe du Taureau et la planète Jupiter, font diffici-
s’approcha des singes, mais l’un d’eux le saisit et le lement sens par rapport à l’histoire racontée. Deux
jeta à terre, le tuant sur le coup99. La violence de la éléments de la description de Jupiter par Maslama
mort de l’oiseau correspond bien sûr tout à fait à la al-Qurṭubī peuvent s’y rattacher toutefois : d’une
symbolique martiale de la figure al-Ḥumra. part, Jupiter est l’astre à qui il « appartient de faci-
Au f.63r (Fig. 3.15) est représentée la scène d’un liter l’obtention des demandes », or, la souveraine
groupe de colombes piégées dans un filet recevant des colombes demande l’aide du rat pour obte-
l’aide d’un rat pour se libérer du piège sous le regard nir la libération de ses sujets du filet du chasseur,
d’un corbeau100. Deux figures se trouvent sur cette d’autre part, à Jupiter correspond « tout oiseau de
illustration : al-ʿUqla [= al-Iʿtikāf] ( ) et al-Laḥyān bonne apparence et beau comme les paons, les
[= al-Ḍāḥika] ( ). Al-ʿUqla est aisément compré- coqs, les colombes sauvages, le francolin »101. Les
hensible : le nom de cette figure signifie « lien, colombes prises au piège peuvent donc aussi tout
nœud, entrave » et peut tout à fait désigner le filet à fait être assimilées à cette figure géomantique.
dans lequel sont piégées les colombes. La seconde, Les « figures géomantiques » que nous avons ici
al-Laḥyān, n’offre pas d’interprétation aussi évi- analysées comportent donc un lien avec les illus-
dente. Sa symbolique astrologique, c’est-à-dire le trations dans lesquelles elles se trouvent.
99 Ibn al-Muqaffaʿ, Le Livre de Kalila et Dimna, trad. 101 Maslama al-Qurṭubī (pseudo-Majrīṭī), Das Ziel des
Miquel, 93‑94. Weisen, 151‑52 ; Chébiri, « Le Livre du but du sage :
100 Ibn al-Muqaffaʿ, Le Livre de Kalila et Dimna, trad. prééminence qu’il convient d’accorder à l’un des deux
Miquel 127‑28. Image consultable sur http://gallica plus dignes résultats », 125‑126 ; Coulon, « Fumigations
.bnf.fr/ark:/12148/btv1b84152188/f137.item. Consulté le et rituels magiques. Le rôle des encens et fumigations
20/12/2016. dans la magie arabe médiévale », 210.
88 Coulon
figure 3.15 Le rat délivre les colombes du filet, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3467, f. 63r
Qu’en est-il d’un éventuel usage divinatoire Ibn al-Muqaffaʿ, ʿAbd Allāh. Kalīla wa-Dimna. Paris,
du livre ? Celui-ci ne saurait être prouvé en l’état Bibliothèque nationale de France, Arabe 3467.
actuel de nos connaissances, mais l’analyse des Pseudo-Idrīs. Traité de géomancie. Paris, Bibliothèque
« figures géomantiques » met en évidence une nationale de France, Arabe 2631.
certaine logique dans leur présence et leur répar-
tition : celles-ci sont en rapport avec les figures ou Sources
le texte à côté desquels elles se trouvent et visent Abū Maʿshar al-Balkhī, Jaʿfar b. Muḥammad. The
à représenter leur idée avec un symbole à carac- Abbreviation of the Introduction to Astrology Together
tère divinatoire. Dans le cas présent, contraire- with the Medieval Latin Translation of Adelard of
ment à des ouvrages de bibliomancie comme les Bath. Édité et traduit par Charles Burnett, Keiji
Fāl-Nāma persans du Dīvān de Ḥāfiẓ, il s’agit d’un Yamamoto et Michio Yano. Leiden : Brill, 1994.
recueil de fables animalières ; or, l’apologue peut Chébiri, Valérie. « Le Livre du but du sage : prééminence
aisément se prêter à la divination. En effet, l’apo- qu’il convient d’accorder à l’un des deux plus dignes
logue est un « court récit imaginaire ou parfois réel résultats ». Thèse de doctorat, École pratique des
dont se dégage une vérité morale103 ». La dimen- hautes études. Section des sciences religieuses, 1998.
sion allégorique, tout à fait assumée dans l’intro- Ibn al-Muqaffaʿ, Calila et Dimna ou fables de Bidpai.
duction de Kalīla et Dimna, et l’usage de symboles Édité par Antoine-Isaac Silvestre de Sacy. Paris :
permet une multiplicité des signifiants, bien qu’en Imprimerie royale, 1816.
général le récit débouche sur un seul précepte Ibn al-Muqaffaʿ, La version arabe de Kalîlah et Dimnah
moral. Un riche potentiel de significations autour d’après le plus ancien manuscrit arabe daté. Édité par
d’une ligne directrice est un des fondements des Louis Cheikho. Beyrouth : Imprimerie Catholique,
symboles des pratiques divinatoires, à l’instar des 1905.
poèmes de Ḥāfiẓ. L’hypothèse d’un usage divina- Ibn al-Muqaffaʿ, Kalīla wa-Dimna. Le Caire : Būlāq, 1937.
toire de ce manuscrit de Kalīla et Dimna est donc Ibn al-Muqaffaʿ, Kalīla wa-Dimna. Édité par Muḥammad
tout à fait possible, bien qu’il faille demeurer pru- Ḥasan Nāʾil al-Marṣafī. Le Caire : Maṭbaʿat Muṣṭafā
dent en l’absence d’autres indices probants tels Muḥammad, 19335.
qu’un texte qui témoignerait qu’une telle pratique Ibn al-Muqaffaʿ, Kalīla wa-Dimna. Édité par ʿAbd
existait ou d’autres manuscrits du même genre. al-Wahhāb ʿAzzām. Beyrouth : Dār al-shurūq-al-
Sharika al-waṭaniyya, 1981.
Ibn al-Muqaffaʿ, Le Livre de Kalila et Dimna. Traduit par
Bibliographie André Miquel. Paris : Klincksieck, 2012.
(Pseudo-)Majrīṭī, Abū al-Qāsim Maslama b. Aḥmad
Manuscrits al-Faraḍī al- [= Maslama al-Qurṭubī]. Das Ziel
Kitāb al-raml. Paris, Bibiothèque nationale de France, des Weisen. Édité par Hellmut Ritter. Berlin : B.G.
Arabe 2731. Teubner, 1933.
al-Bisṭāmī, ʿAbd al-Raḥmān. Rashḥ adhwāq al-ḥikma al-Qabīṣī, Abū al-Saqr ʿAbd al-Aziz b. ʿUthmān. The
al-rabbāniyya fī sharḥ awfāq al‐lumʿa al‐nūrāniyya. Introduction to Astrology. Editions of the Arabic and
Istanbul, Süleymaniye Library, Kadizade Mehmed Latin texts and an English translation. Édité et tra-
333. duit par Charles Burnett, Keiji Yamamoto et Michio
Ibn al-Muqaffaʿ, ʿAbd Allāh. Kalīla wa-Dimna. Munich, Yano. Londres : The Warburg Institute, 2004.
Bayerische Staatsbibliothek, Cod. Arab. 616. al-Qazwīnī, Zakariyā b. Muḥammad. Kosmographie. 1,
Die Wunder der Schöpfung : Aus den Handschriften
der Bibliotheken zu Berlin, Gotha, Dresden und
103 Définition du Centre national des Ressources tex-
tuelles et lexicales. Voir http://www.cnrtl.fr/definition/ Hamburg. Édité par Ferdinand Wüstenfeld.
apologue, consulté le 20/12/2016. Göttingen : Dieterich, 1849.
90 Coulon
al-Zanātī, Muḥammad. al-Faṣl fī uṣūl ʿilm al-raml. Le Coulon, Jean-Charles. « Fumigations et rituels magiques.
Caire : Maṭbaʿat al-jumal al-miṣriyya, 1927. Le rôle des encens et fumigations dans la magie
arabe médiévale ». Bulletin d’études orientales, n° 64
Études (2016) : 179-248.
Abbès, Makram. « L’ami et l’ennemi dans Kalila et Coulon, Jean-Charles. « La magie islamique et le “cor-
Dimna ». Bulletin d’études orientales, n°57 (2008) : pus bunianum” au Moyen Âge ». Thèse de doctorat,
11-41. Université Paris-Sorbonne, 2013.
Alilouche, Sabrina, et Ghazaleh Esmailpour Qouchani. Descola, Philippe. Par-delà nature et culture. Paris :
« Les gloses marginales et le Fālnāma du coran de Gallimard, 2005.
Gwalior, témoignages des usages multiples du coran Dozy, Reinhart. Supplément aux dictionnaires arabes.
dans l’Inde des sultanats ». In Le Coran de Gwalior : 2 vol. Leiden : Brill, 1967.
polysémie d’un manuscrit à peintures. Édité par Fahd, Toufiq. « La magie comme “source” de la sagesse
Éloïse Brac de la Perrière et Monique Burési, 85-110. d’après l’œuvre d’al-Būnī », Res Orientales, n° 14
Paris : Éditions de Boccard, 2016. (2002), 61-108.
Belot, Jean-Baptiste. Al-Farāʾid al-durriyya. ʿArabī – Fahd, Toufiq. La Divination arabe : études religieuses,
faransī. Beyrouth : Dār al-Mashriq, 1971. sociologiques et folkloriques sur le milieu natif d’Is-
Ben Choaib, Aboubekr A. « La bonne aventure chez lam. Paris : Sindbad, 1987.
les musulmans du Moghrib ». Revue africaine, n° 50 Fahd, Toufiq. « Khaṭṭ ». In Encyclopaedia of Islam,
(1906) : 62-71. Second Edition. Édité par Peri Bearman et al. Leiden :
Benhamouda, Ahmed. « Les noms arabes des étoiles : Brill, 1960-2007.
essai d’identification ». Annales de l’Institut d’Études Farhad, Massumeh, et Serpil Bağcı. Falnama : the book
Orientales, n° 9 (1951) : 76-210. of omens. Londres : Thames & Hudson, 2009.
Binsbergen, Wim van. « The Astrological Origin of Ferrand, Gabriel. « Un chapitre d’astrologie arabico-
Islamic Geomancy ». Communication présentée à la malgache ». Journal asiatique, 10e série, t. 6 (1905) :
15e conférence annuelle de la SSIPS (Society for the 193-273.
Study of Islamic Philosophy and Science) et SAGP Fronval, Laetitia. « Islam et bibliomancie : pratiques
(Society of Ancient Greek Philosophy), Binghamton, contemporaines de l’estekhâreh en Iran », Archives
octobre 1996. https://www.labirintoermetico.com/ de sciences sociales des religions, 189 (2020) : 73-93.
07Geomanzia/Binsbergen_van_W_The_astrologi Fronval, Laetitia. « La pratique du fāl-e Ḥāfeẓ dans l’Iran
cal_origin_of_Islamic_Geomancy.pdf. contemporain. Science occulte en mutation ? », dans
Bouché-Leclercq, Auguste. Histoire de la divination Magie et sciences occultes dans le monde islamique,
dans l’Antiquité : divination hellénique et divination dir. Jean-Charles Coulon (Marseille : Diacritiques
italique. Grenoble : J. Millon, 2003. Éditions, 2021).
Caiozzo, Anna. Images du ciel d’Orient au Moyen âge : Gardiner, Noah. « Esotericist Reading Communities and
une histoire du zodiaque et de ses représentations the Early Circulation of the Sufi Occultist Aḥmad
dans les manuscrits du Proche-Orient musulman. al-Būnī’s Works ». Arabica 64, n° 3-4 (2017) : 405-441.
Paris : Presses de l’Université de Paris-Sorbonne, Gardiner, Noah. « Stars and Saints: The Esotericist
2003. Astrology of the Sufi Occultist Aḥmad al-Būnī ».
Caiozzo, Anna. Réminiscences de la royauté cosmique Magic, Ritual, and Witchcraft 12, n° 1 (2017) : 39-65.
dans les représentations de l’Orient médiéval. Le Gardiner, Noah. « The Occultist Encyclopedism of ʿAbd
Caire : Institut français d’archéologie orientale, 2011. al-Raḥmān al-Bisṭāmī ». Mamlūk Studies Review, no 21
Coulon, Jean-Charles. La Magie en terre d’islam au (2017) : 3-38.
Moyen Âge. Paris : CTHS, 2017. Gardiner, Noah. « Forbidden Knowledge? Notes on the
Production, Transmission, and Reception of the
Kalīla, Dimna et la géomancie 91
Major Works of Aḥmad al-Būnī ». Journal of Arabic Melvin-Koushki, Matthew. « Persianate Geomancy from
and Islamic Studies, n° 12 (2012) : 81-143. Ṭūsī to the Millennium: A Preliminary Survey ». In
Gruber, Christiane. « The ‘Restored’ Shīʿī muṣḥaf as The Occult Sciences in Pre-modern Islamic Cultures.
Divine Guide? The Practice of fāl-i Qurʾān in the Édité par Nader El-Bizri et Eva Orthmann, 151-200.
Ṣafavid Period ». Journal of Qurʾanic Studies 13, no 2 Beyrouth : Orient-Institut, 2018.
(2011) : 29-55. Regourd, Anne. « Au sujet des sources manuscrites de
Hartner, Willy. « al-Dj̲awzahar ». In Encyclopaedia of l’ouvrage imprimé au Caire sous le titre d’Al-faṣl fī
Islam, Second Edition. Édité par Peri Bearman et al. uṣūl ʿilm al-raml d’Al-Zanātī ». Annales islamolo-
Leiden : Brill, 1960-2007. giques 35 (2001) : 393-347.
Hauber, Anton. « Ṭomṭom (Ṭimṭim) = Δανδαμις = Regourd, Anne. « La géomancie comme voie d’accès
Dindymus ? ». Zeitschriften der Deutschen Morgen- à un savoir : un cas isolé au nord du Yémen ? ». In
ländischen Gesellschaft 63 (1909) : 457-472. Scienza e islam : atti della giornata di studio, Venezia,
Hébert, Jean-Claude. « Analyse structurale des géoman- 30 gennaio 1999. Édité par Giovanni Canova, 5-16.
cies comoriennes malgaches et africaines ». Journal Rome : Herder, 1999.
des Africanistes 31, no 2 (1961) : 115-208. Regourd, Anne. « Pratiques de géomancie au Yémen ».
Jaulin, Robert. Géomancie et Islam. Paris : Christian In Religion et pratiques de puissance. Édité par Alain
Bourgois, 1991. de Surgy, 105-127. Paris : L’Harmattan, 1997.
Jaulin, Robert. La Géomancie : analyse formelle. Paris : Rehder, Robert M. « New Material for the Text of Ḥāfiẓ ».
Mouton, 1966. Iran 3 (1965) : 109-19.
Kazimirski Biberstein, Albert de. Dictionnaire arabe- Saif, Liana. « From Ġāyat al-ḥakīm to Šams al-maʿārif:
français. 2 vol. Paris : Maisonneuve, 1860. Ways of Knowing and Paths of Power in Medieval
Kerviler, René. Le Chancelier Pierre Séguier, second pro- Islam ». Arabica 64, n° 3-4 (2017) : 297-345.
tecteur de l’Académie française : études sur sa vie pri- Savage-Smith, Emilie, et Marion B. Smith. « Islamic
vée, politique et littéraire, et sur le groupe académique Geomancy and a Thirteenth Century Divinatory:
de ses familiers et commensaux. Paris : Didier, 1875. Another Look ». In Magic and Divination in Early
Klein-Franke, Felix. « The Geomancy of Ahmad B. ʿAli Islam. Édité par Emilie Savage-Smith, 211-276.
Zunbul a Study of the Arabic Corpus Hermeticum ». Aldershot: Ashgate, 2004.
Ambix 20, no 1 (1973) : 26-35. Savage-Smith, Emilie, et Marion B. Smith. Islamic
Lippincott, Kristen. « More on Ibn Al-Hātim ». Journal Geomancy and a Thirteenth-Century Divinatory
of the Warburg and Courtauld Institutes 51 (1988) : Device. Malibu: Undena, 1980.
188-190. Scott Meisami, Julie. « Manuscripts of Hafez ». In
Lippincott, Kristen, et David Pingree. « Ibn al-Hātim on Encyclopaedia Iranica. Edité par Ehsan Yarshater.
the Talismans of the Lunar Mansions ». Journal of the Londres: Routledge, 1982-2020.
Warburg and Courtauld Institutes 50 (1987) : 57-81. Tokatlian, Armen. Falnamah : livre royal des sorts.
Melvin-Koushki, Matthew. « In Defense of Geomancy: Montreuil : Gourcuff Gradenigo, 2007.
Šaraf al-Dīn Yazdī Rebuts Ibn Ḫaldūn’s Critique Villuendas Sabaté, Blanca. La geomancia en los
of the Occult Sciences ». Arabica 64, no 3-4 (2017) : manuscritos judeo-árabes de la Gueniza de El Cairo.
346-403. Cordoue : CNERU-CSIC, 2015.
partie 2
Questions de filiation / Questions of Filiation
∵
4
Intertextual Animals: Illustrated Kalīla wa-Dimna Manuscripts
in Context
Anna Contadini
1 Ethical Wisdom and Textual History patronage and be propagated through the com-
missioning of further copies (see below).
Right at the beginning of his preface (muqad- Although many of the stories are concerned
dima), Ibn al-Muqaffaʿ explains what the Kalīla with how to employ force or trickery to gain the
wa-Dimna is about.1 The book, he says, is com- upper hand or exact vengeance, it is ultimately
posed of exempla and stories set down by learned an ethical wisdom that the reader should gain
men (ʿulamāʾ) from India in which they sought to through pondering their meaning, the core, to use
include the most eloquent statements they could Ibn al-Muqaffaʿ’s own metaphor, that is contained
find to communicate their wisdom. Learned men within the shell of the narrative. Indeed, the need
of every faith (min kull milla) and language strove for an ethical interpretation is stressed as the pref-
ceaselessly to impart their knowledge, and they ace proceeds, aptly enough, to resort to story as
desired to express it in such a way that the vari- the medium to illustrate the themes of a foolish
ous stratagems (ḥiyal) they constructed to convey choice that results in a painful lesson that never-
it, composed in the most eloquent and perfect theless leaves the sufferer wiser, or, more bluntly,
language, would be placed on the tongues of ani- deplorable behavior that leaves the perpetrator
mals and birds. He also adds that the book com- worse off. The latter will culminate in the death
bines wisdom and entertainment, and that the sentence carried out on one of the jackal protago-
wise (ḥukamāʾ) will choose the wisdom and fools nists in the later “Trial of Dimna” (al-faḥṣ ʿan amr
(sukhafāʾ) the entertainment.2 dimna) chapter, which, in Carl Brockelmann’s esti-
Such knowledge, therefore, needs to be trans- mation, “by punishing the traitor satisfies the feel-
mitted, and Ibn al-Muqaffaʿ also expresses a con- ing of justice outraged by the immoral teaching of
cern in the preface that his text should attract this book.”3 This may reflect a feeling that retribu-
tion is too long delayed, but surely tips the balance
too far: there is always a positive precept to be
1 Author’s note: I should like to thank Éloïse Brac de la
extracted from each story. Yet however important
Perrière and Annie Vernay-Nouri for inviting me to par- an element this may be, it is still misleading simply
ticipate in this interesting project; Aïda El Khiari for her to label the Kalīla wa-Dimna a moralizing text, as
help, especially with the images; and Crispin Branfoot and has often been done.
Farouk Yahya for references to the Pañcatantra. I should
Indeed, another prominent theme highlighted,
also like to thank the anonymous readers for their useful
comments. All readings and translations from the Arabic one complementary but different, is the acquisition
sources are mine unless otherwise stated.
2 Louis Cheikho, La version arabe de Kalīlah et Dimnah
d’après le plus ancien manuscrit arabe date (Arabic text. 3 The text of the “Trial of Dimna” is found in Cheikho,
Beirut: Imprimerie Catholique, 1905), 45. Ibn al-Muqaffaʿ La version arabe, 102–24. The quotation is from Carl
(ca. 103H/721-ca. 139H/757), of Persian origin, was an Brockelmann, “Kalīla Wa-Dimna,” in Encyclopaedia of
author and translator working in Basra, Iraq: see also foot- Islam, second edition, ed. P. Bearman et al. (Leiden: Brill,
notes 25 and 26. 1960–2007).
of knowledge as a means of attaining more pro- These characteristics inhere in the ultimately
found understanding. In accordance with the South Asian oral traditions of animal stories on
findings of Burzoy’s journey to India in search of which the Kalīla wa-Dimna draws.5 This pool of
the recipe for immortality, namely that it is knowl- stories – many of them with parallels in the Greek
edge, in Ibn al-Muqaffaʿ’s preface there is a strong Aesop’s fables6 – are recorded in texts such as the
emphasis on the importance of knowledge and, Pali Jātaka tales and the Sanskrit Pañcatantra. The
specifically, that gained through books, including, former contains stories of the previous lives of the
of course, this book. Such knowledge, though, lies Buddha, in which he was reborn in various forms,
beneath the mere surface of literal understanding including those of animals,7 while the Pañcatantra,
and needs to sink into the heart in order to bear which is the more direct ancestor of the Kalīla wa-
fruit (thamarāt), the enlightened wisdom that Dimna, is a collection of animal tales.8 The sto-
such knowledge enables – a metaphor that leads ries they contain are entertaining but also have
into the ethical and religious domains of virtue important spiritual and educational functions: the
and God’s gifts to man.4 Repeatedly described in Jātakas teach key Buddhist concepts and values,
the Qurʾan as God’s sustenance to man, thamarāt while the Pañcatantra imparts lessons on the art
are interpreted as the fruits of knowledge, to be of statecraft (indeed, one of its sources appears
understood ethically as virtue and good works, to have been an arthaśāstra, or technical treatise
and sometimes also, in Sufi tafsīr, as insight. on governing9). The concern with governance is
If this is a text of the “mirrors for princes” type, it echoed in another source of the Kalīla wa-Dimna,
is one that makes one reflect on human behaviors the twelfth book of the war epic the Mahābhārata,
and motivations more generally; and it is vividly which also includes a section on arthaśāstra.10
didactic. At the same time, by showing episodes of However, the Pañcatantra is more than a guide to
gullibility on the part of various persons and ani- politics and administration: it describes itself in its
mals, it is entertaining and, in some respects, even preamble as a nītiśāstra, a technical treatise on the
comic. This gives the text a universal appeal, as
it presents a spectrum of attitudes and emotions
that are part of the life and experience of both 5 For a discussion on the sources of the Kalīla wa-Dimna,
individuals and groups. It provides a mirror within see François de Blois, Burzōy’s Voyage to India and the
Origin of the Book of Kalīlah wa Dimnah (London: Royal
which attitudes and emotions can be seen and rec-
Asiatic Society, 1990), 12–17.
ognized as one’s own. And this is why, I suggest, 6 Aesop’s fables had been translated into Arabic via a
the text, especially when accompanied by paint- Syriac intermediary already by the seventh century. See
ings, has enjoyed such success and has reached Robert Irwin, “The Arabic Beast Fable,” Journal of the
out far beyond a narrow courtly audience: rather Warburg and Courtauld Institutes 55 (1992): 36–50; 38.
7 On the Jātakas, see Sarah Shaw, trans., The Jātakas:
than a mirror for princes, Kalīla wa-Dimna is a mir- Birth Stories of the Bodhisatta (New Delhi: Penguin
ror for humanity. Books, 2006). I thank Farouk Yahya for discussing this
with me.
8 There has been much literature on the Pañcatantra.
See, for instance, Theodor Benfey, Pantschatantra: fünf
4 Cheikho, La version arabe, 47. Qurʾan 2:22, 16:69, 47:15, Bücher indischer Fabeln, Märchen und Erzählungen
etc. For translations of the passages, see Muhammad (Leipzig: Brockhaus, 1859); Johannes Hertel, Das
A.S. Abdel Haleem, trans., The Qurʾan (Oxford: Oxford Pañcatantra, seine Geschichte und seine Verbreitung
University Press, 2004). For a discussion of this point in (Leipzig, Berlin, B.G. Teubner, 1914); Visnu Sarma, The
the context of other Arabic illustrated manuscripts, see Pancatantra (New Delhi: Penguin Books, 1993); Patrick
Anna Contadini, A World of Beasts: A Thirteenth-Century Olivelle, trans., The Pañcatantra: The Book of India’s
Illustrated Arabic Book on Animals (the Kitāb Naʿt Folk Wisdom (Oxford: Oxford University Press, 1997).
al-Ḥayawān) in the Ibn Bakhtīshūʿ Tradition (Leiden: Brill, 9 The Arthaśāstra of Kauṭilya (fl. 300 BCE).
2012), 73–74. 10 De Blois, Burzōy’s Voyage, 16.
Intertextual Animals 97
concept of nīti – “living wisely and well in the tru- also, and in particular, in the same color; this prac-
est sense of these two terms.”11 tice provides a visual counterpart to the textual
But who was the Pañcatantra aimed at? Accord- characterization as well as an aide mémoire. In the
ing to the introduction, it was devised as a book ca. 1220 manuscript in the Bibliothèque nationale
of instruction for three ignorant princes, but de France (BnF) in Paris, Arabe 3465, for example,
as many of the rulers are portrayed as weak and Kalīla’s fur is always blue and Dimna’s always
dimwitted, in contrast to their wise ministers, it red; in the ca. 1300–10 copy in the Bayerische
is possible that it was at the latter group that the Staatsbibliothek (BSB) in Munich, Cod. arab. 616,
text was targeted.12 More generally, it has been Kalīla is always light brown while Dimna is mauve
argued that the teachings contained in the Pañ- (see Fig. 4.1).17 The unvarying characteristics of the
catantra apply to everyone, and that it “holds a
mirror up to society,”13 just as its successor, Kalīla 17 On the first manuscript, see Ernst J. Grube, ed., A
wa-Dimna, does. Mirror for Princes from India: Illustrated Versions of the
Although a direct line to Kalīla wa-Dimna from Kalilah wa Dimnah, Anvar-i Suhayli, Iyar-i Danish, and
the Pañcatantra, the original of which is in any Humayun Nameh (Mumbai: Marg, 1991), 43 and figs. 10,
15, 18, 24, 35, 61; Julian Raby, “The Earliest Illustrations
case no longer extant, cannot be established, the to Kalilah wa Dimnah,” in Grube, A Mirror for Princes,
texts share a structural division into five main sec- 18–22; 26–28 and figs. 10, 15, 18, 24; Wafaa Rizkallah, “The
tions (Pañcatantra meaning “five topics” or “five Earliest Arab Illustrated Kalīla wa Dimna Manuscript
books”14), each with a main story which generates (BN Arabe 3465): A Study of Its Miniatures” (MA
a series of distinct but thematically connected Thesis, American University in Cairo, 1991); Lucy-Anne
Hunt, “The Commissioning of a Late Twelfth Century
stories.15 At first, this nesting structure may seem Gospel Book: The Frontispieces of MS Paris, Bibl. Nat.
complicated and difficult to follow. But the reader Copte 13,” in Byzantium, Eastern Christendom and Islam:
is aided by the fact that each animal has its own Art at the Crossroads of the Medieval Mediterranean,
easily recognizable characteristics. For example, ed. Lucy-Anne Hunt (London: Pindar Press, 1998),
vol. 1, 115–57: 131–32, 149; Marie-Geneviève Guesdon
the two jackals are astute and vengeful; the tortoise and Annie Vernay-Nouri, eds., L’Art du livre arabe (Paris:
is weak and easily influenced; the camel is gullible; Bibliothèque nationale de France, 2001), 113 with two
and the lion is physically strong (and a bully), but color illustrations and no. 95, 132 (entry by Annie
often lacks judgment. These characteristics are Vernay-Nouri); Bernard O’Kane, Early Persian Painting:
Kalila and Dimna Manuscripts of the Late Fourteenth
retained by the particular animal across various
Century (London: I.B. Tauris, 2003), 218, Appendix 1;
stories, and similar characteristics are found in the Anna Contadini, “Ayyubid Illustrated Manuscripts and
Pañcatantra.16 It is interesting to note that pictori- their North Jaziran and ʿAbbasid Neighbours,” in
ally, too, an animal is usually represented not just Ayyubid Jerusalem: The Holy City in Context 1187–1250,
through a particular set of visual conventions but ed. Robert Hillenbrand and Sylvia Auld (London:
Altajir Trust, 2009), 179–194, and col. ills. 17–37: 180,
184–185, pl. 9.5 and col. pls. XXV–XXVII; on the second,
11 On nīti, see Sarma, Pañcatantra, xv, xxvii–xxviii (quota- see Hans-Caspar Graf von Bothmer, Kalila und Dimna:
tion from xxviii). Ibn al-Muqaffaʿ’s Fabelbuch in einer mittelalterlichen
12 Olivelle, Pañcatantra, xxviii–xxix. Bilderhandschrift: Cod. arab. 616 der Bayerischen
13 In Sarma, Pañcatantra, xxvii–xlvii, 4–5 (quotation from Staatsbibliothek München (Wiesbaden: Reichert, 1981);
xxxv). O’Kane, Early Persian Painting, 221; Sabrina Roth,
14 De Blois, Burzōy’s Voyage, 16; Olivelle, Pañcatantra, xiv. “Kalila und Dimna in München: Ibn al-Muqaffaʿ’s Kitāb
15 François de Blois has noted that the framework story Kalīla wa-Dimna und die Handschrift Codex arabicus
of the Kalīla wa-Dimna is more akin to that of the 616 der Bayerischen Staatsbibliothek in München,” in
Mahābhārata (which is also one of its main sources), in Von listigen Schakalen und törichten Kamelen: Die Fabel
that it is a dialogue between a king and a philosopher; in Orient und Okzident, ed. Mamoun Fansa and Eckhard
see de Blois, Burzōy’s Voyage, 18. Grunewald (Wiesbaden: Reichert, 2008), 99–114; Karen
16 For the characteristics of the animals in the Pañcatan- Aydin and Birgit Middendorp, “Ibn al-Muqaffaʿ’s Kalīla
tra, see Olivelle, Pañcatantra, xxii–xxv. wa-Dimna und der Codex arabicus 616 der Bayerischen
98 Contadini
figure 4.1 The two jackals: Kalīla and Dimna. Ibn al-Muqaffaʿ, Kalīla wa-Dimna; a) Egypt or Syria, ca. 1220. BnF Arabe 3465,
fol. 48r; b) Egypt or Syria, ca. 1300–10. BSB Cod. Arab. 616, fol. 44v
animals reflect the presumed oral, popular ante- theatres, where one would expect to find a similar
cedents of the textual tradition – flat characters are emphasis on typological fixity of representation.19
typical of orally transmitted stories, their formu- If the specifics of the material derived from the
laic nature helping the memorization process.18 Pañcatantra remain obscure, the general route
One might also wonder whether stories from of transmission is tolerably clear: from Sanskrit
Kalīla wa-Dimna were acted out in shadow puppet it goes into Middle Persian, as the Kalīla wa-
Dimna itself confirms in an introductory chapter
included in some manuscripts. Entitled “Burzoy
the Physician” (Burzūyih al-mutaṭabbib), this is
Staatsbibliothek München,” in Tierisch moralisch: prefaced by a description of him as the one who
Die Welt der Fabel in Orient und Okzident, ed. Birgit
Middendorp (Oldenburg: Landesmuseum Natur und
Mensch, 2009), 111–115; Helga Rebhan, “Collecting 19 The extant texts written by Ibn Dāniyāl for the shadow
Islamic Manuscripts at the Munich Court Library in theatre indicate its popularity in thirteenth-century
the Nineteenth Century: An Acquisition History,” in Cairo. They do not, however, include animal characters.
Manuscripts, Politics and Oriental Studies: Life and For orality and connections with the shadow theatre,
Collections of Johann Gottfried Wetzstein (1815–1905) though not as related to the Kalīla wa-Dimna stories,
in Context, ed. Boris Liebrenz and Christoph Rauch see Alain George, “Orality, Writing and the Image in
(Leiden: Brill, 2019), 197–227. the Maqāmāt: Arabic Illustrated Books in Context,” Art
18 See also discussion in Irwin, “Arabic Beast Fable,” 47. History 35, no. 1 (2012): 10–37.
Intertextual Animals 99
figure 4.2 Burzoy and physicians setting off on their journey. Ibn al-Muqaffaʿ, Kalīla wa-Dimna; a) Egypt or Syria,
ca. 1300–10. BSB, Cod. Arab. 616, fol. 21r; b) Egypt or Syria, dated 755H/1354. Bodleian Pococke 400, fol. 12v
“undertook the copying and translating of this The production of the Middle Persian version
book” (tawallā istinsākh hādhā al-kitāb wa tarja- is generally attributed to the era of the Sasanian
matah), and consists of an extensive narrative giv- Emperor Khusraw I Anushirwan (r. 531–79).
ing an account of his expedition in search of the His reign was marked by significant cultural
text.20 In some manuscripts, this is accompanied exchange with India, resulting, according to the
by an illustration showing him and his fellow phy- tenth-century historian al-Masʿūdī (280H/893–
sicians setting off on their journey, as, for example, 345H/956),22 in the introduction of both the Kalīla
in BSB Cod. arab. 616 (ca. 1300–10) (fol. 21r) and in wa-Dimna and chess to Persian cultural and social
the manuscript in the Bodleian Library in Oxford, life. The theme is developed further in the eleventh
Pococke 400, of 755H/1354 (fol. 12v) (see Fig. 4.2).21 century by the great Persian poet Firdawsī, who
gives, in the Shahnāma, a story about the introduc- as the basis for a Syriac translation made as early
tion of chess, as well as one about a search for the as the sixth century, and for the eighth-century
elixir of life that yielded the Kalīla wa-Dimna. Arabic version,25 the work of a Persian scholar liv-
The former story tells us that an Indian rajah sent ing and working in Basra, ʿAbdullāh b. al-Muqaffaʿ
an envoy to Khusraw Anushirwan with a chess set (d. c. 139H/756–7 or 142H/759–60).26 In the Fihrist
and a message written on silk to say that unless the of Ibn al-Nadīm (d. 385H/995 or 388H/998), the
Persians could name the pieces and work out their tenth-century Baghdadi bibliophile and writer,
moves, he would cease to pay tribute. Buzurgmihr, Ibn al-Muqaffaʿ is praised as a kātib (a member of
the Persian counsellor, aided somewhat by the the secretarial class) who was a poet and eloquent
indiscretions of the envoy, solved the problem writer, and as a translator who was a master of
within a week. In the latter story, a physician of both Persian and Arabic; indeed, he is said to have
repute, Burzoy, having heard of a plant in a certain translated many other Persian books in addition to
mountain in India that restores the dead to life, the Kalīla wa-Dimna. Ibn al-Nadīm also mentions
asked Anushirwan for permission, and the means, disagreements about its origins, which were vari-
to go and look for it. He set off with a company ously regarded as Indian, Ashkānian (Parthian), or
of physicians but searched to no avail until he was Persian, with contributions by Buzurgmihr. In his
taken to a wise man, who revealed that “the herb own classification, it is placed among the books of
is the scientist; science is the mountain, everlast- the Indians, and he says that it is in 17 or 18 sec-
ingly out of reach of the multitude. The corpse is tions; that it was translated by Ibn al-Muqaffaʿ; but
the man without knowledge, for the uninstructed that there are also other translations (unspecified)
man is everywhere lifeless. Through knowledge and, in addition, versified Arabic versions (again,
man becomes revivified.”23 He further informed unspecified).27
Burzoy of the existence in the king’s treasury of
a book called Kalīla, which was a guide to such
25 De Blois, Burzōy’s Voyage, 1–3; also see Dagmar Riedel,
knowledge. Burzoy eventually obtained the book “Kalila Wa Demna. i. Redactions and Circulation”,
and, once back in Persia, asked Anushirwan to Encylopaedia Iranica XV/4, 386–395, available online at
have it translated by Buzurgmihr, with the request http://www.iranicaonline.org/articles/kalila-demna-i
that in the first chapter mention should be made (last accessed November 5, 2018).
26 On Ibn al-Muqaffaʿ, see Francesco Gabrieli, “Ibn
of his own name, his travels, and the difficulties he
al-Muḳaffaʿ,” in Bearman, Encyclopaedia of Islam;
experienced in finding it, so that he would not be J. Derek Latham, “Ebn Al-Moqaffaʿ, Abū Moḥammad
forgotten.24 The two stories are thus linked in the ʿAbd-Allāh Rōzbeh,” Encyclopaedia Iranica VIII/1, 39–43,
figure of Buzurgmihr, the solver of the chess puz- available online at http://www.iranicaonline.org/arti-
zle who is also regarded as the potential translator cles/ebn-al-moqaffa (last accessed August 25, 2020);
Michael Cooperson, “Ibn al-Muqaffaʿ,” in Medieval
of Kalīla wa-Dimna into Middle Persian. Islamic Civilization: An Encyclopedia, ed. Josef W. Meri
The story of Burzoy’s quest encapsulates a par- (New York; Abingdon: Routledge, 2006), vol. 1, 346–47;
ticular link in the complex transmission history and Michael Cooperson, “Ibn al-Muqaffaʿ,” in The
of the text. The resulting Middle Persian version, Biographical Encyclopedia of Islamic Philosophy, ed.
Oliver Leaman (London: Thoemmes Continuum,
itself also no longer extant, was to serve in turn
2006), vol. 1, 280–86.
27 Muḥammad ibn Isḥāq Ibn al-Nadīm, Fihrist (Beirut:
23 Firdawsī, The Epic of the Kings. Shah-Nama, the Dār al-maʿrifa li-l-ṭibāʿa wa-’l-nashr, 1978), 172 (on Ibn
National Epic of Persia by Ferdowsi, trans. Reuben Levy, al-Muqaffaʿ); 423 (classification); and 424 (on origins
rev. Amin Banani (London: Arkana, 1990), 332. and translations). See also the edition by Ayman Fuʿād
24 In the chapter of “Buzurgmihr the Wise Vizir” there is Sayyid, The Fihrist of al-Nadīm (London: Al-Furqān
a sub-section on “How Burzoy Brought the Kalīla wa Islamic Heritage Foundation, 2009). For the corre-
Dimna from Hindustan”. See, among others, Firdawsī, sponding passages in a translation of the Fihrist, see
The Epic of the Kings, 330–34. Bayard Dodge, ed. and trans., The Fihrist of al-Nadīm:
Intertextual Animals 101
Even though they are no longer extant, these of these that is related the unexpected repre-
further translations and adaptations attest to the sentation of a Kalīla wa-Dimna story in the late
popularity of the Kalīla wa-Dimna stories, while fifteenth-century program of sculptural reliefs on
the predominance of the Ibn al-Muqaffaʿ version the staircase of Palazzo Gondi in Florence, placed
can be attributed to its enduring reputation as an among reliefs of Aesop’s fables (see Fig. 4.3).30
outstanding example of Classical Arabic prose The majority of translations into modern
style. Ibn al-Nadīm refers to Kalīla wa-Dimna European languages come, however, much later,
within a section on “evening conversation and being made during the nineteenth and twenti-
story-telling” (al-asmār wa-’l-khurāfāt), while Ibn eth centuries directly from Ibn al-Muqaffaʿ. His
al-Muqaffaʿ himself calls the stories amthāl (exem- Kalīla wa-Dimna also generated various versions
pla, similitudes), but these categorizations fail in Persian, the most important of which are
to do justice to the depth and complexity of the al-Bukhārī’s Dāstānhā-yi Bīdpāy and Naṣrullāh’s
more extensive sections of the work. In terms of Kalīla wa-Dimna (both twelfth century),
language, there is a degree of contrast between the al-Kāshifī’s Anvār-i Suhaylī (fifteenth century), and
relatively straightforward and quick-moving nar- Abu-’l-Fażl’s ʿIyār-i Dānesh (sixteenth century).31
rative parts and those of a more argumentative or
discursive nature, which, although not as linguisti- ed. N.M. Penzer, vol. 5 (London: Chas J. Sawyer, 1926),
cally difficult as al-Ḥarīrī’s Maqāmāt, still exhibit 232–42, reproduced in Grube, A Mirror for Princes, end-
considerable complexity and ornateness of style, papers. See also Éloïse Brac de la Perrière and Anne
Vernay-Nouri, “Le Fabuleux Voyage de Kalīla et Dimna,”
reflecting the rhetorical and logical adroitness in Intinérance de la fable. Transmissions, transferts et
expected of a kātib versed in the arts of persuasion. transactions, special issue, Le Fablier 26 (2015): 173–83.
The success of Ibn al-Muqaffaʿ’s version is 30 Linda Pellecchia, “From Aesop’s Fables to the Kalila
shown by the number of manuscript copies that wa-Dimna: Giuliano da Sangallo’s Staircase in the
Gondi Palace in Florence,” I Tatti Studies, Essays in the
survive, several of which are illustrated. It was Renaissance 14–15 (2011–12): 137–207. There are four-
also to spawn, directly or indirectly, a vast num- teen sculptural reliefs at the step end of the staircase:
ber of translations. In the medieval West, where five illustrate Aesop’s fables and one is from the Kalīla
the Kalīla wa-Dimna stories became known as wa-Dimna, specifically “The Heron and the Crab”. The
original reliefs are now conserved in the Victoria and
the Bidpai fables – presumably from the Sanskrit
Albert Museum (inv. nos. 26–1891 to 39–1891), London,
vidyāpati (chief scholar)28 – there is a small group while those in the Palazzo Gondi are replicas.
of early translations based upon Old Spanish and 31 For al-Bukhārī’s Dāstānhā-yi Bīdpāy, see Muḥammad
Hebrew intermediaries.29 It is presumably to one ibn ʿAbd Allāh al-Bukhārī, Dāstānhā-yi Bīdpāy /
tarjumah-i. bih taṣḥịh,̣ ed. Parvīz Nātil Khānlarī and
Muḥammad Rawshan (Tehran: Khvārazmī, 1982). For
A Tenth-Century Survey of Muslim Culture, Part 1 (New Abu-‘l-Maʿālī Naṣrullāh Munshī’s Kalīla wa-Dimna, see
York; London: Columbia University Press, 1970), vol. 1, Tarjumah-i Kalīlah va Dimnah, ed. Mujtabā Minūvī
259–60; and vol. 2, 712 and 715–16. (Tehran: Dānishgāh-i Tihrān, 1964); also English
28 For an account of the Arabic animal fables and their translation by Joseph Harris, trans., The Instructive
sources, both East and West, see Irwin, “Arabic Beast and Entertaining Fables of Pilpay: An Ancient Indian
Fable.” The first translation of the Kalīla wa-Dimna in Philosopher (London: S. Birt and D. Browne, 1747); and
English is the one by Thomas North dated 1570, from an Abu-’l-Maʿālī Naṣrullāh Munshī, Kalila and Dimna,
Italian version by Anton Francesco Doni (published in trans. Wheeler Thackston (Indianapolis, IN: Hackett
1552); see Thomas North, The Moral Philosophy of Doni. Publishing, Inc., 2019). For ʿAlī b. Ḥusayn al-Wāʿiẓ
Popularly Known as The Fables of Bidpai, ed. Donald al-Kāshifī’s Anvār-i Suhaylī, see Anvār-i Suhelī or Lights
Beecher, John Butler, and Carmine di Biase (Ottawa: of Canopus being the Persian version of the Fables of
Dovehouse Editions, 2003). Bīdpaī by Husain Vāiz̤ Kāshifī, ed. Jasper Walker Joseph
29 See the family tree by Franklin Edgerton in “Genealogi- Ouseley (Hertford: Stephen Austin, 1851); also English
cal Table of the Panchatantra,” in The Ocean of Story, translation by Edward B. Eastwick, trans., The Anvár-i
102 Contadini
figure 4.3
“The Heron and the Crab”; a)
Giuliano Da Sangallo, step-end
from the Gondi Palace in Florence,
ca. 1490–1501. London, Victoria &
Albert Museum, inv. no. 35–1891;
b) Ibn al-Muqaffaʿ, Kalīla wa-Dimna,
Egypt or Syria, ca. 1300–10. BSB Cod.
Arab. 616, fol. 52r (detail)
2 Visual Intertextuality
Suhailí: Or, the Lights of Canopus, Being the Persian
Version of the Fables of Pilpay, or the Book of “Kalílah
and Damnah” (Hertford: Stephen Austin, 1854); for A yet further form of intertextuality arises from
an analysis of the text, see Christine van Ruymbeke, the tradition of illustrations, which includes not
Kāshefi’s Anvar-e Sohayli: Rewriting Kalila wa-Dimna only those in manuscripts but also wall paint-
in Timurid Herat (Leiden: Brill, 2016). For Abu-’l-Fażl
b. Mubārak’s ʿIyār-i Dānesh, see Antoine Isaac Silvestre
ings and reliefs. A striking, if completely different,
de Sacy, “Eyari danisch ou le paragon de la science. parallel to the Palazzo Gondi reliefs, for example,
Traduction persane du Livre de Calila, faite par le occurs in similar juxtapositions of scenes from
vazir Abou’lfazl,” Notices et Extraits des manuscrits Aesop and the Pañcatantra in the seventh- and
de la Bibliothèque du roi 10, no. 1 (1881): 197–225; Abu-
eighth-century wall paintings of Panjikent, an
’l-Fażl b. Mubārak, ʿIyār-i dānish (Dushanbe: Dānish,
1988); also English translation in Pilpay, The Fables of important Sogdian center along the Silk Road (see
Pilpay (London: Baldwin, Cradock, and Joy et al., 1818); Fig. 4.4a).32 These include both individual scenes
and Christine van Ruymbeke, “Authorship, Ownership
and Rewriting: Vāʿiz Kāshifī and Abū’l-Fazl b. Mubārak
within the Hereditary Line of Kalīla wa-Dimna 32 Raby, “The Earliest Illustrations,” 18–22; Boris Marshak,
Authors,” Jerusalem Studies in Arabic and Islam 45 Legends, Tales, and Fables in the Art of Sogdiana
(2018): 181–210. For an overview of the Persian versions, (New York: Bibliotheca Persica Press, 2002); Matteo
see De Blois, Burzōy’s Voyage, 5–6. Compareti, “Classical Elements in Sogdian Art:
Intertextual Animals 103
books to their libraries, and they also entered pub- 4 Patronage and Readership
lic libraries, such as those of madrasas. Interesting
in this context is the catalog of a thirteenth- In the case of three manuscripts – the ca. 1224–25
century library in Damascus. Studied in detail by Kitāb Naʿt al-Ḥayawān by Ibn Bakhīshūʿ (London,
Konrad Hirschler, this provides an inventory of the British Library, Or. 2784); the ca. 1224–25 Risālat
books in the Ashrafiyya Mausoleum, whose patron al-Ṣūfī fī al-kawākib (“Epistle of al-Sufi on the
was al-Malik al-Ashraf (d. 635H/1237). The librar- Stars”), probably by Ibn al-Ṣūfī (Tehran, Reza
ian was a certain Muḥammad b. ʿAbd al-Qādir Abbasi Museum, M. 570); and the 621H/1224 Kitāb
Muḥy al-Dīn al-Anṣārī (d. 683H/1284), and the al-ḥashāʾish fī hayūlā ʿilāj al-ṭibb (“Book of Simples,
inventory, for reasons internal to the document on the Substances Used for Medical Treatment”)
itself, can be dated to ca. 1250. This extraordinary by Dioscorides, copied by ʿAbdallāh b. al-Faḍl
text includes 2,100 titles of varying types: there (Istanbul, Süleymaniye Kütüphanesi, Aya Sofia
are works of poetry, adab literature, religious and 3703 and various detached leaves) – I have recently
scholarly treatises, and several illuminated (mu proposed that they all come from the same artis-
dhahhab) and illustrated (muṣawwar) manu- tic circle (if not from the same painter), and that
scripts, one an illustrated anthology entitled the first two are contemporary with the third, for
Emulation of Kalīla wa-Dimna (muḍāhāt kalīla wa- which the colophon provides the date.40 They
dimna muṣawwar).38 The Ashrafiyya inventory – were also most probably produced in or around
which must have been paralleled by inventories of Baghdad, during the cultural revival that took place
books available in other institutions39 – thus pro- under Caliph al-Nāṣir (575H/1180–622H/1225),
vides an insight into the eclectic variety of books himself a bibliophile.41 Although of different texts,
found in a library that was presumably open to the they can be looked at together not just because
general public; it suggests the possibility of a wide of their strong stylistic similarities, even down to
readership for the Kalīla wa-Dimna. tiny details, but also because of their implications
for the production of illustrated manuscripts for
an intellectual public with wide-ranging cultural
Islamic Art and Architecture, ed. Gülru Necipoğlu and interests embracing the wonders of the natural
Finbarr Barry Flood (Oxford: Wiley-Blackwell, 2017), world, astronomy, and zoology and botany (the
vol. 1, 431–52; 441–47.
last two also considered for their medical value).
38 Konrad Hirschler, The Written Word in the Medieval
Arabic Lands: A Social and Cultural History of Reading I have suggested that such manuscripts were
Practices (Edinburgh: Edinburgh University Press, probably made for scholars’ libraries.42 They stand
2012), 147; Konrad Hirschler, Medieval Damascus, as testimony to the lively intellectual life that was
Plurality and Diversity in an Arabic Library. The
Ashrafīya Library Catalogue (Edinburgh: Edinburgh
University Press, 2016), 286, no. 1021. The Emulation of 40 See Contadini, A World of Beasts, especially 130–31; and
Kalīla wa Dimna is by an author operating in Fatimid Anna Contadini, “A Question in Arab Painting: The
Egypt, Muḥammad b. Ḥusayn al-Yamanī (fl. fourth Ibn al-Ṣūfī Manuscript in Tehran and Its Art-Historical
H/tenth century, d. 400H/1009); see Carl Brockelmann, Connections,” Muqarnas 23 (2006): 47–84. For the
Geschichte der arabischen Litteratur, supplement 1 Dioscorides, see also Alain Touwaide, Farmacopea
(Leiden: Brill, 1937), 202, no. 3a. An edition of this man- Araba Medievale (Dioscorides De Materia Medica):
uscript is mentioned by Hirshler: Muḥammad Yūsuf Codice Ayasofia 3703 (Milan: Antea Edizioni, 1992–93).
Najm, Kitāb muḍāhāt kalīla wa dimna bimā ashbahahā 41 Angelika Hartmann, An-Nāṣir li-Dīn Allāh (1180–1225):
min ashʿār al-ʿarab (Beirut: Dār al-Thaqāfa, 1961). A Politik, Religion, Kultur in der späten ʿAbbasidenzeit,
non-illustrated copy of the manuscript is presently in Studien zur Sprache, Geschichte und Kultur des isla-
the Vatican Library, Vat. Ar. 1177; this is a compilation mischen Orients, 8. Berlin, New York: de Gruyter, 1975;
of texts, with the muḍāhāt kalīla wa dimna on fols. 25r- also Angelika Hartmann, “al-Nāṣir li-Dīn Allāh, Abū
52v, dated on fol. 52v as Rabī‘ al-Awwal 862H/1457. ’l-ʿAbbās Aḥmad,” in Bearman, Encyclopaedia of Islam.
39 Hirschler, Medieval Damascus, 1. 42 Contadini, A World of Beasts, 164.
106 Contadini
revived during the period in question, and their At the same time, this is not to argue for an
quality should not be read, without supporting absence of princely patronage. Ibn al-Muqaffaʿ is
evidence, as an argument for princely patron- clear that he expects rulers to be attracted by the
age. For example, the splendid 595H/1199 Kitāb illustrations, and confirmation of princely interest
al-Diryāq in the BnF (Arabe 2964) tells us, on fol. is provided by the production, probably between
38, that it was made for Abu-’l-Fatḥ Maḥmūd “the 1144 and 1152, of a versified Persian translation
knowledgeable imam, king of scholars” (malik for the Ghaznavid Bahrām Shāh,45 and of the
al-ʿulamāʾ).43 Later, after the Mongol invasion, the commissioning of Rūdakī by Naṣr II ibn Aḥmad
ʿAjāʾib al-makhlūqāt by Zakarīyā ibn Muḥammad (319H/913–331H/942), one of the Samanid rulers
Qazwīnī now in the BSB (Cod. arab. 464), which is of Bukhara and Samarkand, to compose a metrical
dated 678H/1280 and was therefore made during version, which was then illustrated.46
Qazwīnī’s lifetime, says that it was written by the Insights into the nature of patronage can often
doctor Muḥammad b. ʿAlī al-Dimashqī, who was be gained from frontispieces. In the case of Kalīla
living in Wasit at the time of the completion of wa-Dimna manuscripts, one example is that of
the manuscript.44 As Qazwīnī was a qāḍī at Wasit BnF Arabe 3465 (ca. 1220); here, the frontispiece is
at that time it is reasonable to assume that it was placed, at fol. 34r, between the statement of aims
made under his supervision and probably for a (fol. 33v) and the list of contents (fols. 34v and 35r)
scholar of his circle; certainly, the indication that (Fig. 4.6a). It portrays an unidentified, enthroned,
the scribe too belonged to the intellectual stratum princely figure, and as one of the stated aims is
of society is interesting in this context. It would to make the book attractive to princes it would
be reasonable to consider the Kalīla wa-Dimna seem to make sense for the manuscript to be
manuscripts against this background, for despite a commission by or for a ruler. Its format, of a
the work’s didactic elements and the entertain- seated central figure flanked by two attendants,
ment value of the stories it is by no means devoid is also encountered in one of the double fron-
of intellectual depth. Added to this, the associated tispieces of the ca. 1224–25 British Library Kitāb
representations of animals are analogous in their Naʿt al-Ḥayawān of Ibn Bakhtīshūʿ (Or. 2784,
narrative appeal to those in the books of animals fol. 3v, reconstructed 2v) (Fig. 4.6b), and in the
and ʿajāʾib literatures. frontispiece of the 642H/1245 Dioscorides in the
Biblioteca Universitaria in Bologna (Cod. Arab.
2954, fol. 141r) (Fig. 4.6c). These parallels raise
43 See Bishr Farès, Le livre de la thériaque (Cairo: doubts about an easy identification of the central
Imprimerie de l’Institut français d’archéologie orien- figure in the Kalīla wa-Dimna manuscript. That in
tale, 1953), 8–9 and pl. V. On this manuscript, see also the Dioscorides represents the author of the book,
Oya Pancaroğlu, “Socializing Medicine: Illustrations
of the Kitāb al-diryāq,” Muqarnas 18 (2001): 155–72;
with, flanking him, Aristotle and Luqmān, as iden-
Jaclynne Kerner, “Art in the Name of Science: The Kitāb tified by inscriptions at the top of the page.47 In
al-diryāq in Text and Image,” in Arab Painting: Text the Naʿt there are no comparable specific refer-
and Image in Illustrated Arabic Manuscripts, ed. Anna ences: one princely figure wears a crown and holds
Contadini (Leiden: Brill, 2007 [revised edition 2010]),
a regal spear, while on the opposite page the cen-
25–39; Kitāb al-Diryāq (Thériaque de Paris), facsimile
edition of ms. Arabe 2964 in the Bibliothèque natio- tral figure is, in contrast, a scholar-prince, a bowl
nale, Paris (Sansepolcro: Aboca Edizioni, 2010).
44 See Persis Berlekamp, Wonder, Image, & Cosmos in 45 By Abu-’l-Maʿālī Naṣrullāh ibn Muḥammad ibn ʿAbd
Medieval Islam (New Haven: Yale University Press, al-Ḥamīd, who also states in his introduction that
2011), 14–15; Contadini, A World of Beasts, 160; Stefano he would include pictures. See Raby, “The Earliest
Carboni, The Wonders of Creation and the Singularities Illustrations,” 18.
of Painting: A Study of the Ilkhanid London Qazvīnī 46 Ibid.
(Edinburgh: Edinburgh University Press, 2015), 128. 47 Contadini, A World of Beasts, plates 7, 8, 15.
Intertextual Animals 107
figure 4.6 a) Ruler-prince, enthroned with two attendants. Ibn al-Muqaffaʿ, Kalīla wa-Dimna,
Egypt or Syria, ca. 1220. BnF Arabe 3465, fol. 34r; b) Ruler-prince, enthroned with two
attendants. Ibn Bakhtīshūʿ, Kitāb Naʿt al- Ḥayawān (“Book on the Characteristics of
Animals”), probably Baghdad, datable to ca. 1224-25. London, British Library, Or. 2784,
fol. 2v (reconstructed fol. 3v); c) Dioscorides as scholar-prince (with Aristotle and
Luqmān), frontispiece. Dioscorides, Hayūlā ʿIlāj al-Ṭibb, Syria?, dated 642H / 1245.
Bologna, Biblioteca Universitaria, Cod. arab. 2954, fol. 141r; d) Joseph of Arimathea
claiming the body of Christ. Gospel, Damietta, Egypt, dated 1178–80. BnF Copte 13,
fol. 131r
108 Contadini
of fruit at his feet symbolizing knowledge and its Al-Ṭabarī (d. 310H/923), for example, refers to a
transmission.48 Various identities and relation- copy of Kalīla wa-Dimna owned by a judge,52 and
ships may be suggested by them. Apart from being in parallel to the possibility of others being cre-
inherently a figure of authority to whom homage ated at the behest of scholarly or at least educated
is due, the prince may be the patron or dedicatee urban patrons, we may note other commonalities
of the book, whereas the representation of knowl- that illustrated Kalīla wa-Dimna manuscripts share
edge in the other illustration may embrace both with those with more broadly scientific contents,
the author and the ancient sage who is his source, namely the persistence of certain conventions of
investing them with a competing form of author- representation across different periods and places.
ity, this time intellectual.49 All this suggests a widespread interest among
Further connections are suggested by the pos- the educated classes in illustrated manuscripts of
ture of the prince in the Kalīla wa-Dimna manu- various types and the existence of a lively indus-
script. He is seated cross-legged on a low throne, try to service it, with connections between ate-
and his hand, emerging from the upper aperture liers, scribes, painters, and bookbinders moving
of his mantle, makes a gesture similar to the between different production centers and working
benedictory one made by Christ or a patriarch in on diverse projects for different patrons.53
Byzantine manuscripts.50 Further, his crown and
facial features bear a striking resemblance to those
of Pilate in the Damietta Gospel of 1178–80 in the 5 Text and Image
BnF (Copte 13, fol. 131r) (Fig. 4.6d).51 These various
connections to different but interconnected tradi- Among the various gruesome stories in Kalīla
tions of representation widen the interpretative wa-Dimna (certainly not to be read to children at
possibilities: the figure in the frontispiece of BnF bedtime!), there is one – my favorite – that por-
Arabe 3465 may have been seen as a patron, as a trays freedom attained through the virtues of wise
princely figure in a scholarly role, or as an ancient leadership, friendship, and generosity. It forms
king. It has been suggested, indeed, that it rep- the first part of the complex narrative of what
resents Dabshalīm, the king of the Pañcatantra the index calls the “Ring-dove Chapter,” but in the
(himself a scholar interested in astronomy), and body of the text is headed in more detail as “The
therefore that it is an homage to an ancient source Crow, the Ring-dove, the Rat, the Tortoise and the
related to the origin of the Kalīla wa-Dimna. The Gazelle” (al-ghurāb wa-’l-muṭawwaqa wa-’l-juradh
text provides no clue, but it is at least clear that wa-’l-sulaḥfāh wa-’l-ẓaby).54 Its springboard is a
the presence of this frontispiece does not neces- request made by the king, who has been told how
sarily indicate that the manuscript was a royal the perfidious can undermine friendship, for a tale
commission, and there can be no doubt as to that shows how sincere brethren (ikhwān al-ṣafāʾ)
the circulation of copies beyond court circles. first form a bond.55
48 Contadini, A World of Beasts, 58. 52 Abū Jaʿfar Muḥammad ibn Jarīr al-Ṭabarī, Annales quos
49 Contadini, A World of Beasts, 58, also 72–75. scripsit Abu Djafar Mohammed Ibn Djarir at-Tabari, ed.
50 Contadini, A World of Beasts, 58. Michael Jan de Goeje (Leiden: Brill, 1881), vol. 2, 1308–9,
51 On this manuscript, see Jules Leroy, Les manuscrits 1318; see also Raby, “The Earliest Illustrations,” 18.
syriaque à peintures, conservés dans les Bibliothèques 53 See Contadini, A World of Beasts, esp. 164.
d’Europe et d’Orient (Paris: Geuthner, 1964), vol. 1, 99; 54 Cheikho, La version arabe, 125.
Jules Leroy, Les manuscrits coptes et coptes-arabes illus- 55 It has been argued that it is from this passage that the
trés (Paris: Geuthner, 1974), nos. XVI and XVII, 113–48, collective name Ikhwān al-Ṣafāʾ is derived; see Irwin,
col. pls. C and D, figs. 41–73; Hunt, “The Commissioning”; “Arabic Beast Fable,” 50. See also Ikhwān al-Ṣafāʾ, The
Contadini, “Ayyubid Illustrated Manuscripts,” 180, and Case of the Animals versus Man Before the King of the
col. pls. XXIX and XXX. Jinn: A Translation from the Epistles of the Brethren of
Intertextual Animals 109
figure 4.7 Pigeons trapped in the net by the fowler, and rat gnawing the net, while the crow looks down from the tree.
Ibn al-Muqaffaʿ, Kalīla wa-Dimna; a) Egypt or Syria, ca. 1220. BnF Arabe 3465, fol. 84v; b) Egypt or Syria,
mid-fourteenth century. BnF Arabe 3467, fol. 63r; c) Egypt or Syria, dated 755H / 1354. Bodleian Pococke 400,
fol. 82v
It begins with a crow, seated in its tree, watching a birds are trapped in the fowler’s net (Fig. 4.7a). In
fowler set a trap with a net and grain. After a while, BSB Cod. arab 616 (ca. 1300–10), on the other hand,
a flock of hungry pigeons arrive, and despite the the story is illustrated by three paintings, on fols.
wise advice of the ring-dove, their chief, they are 76r, 76v, and 77v, illustrating different moments
unable to resist the lure of the grain. Caught in the in the story: the birds trapped in the net; the birds
fowler’s net, they flap their wings in vain until the managing to fly off; and the rat gnawing through
ring-dove persuades them to unite their efforts, the net (Fig. 4.8). In BnF Arabe 3467 (ca. mid-
thus allowing them to fly off, still in the net. The fourteenth century) on fol. 63r (Fig. 4.7b), and in
fowler follows, expecting them to tire and fall. To Bodleian Pococke 400 (755H/1354), on fol. 82v, we
shake him off, the pigeons fly among buildings, have one painting showing the moment in which
coming at last to the hole of a rat, a loyal friend of the rat gnaws through the net, while the crow looks
the ring-dove, who gnaws through the net and sets down from the tree – the moment, just prior to the
them free. Much impressed by the actions of the freeing of the pigeons, which underlines the gen-
rat, the crow then sets about the laborious task of erosity and friendship shown by the rat (Fig. 4.7c).
gaining his friendship.56 It is interesting to note the differences between
This story is variously illustrated in different the choices made by the painters. Often, in Arab
manuscripts. In BnF Arabe 3465 (ca. 1220) on painting, the moment chosen for illustration is
fol. 84v the moment chosen is that in which the the one preceding the resolution of the narra-
tive. Dramatically, it is a moment of tension, of
anticipation. In this story, however, three crucial
and dramatically charged moments may be distin-
guished, the initial predicament (the moment of
Purity, trans. Richard J.A. McGregor and Lenn Evan
Goodman (Oxford: Oxford University Press with The
capture by the fowler), followed by two of escape,
Institute of Ismaili Studies, 2009), 156–58. first partial (flight inside the net) and then com-
56 Cheikho, La version arabe, 128–30. plete (release from the net), and various selections
110 Contadini
figure 4.8 a) Pigeons trapped in the net, b) they fly off while still in the net, c) and rat gnawing through the net. Ibn
al-Muqaffaʿ, Kalīla wa-Dimna; Egypt or Syria, ca. 1300–10. BSB Cod. arab. 616, fols. 76r, 76v and 77v respectively
from these are made in the different manuscripts, 978H/23 September 1570, it represents one of
with only BSB Cod. arab 616 including all three. the earliest examples of Mughal painting during
Other manuscripts choose either the first or the Akbar’s reign. It has generally been assumed that
last. It is the second, the intermediate stage, that is
the one selected in the manuscript of Ḥusayn Vāʿiẓ 2007, https://www.soas.ac.uk/gallery/treasuresofsoas/.
Kāshifī’s Anvār-i Suhaylī (“Lights of Canopus”) My summary of the stories there are based on both
in the Special Collections of SOAS Library (Ms. Eastwick’s translation and my direct analysis of the
10102). It has a beautiful illustration of the pigeons SOAS text; Eastwick, The Anvár-i Suhailí. See also
Anna Contadini, “A Jewel of Mughal Painting: The
managing to fly off, still in the net, while the fowler
SOAS Anvār-i Suhaylī Manuscript,” Orientations 38,
runs after them and the crow observes the scene no. 8 (2007): 51–56; Contadini, “The SOAS Anvār-i
from the top of the tree (fol. 123r) (Figure 4.9).57 Suhaylī Manuscript”, 22–25; John Seyller, “The School
This manuscript, a particularly impressive copy, of Oriental and African Studies Anvār-i Suhaylī: The
is distinguished by the outstanding quality of Illustration of a De Luxe Mughal Manuscript,” Ars
Orientalis 16 (1986): 119–51; Mika Natif, “The SOAS
its 27 sumptuous paintings.58 Dated 22 Rabīʿ II Anvār-i Suhaylī: The Journey of a “Reincarnated”
Manuscript,” Muqarnas 25 (2008): 331–58. Two paint-
57 Anna Contadini, “The SOAS Anvār-i Suhaylī Manu ings are in a Persian, so-called Bukhara style, the first
script,” in Objects of Instruction: Treasures of the School of which has been attributed on stylistic grounds to
of Oriental and African Studies, ed. Anna Contadini Shahm, who signed paintings in a 975H/1567 manu-
(London: SOAS, 2007 [reprinted 2016]),” 25. script of Saʿdī’s Gulistān. The others are Mughal in style,
58 The manuscript has 349 folios, measuring 335 × and attributed to approximately eight painters, indi-
215 mm, with nineteen lines of text in a very good cating the involvement of different artists working in
calligraphic nastaʿlīq hand, in black ink with rubrics distinct styles, the more recent developing earlier con-
in red, green, and gold. For the complete digitized ceptual ideas of space to arrive at a more forceful and
manuscript with an introduction, bibliography, and distinctively Mughal mode of expression. The device of
summaries of the stories related to the paintings, see “continuous narrative,” where more than one moment
Anna Contadini, Anvār-i Suhaylī (silverlight version, in the story is represented in the same painting, is also
cross platform) and Anvār-i Suhaylī (xbap version), effectively used in some paintings.
Intertextual Animals 111
figure 4.10 Geese carrying the tortoise in flight. a) Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca. 1220. BnF Arabe
3465, fol. 67r; b) Egypt or Syria, ca. 1300–10. BSB Cod. Arab. 616, fol. 60v; c) Egypt or Syria, dated 755H / 1354.
Bodleian Pococke 400, fol. 61v
(Fig. 4.11b).62 It seems that in the Buddhist context (and, according to the vertical comment at the
the didactic moment is the one that needed to be side of the BSB illustration, reproving him). One
highlighted, while the Arab painters preferred the might conjecture that the painters wished to
wondrous and somewhat comical scene of a flying emphasize the final moments in which Kalīla and
tortoise hanging from a stick. Dimna were together, and avoid representing their
No such lighter moments alleviate the chapter miserable, separate deaths.
in which Dimna, thrown into prison, is visited by The BnF Arabe 3465 ca. 1220 manuscript, how-
Kalīla, after which Kalīla dies of sorrow and Dimna ever, veers off to follow a variant of the text and
is finally condemned to death by starvation. It is has a very interesting illustration on fol. 80v that
illustrated, among others, in the BSB Cod. arab shows, opposite Dimna, not Kalīla but another
616 manuscript (ca. 1300–10, fol. 70v), and in the jackal, shaʿhar, identified as such in the picture
Bodleian Pococke 400 manuscript of 755H/1354 and in the text. In other manuscripts, this jackal
(fol. 73v) (Figs. 4.12a and 4.12b).63 They choose is called Rūzabī. It is Rūzabī who had reported
the moment of the prison visit, and have two the death of Kalīla to Dimna, who then asked
jackals confronting each other, but the emphases Rūzabī to go and fetch their belongings. Dimna
are quite different. Dimna, on the left, within the gives Rūzabī half in return for its help in support-
prison frame, is lightly chained in one (Bodleian), ing a plea of innocence, vainly hoping to influence
and confined in a stoutly barred cage in another the outcome of the trial. The painting depicts the
(BSB). Kalīla, on the right, is simply facing Dimna shaʿhar/Rūzabī in a strenuous standing posture on
hind legs, carrying a heavy bag (Fig. 4.12c).64 The
figure 4.11
a) The tortoise falls to the ground.
Naṣrullāh, Kalīla wa-Dimna, Herat,
ca. 1510. Rampur, Raza Library, MS
2982; b) The tortoise, having fallen
to the ground, is beaten to death.
Stone relief, Mathura, third century
ce. Mathura Museum, no. J 36 (After
Grube, A Mirror for Princes, figs. 27 and
23 respectively)
figure 4.12 Dimna in prison. Ibn al-Muqaffaʿ, Kalīla wa-Dimna; a) Dimna visited by Kalīla. Egypt or Syria, ca. 1300–10.
BSB Cod. Arab. 616, fol. 70v; b) Dimna visited by Kalīla. Egypt or Syria, dated 755H/1354. Bodleian Pococke 400,
fol. 73v; c) shaʿhar/Rūzabī carrying off a heavy bag given to it by Dimna. Ibn al-Muqaffaʿ, Kalīla wa-Dimna,
Egypt or Syria, ca. 1220. BnF Arabe 3465, fol. 80v
114 Contadini
figure 4.14 a) Two goats. Ibn Bakhtīshūʿ, Kitāb Naʿt al-Ḥayawān (“Book on the Characteristics of Animals”), probably
Baghdad, ca. 1224-25. London, British Library, Or. 2784, fol. 113r (reconstructed fol. 22r); b) Two goats. Ibn
al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca. 1220. BnF Arabe 3465, fol. 110r
although no firm conclusions have been reached, The likelihood of an early date for BnF Arabe
it is generally accepted that the manuscript is dat- 3465 is further reinforced by the connections that
able to the early thirteenth century. It is relevant can be made with the Damietta Gospel of 1178–80
here to note that it is the only known illustrated in the BnF (Copte 13).66 There are similarities in
Kalīla wa-Dimna manuscript that has all the the vegetal elements of the plants and flowers
names of the animals entered above or beside (Fig. 4.15), in the peculiar way in which the grass
their illustrations. In this it resembles thirteenth- is depicted, by a band of light green onto which
century Maqāmāt manuscripts, the implication darker blades of grass have been painted, in the
being therefore that it may be from the same domed arches (Fig. 4.16),67 and, as noted above,
period, maintaining an archaic feature no longer between the central figure in the frontispiece
found in later manuscripts.65
the fables were probably closer to their Indo-Persian
65 Irwin, “Arabic Beast Fable,” 46: “In fact there is very prototypes, but in the second version, written in Sicily,
little evidence of characterization in the fables. After most of the animals lost them.”
Kalīla wa-Dimna, it was rare to find animals even with 66 See footnote 50.
personal names; they tend to be identified only by their 67 For instance, see the arch with a ribbed cupola in
species name or species nickname. They had names in the painting of the birth of St. John the Baptist in the
the first version of Ibn Ẓafar’s Sulwān al-Muṭāʿ, in which Damietta Gospel (fol. 138v).
116 Contadini
figure 4.15 Similar-shaped plants in: a) Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca. 1220. BnF Arabe 3465, fol.
107v; b) Gospel, Damietta, Egypt, dated 1178–80. BnF Copte 13, fol. 167v
figure 4.16 Ribbed blue domes on arches in: a) Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca. 1220. BnF Arabe 3465,
fol. 30v; b) Gospel, Damietta, Egypt, dated 1178–80. BnF Copte 13, fol. 138v
Intertextual Animals 117
figure 4.17 Similar-shaped conceptual architecture in: a) Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, ca. 1220. BnF
Arabe 3465, fol. 131v; b) Al-Ḥarīrī, Maqāmāt (“Assemblies”), Syria(?), dated 619H/1222. BnF Arabe 6094, fol. 180r
(fol. 34r) and the figure of Pilate (fol. 131r) (Figs. 4.6a stylistically remote, and the connections made
and 4.6d). Further, strong stylistic and iconograph- with the two previously mentioned manuscripts
ical connections can be made with the 619H/1222 may be taken as more indicative. If anything, the
Maqāmāt (BnF Arabe 6094):68 there are parallels archaizing inclusion of the names of the animals,
in the vegetation again, in the conceptual archi- when considered together with the back-attack
tecture – elements of it being identical (Fig. 4.17) – of the lion, which reminds us of the old tradi-
in the stance of the human figures, and in aspects tion of this iconographical motif,70 reinforces the
of the clothing and furniture.
Some of these iconographical and design ele- of the Twelfth to Mid-Thirteenth Centuries: Sources
ments can be identified in later manuscripts, such of Wall-Painting at Deir es-Suriani and the Illustration
as the Gospel in the Institut Catholique in Paris of the New Testament MS Paris, Copte-Arabe 1/Cairo,
Bibl. 94,” in Byzantium, Eastern Christendom and Islam:
(Copte-Arabe 1),69 but this Gospel is otherwise
Art at the Crossroads of the Medieval Mediterranean,
ed. Lucy-Anne Hunt (London: Pindar Press, 1998,
68 Paris, BnF, Arabe 6094. For the date, see Contadini, 2000), vol. 1, 205–81; Contadini, “Ayyubid Illustrated
“Ayyubid Illustrated Manuscripts,” 83–84, the date Manuscripts,” 180 and col. pl. XXVIII.
being found in Kufic script on the boat (fol. 68r) and a 70 As seen, for example, in the Persepolis relief of the fifth
writing tablet (fol. 167r). century BCe: see Grube, A Mirror for Princes, 40, fig.
69 On this manuscript, see Leroy, Les manuscrits coptes, no. 30. The back-attack of the lion, however, is also found
XXI, 157–74, col. pls. E, F, G and figs. 75–91; Lucy-Anne in other manuscripts, as, for example, the Kalīla wa-
Hunt, “Christian-Muslim Relations in Painting in Egypt Dimna in Cambridge, Corpus Christi Library, MS 578,
118 Contadini
likelihood of an early date, and it would be rea- date. It needs further study (forthcoming by the
sonable to ascribe BnF Arabe 3465 to the early author), and will not be discussed here): BnF
thirteenth-century Ayyubid cultural area. This Arabe 3467 (ca. mid-fourteenth century); BSB Cod.
was a time of considerable political complexity, arab. 616 (ca. 1300–10); and Bodleian Pococke 400
with the Ayyubids of Damascus, the Zangids of (dated 755H/1354). They exhibit different stylis-
Mosul, the Artuqids of Diyarbekir and, of course, tic characteristics to BnF Arabe 3465 of ca. 1220,
the Abbasids of Baghdad, and it is consequently but are relatable in the same way to illustrations
no easy task to place a manuscript among these in other fourteenth-century Mamluk manuscripts
various potential sites of patronage.71 Indeed, of different texts. These include, among oth-
there are no manuscripts that can be securely ers, an al-Ḥarīrī Maqāmāt copied in Damascus
labeled Ayyubid, apart perhaps from one, the ca. 1300–10, in the British Library (Or. 9718);73 an
extraordinary al-Ṭarsūsī, Al-Tabṣira fī al-ḥurūb al-Ḥarīrī Maqāmāt, probably from Damascus,
(a manual on warfare), in the Bodleian Library dated 723H/1323, also in the British Library
(Huntington 264), made in Egypt or Syria for the (Add. 7293);74 an Ibn Buṭlān Risalāt Daʿwat
library of Saladin, as the text itself tells us, and al-Aṭibbāʾ (“Epistle of the Banquet of Physicians”)
therefore datable to the second half of the twelfth in the L.A. Mayer Museum of Islamic Art in
century.72 Unfortunately, given the nature of this Jerusalem (ms. 39–69), produced in Egypt or Syria
manuscript’s paintings, which are of a technical ca. 1300;75 a Qazwīnī ʿAjāʾib al-Makhlūqāt in the
and non-narrative type, it is impossible to make Museum of Islamic Art in Doha (Ms. 647), possi-
any fruitful comparison with the Kalīla wa-Dimna. bly dated to early fourteenth-century Syria;76 an
al-Jāḥiẓ Kitāb al-Ḥayawān (“Book of Animals”),
probably from Egypt or Syria ca. 1315, in the
7 Mamluk Painting Ambrosiana Library in Milan (Ar. 140 Inf.);77 an
Ibn Bakhtīshūʿ Kitāb Manāfīʿ al-Ḥayawān (“Book
Just as the proposal for an early Ayyubid dat-
ing for this earliest illustrated Kalīla wa-Dimna 73 L.A. Mayer, “A Hitherto Unknown Damascene Artist,”
manuscript relies on similarities with other, dated Ars Islamica 9 (1942): 168; also D.S. Rice, “A Miniature in
manuscripts, so, too, parallels can be drawn to an Autograph of Shihāb al-Dīn Ibn Faḍlallāh al-ʿUmarī,”
Bulletin of the School of Oriental and African Studies 13,
help situate the three fourteenth-century Mamluk
no. 4 (1951): 856–67; 862–63.
successors under consideration in this article (the 74 Oleg Grabar, The Illustrations of the Maqamat (Chicago:
manuscript in Cambridge, Corpus Christi Library, University of Chicago Press, 1984), 14–15.
MS 578, see footnote 70 and Fig. 4.18, is of uncertain 75 Eva Baer, “The Illustrations for an Early Manuscript of
Ibn Butlan’s Daʿwat al-aṭibbāʾ in the L.A. Mayer Memorial
in Jerusalem,” Muqarnas 19 (2002): 1–11.
fol. 52v. The colophon of this manuscript (on fol. 136r) 76 Stefano Carboni and Anna Contadini, “An Illustrated
gives a date to 23 Jumādā al-Ākhira 7** (the second and Copy of al-Qazwīnī’s ‘The Wonders of Creation’,” in
third digits are damaged) corresponding to sometime Sotheby’s Art at Auction: 1989–1990, ed. Sally Prideaux
in the 1300s. On a flyleaf, another Arabic inscription (London: Sotheby’s Publications, 1990), 228–33.
gives the date to end of the month of Rabīʿ al-Awwal 77 Oscar Löfgren and Carl J. Lamm, Ambrosian Fragment
848/March 1444. Further study is needed to ascertain a of an Illuminated Manuscript Containing the Zoology of
more exact date for this manuscript. al-Ğāḥiẓ (Uppsala: Lundequistska, 1946); Oscar Löfgren
71 See discussion in Contadini, “Ayyubid Illustrated and Renato Traini, Catalogue of the Arabic Manuscripts
Manuscripts”. in the Biblioteca Ambrosiana (Vicenza: N. Pozza, 1975–
72 Fol. 208v: Bi-rasm khizānat al-malik al-nāṣir … ṣalāḥ 81), vol. 1, no. LXX, 50 and col. pls; Anna Contadini,
al-dunyā wa-’l-dīn. For this manuscript, see Contadini, “Two Illustrated Mamluk Manuscripts in the Biblioteca
“Ayyubid Illustrated Manuscripts,” 180–82, pl. 9.1 and Ambrosiana: The al-Jāḥiẓ Kitāb al-Ḥayawān and the Ibn
col. pls. XVII–XXIII, and previous literature on the Buṭlān Risalat Dawʿat al-Aṭibbāʾ,” in Le arti del Vicino
manuscript and its author. Oriente / Le scienze del Vicino Oriente, ed. R.B. Finazzi,
Intertextual Animals 119
FIGURE 4.18 “The merchant and the musician” (right, fol. 24v) and “The dog and the bone” (left,
fol. 25r). Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Egypt or Syria, late fourteenth century (?).
Cambridge, Corpus Christi Library, MS 578
figure 4.19 Qurʾan produced in the Great Umayyad Mosque in Damascus, ca. 731–741H/1330–40, London,
Khalili Collection, QUR 807, fols. 295v-296r
Intertextual Animals 121
figure 4.20 a) Ibn Ẓafar al-Siqillī, Sulwān al-Muṭāʿ fī ʿUdwān al-Atbāʿ (“The Consolation of the Prince with Hostile
Subjects”), probably Damascus, second quarter of the fourteenth century. Doha, Museum of Islamic Art,
MS.27.1999, fol. 1r (After Melikian-Chirvani, Sulwān al-Muṭāʿ); b) Al-Ḥarīrī, Maqāmāt (“Assemblies”), probably
Damascus, dated 734H/1334. Vienna, Nationalbibliothek, A.F. 9, fol. 8v; c) Ibn Bakhtīshūʿ, Kitāb Manāfīʿ
al-Ḥayawān (“Book of the Usefulness of Animals”), probably Damascus, dated 755H/1354. San Lorenzo del
Escorial, Biblioteca Real, Ar. 898, fol. 80r; d) Al-Ḥarīrī, Maqāmāt (“Assemblies”), probably Damascus, dated
738H/1337. Bodleian Marsh 458, fol. 45r
122 Contadini
figure 4.22 Ibn Ẓafar al-Siqillī, Sulwān al-Muṭāʿ fī ʿUdwān al-Atbāʿ (“The Consolation of the Prince with Hostile Subjects”),
probably Damascus, second quarter of the fourteenth century. Doha, Museum of Islamic Art, MS.27.1999,
fol. 44r (After Melikian-Chirvani, Sulwān al-Muṭāʿ)
Mamluk glasses of the fourteenth century – the produced in Syria, with Aleppo and Damascus,
enamels and gold are applied with a brush on the for example, being mentioned in the sources and
surface of the glass.90 More importantly, there is a attested to by material culture from as early as the
similar visual, coloristic approach; there are strong late twelfth century.91 (Fig. 4.21)
parallels in the color range, with gold, blue, red, The style and spatial approach of Mamluk
and white particularly prominent. Such glass was paintings can differ somewhat from manuscript
to manuscript, but there are certain character-
istics that are common to several of them – for
90 Such as with a vase, fourteenth century, now in Kuwait, example, a plain background and an absence of
the Al-Ṣabāḥ Collection, Dār al-Āthār al-Islāmiyyah,
frames around the paintings, as in the case of the
LNS 69 G; see Esin Atıl, ed., Arte islamica e mecenatismo:
Tesori dal Kuwait (New York: Rizzoli, 1990), 212–13, cat. three Kalīla wa-Dimna92 and other manuscripts
64. For connections between manuscript painting and
gilded and enameled glass, see Anna Contadini, “Text
and Image on Middle Eastern Objects: The Palmer Cup 91 Contadini, “Urban Bourgeoisie,” 433; Contadini, “Text
in Context,” in A Rothschild Renaissance: A New Look at and Image,” 140–41.
the Waddesdon Bequest in the British Museum, Research 92 The 755H/1354 Bodleian Pococke 400 Kalīla wa-Dimna
Publications 212, ed. Pippa Shirley and Dora Thornton has thin red frames around the whole text block, but
(London: British Museum, 2017), 124–45. not around the paintings.
124 Contadini
of different texts, including the only known illus- and the position of the ṭirāz band (Figs 4.20a and
trated copy of al-Jāḥiẓ’s Kitāb al-Ḥayawān, the b). All these make the two particularly close, and
aforementioned copy in the Ambrosiana Library although confirmation will require further inves-
in Milan (Ar. 140 Inf.), datable to ca. 1315. tigation and technical examination, I should like
The group of four, on the other hand, is distin- to venture the hypothesis that they may have been
guished by paintings being spatially organized the work of the same painter(s).
within blue frames with surrounding decorative In contrast, the parallels between the Sulwān
motifs, and by having solid gold backgrounds. In and the three Kalīla wa-Dimna manuscripts are
addition, these manuscripts are innovative, the primarily textual. The Sulwān, in fact, adopts the
Sulwān particularly so, in terms of both style and same five-part structure; the text, as in the Kalīla
the overall concept of the relationships between wa-Dimna, is ethical and didactic; and it has
the various elements of the composition. For more recourse to the same vehicle of the animal fable.
specific similarities within the group, we may cite In this, it provides abundant material for compara-
the rendering of the gazelle and antelope in the tive study of the rendering of animals, whether in
Sulwān (Fig. 4.22) and the way in which one turns isolation or interacting, across manuscripts of this
its head back to the other, which closely resem- period.
bles that in the miniature of the gazelles in the
755H/1354 Escorial Manāfi‘ (Ar. 898, fol. 36v); the
painting of the wild and tame elephants, where the 8 Cultural Attitudes and Analytical
colorful lotus flowers are another feature shared Paintings
with the other three manuscripts; and the scene of
Nuʿmān, Arab king of Hira, greeting the poet ʿAdī There is, further, a parallel to be seen in the social
b. Zayd in his tent, with horse and camel heads environments in which these texts circulated,
projecting from behind the tent, which is very for just as a Kalīla wa-Dimna copy was owned by
reminiscent of the scene in the Nationalbibliothek a judge and a text derived from it was placed in
Maqāmāt of 734H/1334 (A.F. 9), fol. 87v. It is clear a madrasa library, so there is evidence that the
that these four manuscripts were produced in Sulwān, despite its princely double frontispiece,
a common artistic environment, probably in indicating that it could have been commissioned
Damascus. Further study will be needed to clarify by or for a prince, belonged to a religious institu-
the relationships, and the differences, between tion, and later to a qāḍī teaching at the famous
the diverse types of fourteenth-century Mamluk ʿAdiliyya madrasa in Aleppo.
painting that they and the other manuscripts We do not have sufficient information to decide
exhibit. whether it was originally intended for one envi-
The Sulwān is the only manuscript to lack a ronment or the other, and it could have moved
date, but the close stylistic connections with the between them: what is clear is that appreciation of
other three place it squarely within the second the frequent subtlety of the relationship between
quarter of the fourteenth century. To be more text and image required a high degree of cultural
precise, I would suggest a closeness in date to alertness, and that this was just as likely to be
the Nationalbibliothek Maqāmāt of 734H/1334 found among religious and other scholars as it was
(A. F. 9) because of the additional similarities of in a court environment. The paintings, it should
the human figures, as we can see from details be emphasized, are not simply descriptive, but are
such as the treatment of the stocky figures and also analytical, interpreting the text and making
especially the heads, which have almost identi- many subtle references to it, even when they take
cal features, and from elements of the costume, on the familiar garb of a reception or a tent scene.
such as the turban, in the way that it is folded Other scenes, such as the wild elephant dressed
Intertextual Animals 125
as a warrior elephant, or that depicting absence, culture, and their illustrations formed an integral
are strikingly original, and imply a highly sophisti- part of the world of visual representation.
cated readership.
A further aspect to be considered with regard
to the Sulwān is what it reveals about attitudes Bibliography
to figurative painting. If it was used in religious,
madrasa teaching and owned by a qāḍī, its illus- À l’ombre d’Avicenne : La médicine au temps des califes.
trations cannot have been regarded as objection- Paris: Institut du Monde Arabe; Gand: Snoek-Ducaju
able, and I am inclined to hypothesize that they & Zoon, 1996.
not only provided a useful aid to the interpreta- Abdel Haleem, Muhammad A.S., trans. The Qurʾan.
tion of the text, but also an element of wonder and Oxford: Oxford University Press, 2004.
contemplation. It would appear that at least some Abu-’l-Fażl b. Mubārak. ʿIyār-i dānish. Dushanbe:
of the more religious sections of Mamluk society Dānish, 1988.
were not opposed to figurative painting, or could al-Masʿūdī, ʿAlī ibn al-Ḥusayn. Murūj adh-dhahab wa
at least tolerate it in an ethically driven text when maʿādin al-jawhar / Les prairies d’or, translated by
it was associated with a type of narrative for which Charles Barbier de Meynard and Abel Pavet de
there was a tradition of illustration. It would not Courteille. Paris: Imprimerie impériale, 1863.
have been out of place in the Ashrafiyya Library al-Ṭabarī, Abū Jaʿfar Muḥammad ibn Jarīr. Annales quos
catalog discussed above, which lists several illus- scripsit Abu Djafar Mohammed Ibn Djarir at-Tabari.
trated manuscripts. Edited by Michael Jan de Goeje. Leiden: Brill, 1881.
All this suggests that we need to explore con- Atıl, Esin. Kalila wa Dimna. Fables from a Fourteenth-
texts of production, consumption, and reception Century Arabic Manuscript. Washington, DC: Smith
further, as well as relationships with other manu- sonian Institution Press, 1981.
scripts, in order to arrive at a better understand- Atıl, Esin, ed. Arte islamica e mecenatismo: Tesori dal
ing of the significance and agency of illustrated Kuwait. New York: Rizzoli, 1994.
Kalīla wa-Dimna manuscripts. This implies a more Aydin, Karen and Birgit Middendorp. “Ibn al-Muqaffaʿ’s
comprehensive study of the surviving body of Kalīla wa-Dimna und der Codex arabicus 616 der
twelfth- to fourteenth-century illustrated manu- Bayerischen Staatsbibliothek München.” In Tierisch
scripts than could be undertaken here, but at moralisch: Die Welt der Fabel in Orient und Okzident.
least the discussion so far of the scope and aims Edited by Birgit Middendorp, 111–115. Oldenburg:
of these early Kalīla wa-Dimna manuscripts and Landesmuseum Natur und Mensch, 2009.
their connections with other manuscripts leads Baer, Eva. “The Illustrations for an Early Manuscript
to the conclusion that they, too, belong to a style of Ibn Butlan’s Daʿwat al-aṭibbāʾ in the L.A. Mayer
world that reveals cross-pollination, at the same Memorial in Jerusalem.” Muqarnas 19 (2002): 1–11.
time as it exhibits differences, between the various Baker, Colin F. Qurʾan Manuscripts: Calligraphy, Illumi-
centers of Ayyubid, Abbasid, and early Mamluk nation, Design. London: British Library, 2007.
production. Along with works on astronomy, med- Benfey, Theodor. Pantschatantra: fünf Bücher indischer
icine, pharmacology, and the wonders of nature, Fabeln, Märchen und Erzählungen. Leipzig: Brock
such manuscripts served the complex cultural haus, 1859.
needs of an urban clientèle stretching from the Berlekamp, Persis. Wonder, Image, and Cosmos in
court through the scholarly community, with its Medieval Islam. New Haven, CT: Yale University
various specialisms, and beyond. Kalīla wa-Dimna Press, 2011.
and other narrative works with an ethical con- Bothmer, Hans-Caspar Graf von. Kalila und Dimna:
tent, such as the Sulwān, were, together with the Ibn al-Muqaffaʿ’s Fabelbuch in einer mittelalterlichen
Maqāmāt, indispensable elements of intellectual Bilderhandschrift: Cod. arab. 616 der Bayerischen
126 Contadini
Eastwick, Edward B. trans. The Anvár-i Suhailí: Or, the Hirschler, Konrad. Medieval Damascus, Plurality and
Lights of Canopus, Being the Persian Version of the Diversity in an Arabic Library. The Ashrafīya Library
Fables of Pilpay, or the Book of “Kalílah and Damnah”. Catalogue. Edinburgh: Edinburgh University Press,
Hertford: Stephen Austin, 1854. 2016.
Edgerton, Franklin. “Genealogical Table of the Hourihane, Colum P. ed. The Grove Encyclopedia of
Panchatantra.” In The Ocean of Story. Edited by Medieval Art and Architecture. Oxford: Oxford
N.M. Penzer, vol. 5, 232–42. London: Chas J. Sawyer, University Press, 2012.
1926. Hunt, Lucy-Anne. “Christian-Muslim Relations in Paint-
Farès, Bishr. Le livre de la thériaque. Cairo: Imprimerie ing in Egypt of the Twelfth to Mid-Thirteenth Cen-
de l’Institut français d’archéologie orientale, 1953. turies: Sources of Wall-Painting at Deir es-Suriani
Firdawsī, The Epic of the Kings. Shah-Nama, the National and the Illustration of the New Testament MS Paris,
Epic of Persia by Ferdowsi, translated by Reuben Levy, Copte-Arabe 1/Cairo, Bibl. 94.” In Byzantium, Eastern
revised by Amin Banani. London: Arkana, 1990. Christendom and Islam: Art at the Crossroads of the
Gabrieli, Francesco. “Ibn al- Muḳaffaʿ.” In Encyclopaedia Medieval Mediterranean. Edited by Lucy-Anne Hunt,
of Islam, second edition. Edited by P. Bearman et al. vol. 1, 205–81. London: Pindar Press, 1998, 2000.
Leiden: Brill, 1960–2007. Hunt, Lucy-Anne. “The Commissioning of a Late
George, Alain. “Orality, Writing and the Image in the Twelfth Century Gospel Book: The Frontispieces of
Maqāmāt: Arabic Illustrated Books in Context.” Art MS Paris, Bibl. Nat. Copte 13.” In Byzantium, Eastern
History 35, no. 1 (2012): 10–37. Christendom and Islam: Art at the Crossroads of the
Grabar, Oleg. The Illustrations of the Maqamat. Chicago: Medieval Mediterranean. Edited by Lucy-Anne Hunt,
University of Chicago Press, 1984. vol. 1, 115–57. London: Pindar Press, 1998, 2000.
Grube, Ernst J., ed. A Mirror for Princes from India: Hunt, Lucy-Anne. “A Christian Arab Gospel Book: Cairo,
Illustrated Versions of the Kalilah wa Dimnah, Coptic Museum MS Bibl. 90 in Its Mamluk Context.”
Anvar-i Suhayli, Iyar-i Danish, and Humayun Nameh. Mamluk Studies Review 13, no. 2 (2009): 105–33.
Mumbai: Marg, 1991. Ibn Ḥajar al-ʿAsqalānī, Aḥmad ibn ʿAlī. Al-Durar
Guesdon, Marie-Geneviève and Annie Vernay-Nouri, al-kāmina fī aʿyān al-miʾah al-thāminah. Hyderabad:
eds. L’Art du livre arabe. Du manuscrit au livre d’artiste. Maṭbaʿat Majlis Dāʿirat al-Maʿārif, 1929–32.
Paris: Bibliothèque nationale de France, 2001. Ibn al-Nadīm, Muḥammad ibn Isḥāq. Fihrist, Arabic
Harris, Joseph trans. The Instructive and Entertaining edition. Beirut: Dār al-maʿrifa li-l-ṭibāʿa wa-’l-nashr,
Fables of Pilpay: An Ancient Indian Philosopher. 1978.
London: S. Birt and D. Browne, 1747. Ikhwān al-Ṣafāʾ. The Case of the Animals versus Man
Hartmann, Angelika. “al-Nāṣir li-Dīn Allāh, Abū ’l-ʿAbbās before the King of the Jinn: A Translation from the
Aḥmad.” In Encyclopaedia of Islam, second edition. Epistles of the Brethren of Purity. Translated by
Edited by P. Bearman et al. Leiden: Brill, 1960–2007. Richard J.A. McGregor and Lenn Evan Goodman.
Hartmann, Angelika. An-Nāṣir li-Dīn Allāh (1180–1225): Oxford: Oxford University Press, 2009.
Politik, Religion, Kultur in der späten ʿAbbasidenzeit. Irwin, Robert. “The Arabic Beast Fable.” Journal of the
Berlin; New York: De Gruyter, 1975. Warburg and Courtauld Institutes 55 (1992): 36–50.
Hertel, Johannes. Das Pañcatantra, seine Geschichte James, David. “Some Observations on the Calligrapher
und seine Verbreitung. Leipzig; Berlin: B.G. Teubner, and Illuminators of the Koran of Rukn al-Din
1914. Baybars al-Jashnagir.” Muqarnas 2 (1984): 147–57.
Hirschler, Konrad. The Written Word in the Medieval James, David. Qurʾans of the Mamluks. London:
Arabic Lands: A Social and Cultural History of Alexandria Press, 1988.
Reading Practices. Edinburgh: Edinburgh University James, David. The Master Scribes: Qurʾans of the 10th to
Press, 2012. 14th centuries AD. The Nasser D. Khalili Collection of
128 Contadini
Islamic Art. Volume II. London: The Nour Foundation Minūvī, Mujtabā, ed. Tarjumah-i Kalīlah va Dimnah.
in association with Azimuth Editions and Oxford Tehran: Dānishgāh-i Tihrān, 1964.
University Press, 1992. Muḥammad ibn ʿAbd Allāh al-Bukhārī, Dāstānhā-yi
Kerner, Jaclynne. “Art in the Name of Science: The Bīdpāy / tarjumah-i bih taṣḥịh. Edited by Parvīz
Kitāb al-diryāq in Text and Image.” In Arab Painting: Nātil Khānlarī and Muḥammad Rawshan. Tehran:
Text and Image in Illustrated Arabic Manuscripts. Khvārazmī, 1982.
Edited by Anna Contadini, 25–39. Leiden: Brill, 2007 Najm, Muḥammad Yūsuf. Kitāb muḍāhāt kalīla wa
(revised edition 2010). dimna bimā ashbahahā min ashʿār al-ʿarab. Beirut:
Khalidov, Anas B. “Rasāʾil al-rusul wa aʿmāluhum Dār al-Thaqāfa, 1961.
(The Epistles and Acts of the Apostles).” In Yuri A. Naṣrullāh Munshī, Abu-’l-Maʿālī. Kalila and Dimna.
Pétrosyan et al., Pages of Perfection. Islamic Paintings Translated by Wheeler Thackston. Indianapolis, IN:
and Calligraphy from the Russian Academy of Hackett Publishing, Inc., 2019.
Sciences, St. Petersburg. Lugano; Milan: ARCH Natif, Mika. “What the Fox Told the Snake: Illustrated
Foundation, Electa, 1995. Animal Fables from Akbar’s Age.” In Pearls from
Kitâb al-Diryâq (Thériaque de Paris), facsimile edition Water. Rubies from Stone. Studies in Islamic Art
of ms. Arabe 2964 in the Bibliothèque nationale, in Honor of Priscilla Soucek, Part II, special issue,
Paris. Sansepolcro: Aboca Edizioni, 2010. Artibus Asiae 67, no. 1 (2007): 55–73.
Latham, J. Derek. “Ebn Al-Moqaffaʿ, Abū Moḥammad Natif, Mika. “The SOAS Anvār-i Suhaylī: The Journey of
ʿAbd-Allāh Rōzbeh.” Encyclopaedia Iranica VIII/1, a ‘Reincarnated’ Manuscript.” Muqarnas 25 (2008):
39–43, http://www.iranicaonline.org/articles/ebn-al 331–58.
-moqaffa. North, Thomas. The Moral Philosophy of Doni. Popularly
Leroy, Jules. Les manuscrits syriaque à peintures, con- Known as The Fables of Bidpai. Edited by Donald
servés dans les Bibliothèques d’Europe et d’Orient. Beecher, John Butler, and Carmine di Biase. Ottawa:
Paris: Geuthner, 1964. Dovehouse Editions, 2003.
Leroy, Jules. “Un Évangéliaire Arabe de la Bibliothèque O’Kane, Bernard. Early Persian Painting: Kalila and
de Topqapi Sarayi à Décor Byzantin et Islamique.” Dimna Manuscripts of the Late Fourteenth Century.
Syria 44 (1967): 119–30. London: I.B. Tauris, 2003.
Leroy, Jules. Les manuscrits coptes et coptes-arabes illus- O’Kane, Bernard. “Der Kreis schließt sich: Illustrationen
trés. Paris: Geuthner, 1974. aus Kalīla wa-Dimna von Indien zum Mittleren Osten
Löfgren, Oscar and Carl J. Lamm. Ambrosian Fragment und wieder zurück.” In Von listigen Schakalen und
of an Illuminated Manuscript Containing the Zoology törichten Kamelen: Die Fabel in Orient und Okzident.
of al-Ğāḥiẓ. Uppsala: Lundequistska, 1946. Edited by Mamoun Fansa and Eckhard Grunewald,
Löfgren, Oscar and Renato Traini. Catalogue of the 127–48. Wiesbaden: Reichert, 2008.
Arabic Manuscripts in the Biblioteca Ambrosiana. Olivelle, Patrick, trans. The Pañcatantra: The Book of
Vicenza: N. Pozza, 1975–81. India’s Folk Wisdom. Oxford: Oxford University Press,
Marshak, Boris. Legends, Tales, and Fables in the Art of 1997.
Sogdiana. New York: Bibliotheca Persica Press, 2002. Ouseley, Jasper Walker Joseph, ed. Anvār-i Suhelī or
Mayer, L.A. “A Hitherto Unknown Damascene Artist.” Lights of Canopus being the Persian version of the
Ars Islamica 9 (1942): 168. Fables of Bīdpaī by Husain Vāiz̤ Kāshifī. Hertford:
Melikian-Chirvani, Asadullah Souren. Sulwān al-Muṭāʿ Stephen Austin, 1851.
fī ʿUdwān al-Atbāʿ, A Rediscovered Masterpiece of Pancaroğlu, Oya. “Socializing Medicine: Illustrations of
Arab Literature and Painting, with a translation by the Kitāb al-diryāq.” Muqarnas 18 (2001): 155–72.
Michele Amari. Kuwait: Al-sharika al-thulāthīya lil- Pellecchia, Linda. “From Aesop’s Fables to the Kalila
nashr, 1985. wa-Dimna: Giuliano da Sangallo’s Staircase in the
Intertextual Animals 129
Gondi Palace in Florence.” I Tatti Studies, Essays in Handschrift Codex arabicus 616 der Bayerischen
the Renaissance no. 14–15 (2011–12): 137–207. Staatsbibliothek in München.” In Von listigen
Pilpay. The Fables of Pilpay. London: Baldwin, Cradock, Schakalen und törichten Kamelen: Die Fabel in Orient
and Joy et al., 1818. und Okzident. Edited by Mamoun Fansa and Eckhard
Raby, Julian. “The Earliest Illustrations to Kalilah Grunewald, 99–114. Wiesbaden: Reichert, 2008.
wa Dimnah.” In A Mirror for Princes from India: Ruymbeke, Christine van. Kāshefi’s Anvar-e Sohayli:
Illustrated Versions of the Kalilah wa Dimnah, Anvar-i Rewriting Kalila wa-Dimna in Timurid Herat. Leiden:
Suhayli, Iyar-i Danish, and Humayun Nameh. Edited Brill, 2016.
by Ernst J. Grube. Mumbai: Marg, 1991. Ruymbeke, Christine van. “Authorship, Ownership and
Rebhan, Helga. “Collecting Islamic Manuscripts at the Rewriting: Vāʿiz Kāshifī and Abū’l-Fazl b. Mubārak
Munich Court Library in the Nineteenth Century: within the Hereditary Line of Kalīla wa-Dimna
An Acquisition History.” In Manuscripts, Politics Authors.” Jerusalem Studies in Arabic and Islam 45
and Oriental Studies: Life and Collections of Johann (2018): 181–210.
Gottfried Wetzstein (1815–1905) in Context. Edited Sarma, Visnu, trans. The Pañcatantra. New Delhi:
by Boris Liebrenz and Christoph Rauch, 197–227. Penguin Books, 1993.
Leiden: Brill, 2019. Sayyid, Ayman Fuʾād, ed. The Fihrist of al-Nadīm.
Rice, D.S. “A Miniature in an Autograph of Shihāb al-Dīn London: Al-Furqān Islamic Heritage Foundation,
Ibn Faḍlallāh al-ʿUmari.” Bulletin of the School of 2009.
Oriental and African Studies 13, no. 4 (1951): 856–67. Seyller, John. “The School of Oriental and African
Riedel, Dagmar. “Kalila Wa Demna. i. Redactions Studies Anvār-i Suhaylī: The Illustration of a De Luxe
and Circulation”, Encylopaedia Iranica XV/4, 386– Mughal Manuscript.” Ars Orientalis 16 (1986): 119–51.
395, http://www.iranicaonline.org/articles/kalila Shaw, Sarah, trans. The Jātakas: Birth Stories of the
-demna-i. Bodhisatta. New Delhi: Penguin Books, 2006.
Rizkallah, Wafaa. “The Earliest Arab Illustrated Kalīla Touwaide, Alain. Farmacopea Araba Medievale
wa Dimna Manuscript (BN Arabe 3465): A Study of (Dioscorides De Materia Medica): Codice Ayasofia
Its Miniatures.” MA thesis, American University in 3703. Milan: Antea Edizioni, 1992–93.
Cairo, 1991. Vogel, Jean Philippe. Catalogue of the Archaeological
Roth, Sabrina. “Kalila und Dimna in München: Ibn Museum at Mathura. Allahabad: F. Luker, 1910.
al-Muqaffaʿ’s Kitāb Kalīla wa-Dimna und die
5
Histoires de transmissions : un manuscrit arabe du xiiie siècle
et ses copies tardives
Annie Vernay-Nouri
Il se rattache à une famille de manuscrits arabes vert de style oriental avec un rabat, un motif cen-
qui se différencie par son texte et son cycle icono- tral et des écoinçons en marqueterie de cuir coloré
graphique de celui des manuscrits mamelouks et fut ajoutée en 194724.
du manuscrit de Rabat et de sa copie19. Sur les cent Le manuscrit mesure fermé 290 sur 220 mm
quarante-six feuillets qu’il contient, vingt-deux ont pour une épaisseur de 50 mm côté tranche. Ses
été remplacés à des époques diverses et pas moins feuillets ont été largement rognés. Il compte sept
de sept copistes y ont œuvré. Parmi les quatre- sortes de papiers qui correspondent chacun à un
vingt-dix-huit miniatures, huit ont été refaites copiste différent25. Toutes les pages sont enduites
tardivement en début et en fin d’ouvrage20 ; elles en surface d’une charge minérale blanche des-
sont attribuables, comme nous le démontrerons tinée à recevoir l’écriture et le décor. La partie la
plus loin, à deux artistes différents21. En outre, de plus ancienne est copiée sur un papier oriental
nombreux repeints altèrent les peintures d’ori- comparable à celui que l’on trouve dans d’autres
gine. Tel qu’il se présente aujourd’hui, le manus- manuscrits du xiiie siècle26. Deux autres papiers
crit constitue un objet hétérogène dont seules les orientaux de mains différentes ont remplacé à
composantes anciennes ont suscité l’attention une époque ancienne des pages manquantes. Tous
des chercheurs. trois se rattachent à la production papetière égyp-
Plus aucune indication ne subsiste sur l’état tienne ou syrienne du xie au xive siècle27 ; l’espa-
dans lequel le volume est arrivé à la Bibliothèque cement des vergeures n’est pas discernable du fait
royale. Le manuscrit a sans doute été démem- de leur grande détérioration.
bré pour être présenté en feuillets isolés lors de Le reste des feuillets est copié sur papier fili-
l’exposition « L’art de l’Iran, l’ancienne Perse et grané qu’on trouve au début et à la fin (là où se
Bagdad » qui s’est tenue en 1938 à la Bibliothèque placent les peintures remplacées) mais aussi sur
nationale à Paris22. Soixante-deux miniatures y des feuillets dispersés au sein du volume. Des res-
étaient regroupées en deux panneaux théma- taurations partielles figurent également sur un
tiques23. L’exposition finie, l’ensemble des feuillets certain nombre de pages, sans qu’on puisse iden-
fut remonté sur onglets et une reliure d’art en cuir tifier ni les papiers ni la chronologie de ces réfec-
tions. La réglure est absente, hormis dans la partie
finale où l’on distingue nettement des lignes tra-
19 François de Blois, Burzōy’s Voyage to India and the
cées à la misṭara. Les réclames sont postérieures
Origin of the Book of Kalīlah Wa Dimnah (Londres :
Royal Asiatic Society, 1990). au texte ancien.
20 Ff.3, 3v, 22, 25 pour les peintures refaites en début de En dehors de la foliotation européenne ajou-
volume et ff.138v, 139v, 141, 143 pour celles de la fin. tée en France, le volume est doté d’une foliotation
21 Voir dans ce volume le chapitre 9, Nathalie Buisson copte qui a servi d’argument en faveur de l’origine
et Annie Vernay-Nouri, « Étude de la matière pic-
turale de six Kalīla wa Dimna et d’un Maqāmāt
égyptienne du manuscrit. L’observation atten-
(XIIIe-XVIIIe siècle) ». Benouniche et O’Kane attri- tive n’étaye pourtant cette hypothèse qu’impar-
buent cet ensemble à un seul artiste : pour la première, faitement. Les chiffres coptes sont visibles sur la
il s’agit du copiste d’un volume des Mille et Une Nuits
(Manchester, John Rylands Library, Arabic 646 [706])
auquel elle attribue aussi Paris, BnF, Arabe 3470. Pour 24 C’est la seule mention de restauration qui figure sur les
le second, c’est celui d’un autre manuscrit de Kalīla registres de la Bibliothèque.
wa-Dimna (Oxford, Bodleian Library, E.D. Clarke Or.9) : 25 Voir détail en tableau 4.1.
Benouniche, Le « Kalila et Dimna » de Genève, 215-216 ; 26 Par exemple, dans les trois manuscrits contemporains
O’Kane, Early Persian Painting, 218, 230. des Maqāmāt conservés à la BnF : Arabe 6094, Arabe
22 Paul Pelliot, Les arts de l’Iran, l’ancienne Perse et Bagdad 3929 et Arabe 5847.
(Paris : Bibliothèque nationale, 1938), 10-11. 27 Geneviève Humbert, « Le manuscrit arabe et ses
23 Panneaux XII et XVIII : Pelliot, Les arts de l’Iran, papiers », Revue des mondes musulmans et de la
125-126. Méditerranée, no 99-100 (2002) : 55-77.
Histoires de transmissions 133
tableau 5.1 Caractéristiques codicologiques du manuscrit Arabe 3465 (Paris, Bibliothèque nationale de France)
majorité des feuillets d’origine (à l’exception des L’écriture du manuscrit présente, selon les
pages où ils se trouvaient sur des zones rognées copistes, de grandes différences de qualité. Celle
ou abîmées) et sur les réfections en papier oriental de la partie ancienne, très soignée, est copiée à
ainsi que sur certains feuillets refaits en papier fili- l’encre marron dans un naskhī large et régulier.
grané. Ils sont néanmoins, sur ces derniers, d’une La graphie présente néanmoins quelques parti-
graphie différente et d’une encre plus noire que cularités orthographiques : dans les paires d’ho-
sur les feuillets d’origine et que sur les réfections mographes, la lettre qui ne porte pas de point
en papier oriental. La foliotation copte a donc été est signalée soit par une lettre miniature pla-
introduite en deux étapes : la première lors d’une cée en dessous ou au-dessus, soit par l’inversion
restauration ancienne28 postérieure à la copie des points diacritiques29. L’écriture des parties
originelle et la seconde complétant la première, refaites sur papier oriental est de moins bonne
contemporaine de l’ajout des peintures tardives en tenue. Celle sur papier occidental est irrégulière,
début de volume. Elle ne peut donc constituer une particulièrement au début du volume où le tracé
preuve formelle de la réalisation du manuscrit en manque de fermeté ; pour les dernières pages, elle
Égypte mais témoigne seulement de sa présence est plus aérée, avec des points rouges séparateurs.
précoce, dès les premières réfections. L’ensemble du manuscrit est vocalisé.
28 Antérieure au milieu du xvie siècle, époque à laquelle 29 François Déroche, éd., Manuel de codicologie des
le papier filigrané a majoritairement supplanté le manuscrits en écriture arabe (Paris : Bibliothèque natio-
papier oriental en Égypte et en Syrie. nale de France, 2000).
134 Vernay-Nouri
figure 5.1a Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3465, f. 114v
figure 5.1b Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Paris, BnF, Arabe 3465, f. 139v
sept cartouches où l’on peut lire les dix-huit titres. une vignette circulaire, ils sont surmontés de
Le décor s’agrémente dans la marge de vignettes demi-cercles (f.139v) ou de petits triangles (f.140v
tournées vers l’extérieur du livre. Copiés dans et 142v). (Figs. 5.1A et 5.1B)
des bandeaux plus maladroits, les trois derniers
titres34 dans les pages refaites finales contrastent
avec les précédents. Sur un fond de rinceau se 4 Les peintures anciennes
détachent les lettres en naskhī bleu, entourées
de petits motifs rouges et verts. Prolongés par Malgré un état inégal de conservation, les quatre-
vingt-dix miniatures anciennes restent l’un des
34 F.139, 140v, 142v. plus beaux témoignages de la peinture arabe du
136 Vernay-Nouri
xiiie siècle. Certaines sont restées intactes tandis tradition des frontispices princiers, le manuscrit
que d’autres portent la trace de retouches et de est l’un des rares Kalīla arabes à posséder cette
repeints assez grossiers. Peintes sans cadre ni fond, représentation placée à la fin des trois introduc-
elles occupent une large partie du feuillet (entre un tions, juste avant la table des chapitres35.
tiers et la moitié). Seul le frontispice (f.34) utilise Dans la plupart des miniatures, on distingue
la pleine page (Fig. 5.2). Il montre un prince assis nettement un tracé préparatoire rouge. La palette
entouré de deux autres personnages. Encadrée est riche et variée : composée d’une gamme
d’une tresse à décor d’arabesques, la peinture qui
porte la trace de plusieurs réfections est la seule à 35 L’autre se trouve à Cambridge, Parker Library, Corpus
posséder un fond de couleur. S’inscrivant dans la Christi College, MS 578, ff.1v-2.
Histoires de transmissions 137
figure 5.4 Le moine, le voleur, le renard et l’épouse du cordonnier, Paris, BnF, Arabe 3465, f. 55
importante de pigments, elle se caractérise par le d’herbe figure le sol et des plantes aux formes très
nombre de nuances apportées à chaque couleur36. travaillées structurent la composition. Les scènes
La peinture des décors est conforme aux codes d’intérieur, qui prennent place dans des formes
iconographiques du début du xiiie siècle et architecturales à base d’arcs polylobés, ne com-
montre de nombreuses similitudes avec celle portent que les éléments nécessaires à la compré-
des Maqāmāt de 1222. Les scènes d’extérieur sont hension des scènes, l’ameublement reste succinct :
signifiées par les éléments végétaux, une bande rideaux noués, tapis et coussins, quelques acces-
soires comme les coupes de fleurs ou les lampes,
36 Voir dans ce volume le chapitre 9, Buisson et Vernay- le trône à dais qui symbolise le pouvoir royal. Les
Nouri, « Étude de la matière picturale ». animaux sont dépeints avec beaucoup de vivacité
138 Vernay-Nouri
et d’expressivité. Les images ne portent pas de souligné de rouge. Des moustaches et des barbes
légendes37, mais des inscriptions d’une autre main d’un noir vif défigurent les visages d’origine. Ceux
mentionnent le nom des personnages ou des ani- des rois Dabshalīm et Khusraw ont tous été retou-
maux. Ces indications pourraient être destinées à chés et le blanc de leur face s’est oxydé et trans-
l’illustrateur et lui indiquer l’emplacement des dif- formé en marron. (Fig. 5.5)
férents éléments iconographiques, à moins d’être
des mentions plus tardives destinées au lecteur.
(Figs. 5.3 et 5.4) 5 Les peintures tardives du début
Les repeints portent principalement sur les
visages des personnages : les yeux sont souvent Huit peintures plus récentes remplacent les minia-
accentués d’un trait plus épais, le menton parfois tures manquantes du volume originel. Situées au
début de l’ouvrage, les quatre premières (P2)38 qui
figure 5.6 Alexandre consultant les astrologues, Paris, BnF, Arabe 3465, f. 3
illustrent la préface de ʿAlī b. al-Shāh forment un on ne possède que très peu d’exemples illustrés39
contraste étonnant avec les plus anciennes aux- (Fig. 5.6). La scène suivante (f.22) concerne la
quelles elles s’intercalent. L’image est mal centrée recherche du livre en Inde tandis que la dernière
et une esquisse non terminée est visible à l’envers (f.25v) montre une composition en pleine page sur
d’une peinture (f.22v). Très diluées, les couleurs, deux niveaux. Aucune indication de date ou de
limitées au vert, aux ocres et au gris, remplissent lieu ne figure sur ces pages et c’est en se tournant
maladroitement le dessin fait à l’encre noire. Le
nez, la bouche et les yeux, ainsi que les bras et
39 Les seuls exemples connus dans des manuscrits arabes
les jambes, sont tracés en rouge sur le papier en de Kalīla wa-Dimna figurent dans deux manuscrits
réserve. Les deux premières miniatures ornent conservés à Oxford, Bodleian Library : E.D. Clarke 09 et
l’histoire d’Alexandre (ff.3, 3v), un épisode dont Manchester, John Rylands Library, Arabic MSS 487[537].
140 Vernay-Nouri
figure 5.7
Histoire du roi Nuʿamān, Mille
et une Nuits, Manchester, John
Rylands Library, Arabic 646, f. 78v
vers deux autres manuscrits qu’on en apprend ressemblent étrangement aux nôtres. Le dessin
davantage. Articulés autour de l’histoire épique malhabile, la palette de couleurs, la graphie irrégu-
du roi Nuʿmān, ces volumes appartiennent au cor- lière, les motifs ornementaux y sont semblables et
pus des Mille et Une Nuits, un texte ne faisant pas on peut raisonnablement penser que c’est le même
d’ordinaire l’objet de peintures. L’un est conservé peintre copiste42 qui a œuvré dans les manuscrits
à Manchester40, l’autre à Tübingen41. Leurs de Paris et de Manchester. Produit vraisemblable-
miniatures, particulièrement celles du premier, ment en Égypte, le volume de Manchester y aurait
été acquis par Jean-Georges Varsy (1775-1859),
40 Manchester, John Rylands Library, Arabic MSS 646
[706], 263 f., 29 miniatures. 42 La mise en page de ces trois manuscrits laisse à pen-
41 Tübingen (Allemagne), Universitätsbibliothek M.a ser qu’un même intervenant a effectué la copie et les
VI.32, 210 f., 46 miniatures. peintures.
Histoires de transmissions 141
négociant, fin arabisant et collectionneur de dans toutes les études ultérieures. La nouvelle lec-
manuscrits qui y vécut près de dix-huit ans. (Fig. ture qu’en fait Ibrahim Akel44 « Ibn Abū al-Minā »
5.7) L‘orientaliste compléta de sa main les lacunes lui permet de rattacher ce scribe à une famille de
du texte et y ajouta la traduction française de copistes coptes active aux xviie et xviiie siècles et
certains passages. Dans le colophon (f.124) sont c’est cette interprétation que je suivrais.
précisés les noms du commanditaire al-muʿallim Présentant un découpage textuel un peu dif-
al-Miṣrī et du copiste. Alphonse Mingana, dans férent, le volume de Tübingen, bien que très
son Catalogue43, interprète ce nom comme Nasīm proche, montre des dissemblances dans le des-
b. Yuḥannā b. Abū al-Masā, transcription reprise sin des visages et dans la palette chromatique.
figure 5.8
Le saint homme et son
hôte, Paris, BnF, Arabe
3465, f. 139v
figure 5.9
Le saint homme et son
hôte, Paris, BnF, Arabe 3470,
f. 116v
Ayant appartenu à la bibliothèque d’Aḥmad laquelle a été copiée la majorité des Kalīla illustrés
al-Rabbāṭ al-Ḥalabī, un célèbre libraire damas- et datés46.
cène du xixe siècle, le livre possède toutefois des
indications à des endroits communs qui laissent
supposer que les deux manuscrits se sont trouvés 6 Les peintures tardives de la fin
ensemble à un moment de leur histoire45. Ils pour-
raient être issus d’un atelier commun ou bien être Les quatre autres peintures tardives se situent dans
la copie l’un de l’autre. Malgré l’absence d’éléments la partie refaite ; elles débutent au f.138 après le
tangibles de datation, on peut penser que les trois
ont été réalisés au cours du xviie siècle, période à
46 L’année 1708, date à laquelle Benoît de Maillet a quitté
45 Akel, vol. 2, 44. l’Égypte, constitue un terminus ante quem pour A3465.
Histoires de transmissions 143
bandeau de « La lionne et le chacal » et s’achèvent 3465. Deux volumes, Paris, BnF, Arabe 3470 et
au f.146v, à la fin du « Fils du roi et ses compa- Oxford, Bodleian Library, E.D. Clarke Or. 0948, s’en
gnons », le dernier chapitre du livre. Les minia- rapprochent tout particulièrement. Leur table des
tures, qui occupent une surface plus petite que chapitres est similaire et leurs cycles iconogra-
les précédentes (environ un tiers de la page), s’en phiques, malgré de légères différences, sont très
distinguent également par leur gamme chroma- proches. Ont-ils été copiés directement ou à partir
tique plus étendue et plus éclatante. En l’absence d’une copie de la même famille ? Quels rapports
de légendes, le nom de chaque personnage a été entretiennent-ils l’un et l’autre avec les parties
noté au-dessus en noir, peut-être à l’imitation des refaites de ce manuscrit ? Peuvent-ils nous aider à
peintures médiévales. Bien que le texte paraisse reconstituer partiellement l’histoire complexe du
complet, le copiste n’a étonnamment pas laissé de nôtre ? Que nous laissent-ils entrevoir de la pro-
colophon et termine sur la formule habituelle de duction artistique manuscrite dans les provinces
fin de chapitre : cela veut-il dire que le livre n’était arabes de l’Empire ottoman ?
pas terminé ? Une double marque de possession
non datée à la dernière page (f.146v), à lire comme 7.1 Paris, BnF, Arabe 3470
Milk kātibihi Miftāḥ Allāh b. Jād, pourrait signifier C’est plus de deux siècles après l’arrivée d’A3465
en outre que le possesseur Miftāḥ b. Jād est le qu’A3470 entre à la Bibliothèque royale. Trois
copiste de cette partie ; l’écriture n’est pas exac- autres Kalīla illustrés y figurent déjà49. L’ouvrage50
tement la même, mais il semble que les scribes a été acquis par Asselin de Cherville (1772-1822),
pouvaient avoir des graphies un peu différentes agent consulaire en Égypte entre 1806 et 182251.
selon le contexte. Les analyses physico-chimiques Riche de plus de mille cinq cents manuscrits orien-
et la comparaison stylistique montrent en tout cas taux, sa collection est achetée par la Bibliothèque
que ce peintre copiste est probablement celui d’un en 1833. Dans le catalogue bilingue dressé après
autre Kalīla, Paris, BnF, Arabe 3470, dont le cycle sa mort, la présence de neuf « Kalilé et Dimné »52
iconographique est très proche de la partie origi- prouve combien les manuscrits de cette œuvre
nale d’Arabe 346547. (Figs. 5.8 et 5.9) étaient alors recherchés. Vu le peu d’indications
figurant sur ce catalogue, il est difficile de savoir
sous quel numéro se cache notre volume.
7 Deux copies très proches d’Arabe 3465 Le manuscrit, dont on ignore l’aspect à son arri-
vée en France, compte aujourd’hui cent vingt-deux
L’examen des parties refaites du manuscrit montre feuillets. Il présente quantité de lacunes dont le
que, bien loin de se limiter à la période médiévale,
la tradition d’illustrer le livre de Kalīla wa Dimna 48 Abrégés en A3470 et Or.9.
se perpétue dans les provinces sous domination 49 Arabe 3467, Arabe 3472 et Arabe 3466 (figures non
ottomane au-delà de la chute des Mameluks. En exécutées).
recensant plus d’une vingtaine de manuscrits 50 Anciennes cotes : supplément arabe 1803, 974 (f.10v).
Notice dans Mac Guckin de Slane, Catalogue des
illustrés aux xviie et xviiie siècles, le programme
manuscrits arabes, 603, n° 3470.
de recherche a mis en lumière l’extrême diversité 51 Henri Dehérain, « Asselin de Cherville, drogman du
stylistique de ces ouvrages. Plusieurs d’entre eux consulat de France en Égypte et orientaliste », Journal
s’inscrivent néanmoins dans une tradition trou- des savants 14, no 4 (1916) : 176-187.
vant sa source dans la partie originelle d’Arabe 52 Paris, BnF, Arabe 4481 : « Inventaire des manuscrits de
la bibliothèque d’Asselin de Cherville, en arabe et en
français » consultable en ligne sur http://gallica.bnf.fr/
47 Voir chapitre 9, Buisson et Vernay-Nouri, « Étude de la ark:/12148/btv1b11003975g. Le numéro barré sur la page
matière picturale ». de garde « Asselin 1067 » est erroné.
144 Vernay-Nouri
décompte a été soigneusement noté au crayon fautives ou déformées (f.357, f.33 et 3558, f.4559 et
à papier par une main européenne53. En outre, f.61). Deux basmala chrétiennes sont d’une autre
de nombreux feuillets ont été reliés en désordre, main (f.4 et 88). Juste sous le jadval, figure une
probablement lorsqu’on a remonté l’ouvrage prière du copiste au lecteur dont les derniers
sur onglets et qu’il a été recouvert d’une reliure mots sont grattés60 et la mention, à côté, dans une
demi-dos cuir au chiffre de Louis-Philippe. Fermé, autre écriture, d’un scribe au patronyme chrétien
le volume a pour dimensions 320 × 235 mm pour (al-nāsiḥ Naṣrānī). On ne sait si elle est le fait de
une épaisseur de 35 mm ; chaque feuillet mesure notre copiste ou si elle est ultérieure. L’abondance
297 × 207 mm et le texte, sur dix-sept lignes, est de qualificatifs dépréciatifs dont il se qualifie ren-
encadré d’un jadval fait de deux fins traits bleus de voie en tout cas à un contexte chrétien61.
225 × 130 mm. À la double foliotation européenne Marianne Barrucand, la première, a relevé la res-
récente54 s’ajoute une plus ancienne en chiffres semblance du volume avec A346562 dans son orne-
coptes, montrant qu’il a été copié en Égypte. Grâce mentation et ses illustrations. Les couleurs vives
à celle-ci, on peut reconstituer l’ordre originel remplacent l’or et les différents motifs, tresses,
et les feuillets absents. Bien vocalisé, le texte est points groupés en triangle, demi-médaillons ou
copié à l’encre noire en naskhī sur un papier blanc vignettes peints en rouge, ocre et vert, transcrivent
lissé à filigrane tre lune. L’écriture assez régulière à leur manière le vocabulaire décoratif médiéval
se rapproche de celle des pages finales d’A3465.
Les points de séparation sont rubriqués ainsi que
les légendes. Les lignes sont tracées à la misṭara. Le
manuscrit possède également des réclames situées 57 Le premier vers, attribué souvent à al-Mutanabbī, ne
à la place habituelle au bas gauche du verso et sur- figure pourtant pas dans son Diwān. Le second vers est
montées parfois d’un petit signe rouge ; d’autres se attribué communément à ʿAlī b. Abī Ṭālib.
58 Ces citations imparfaitement copiées proviennent
trouvent en bas du recto (f.4, 5,9, 18), ce qui est plus toutes deux du livre d’al-Yamanī, Muḍāhat amthāl
surprenant. Des marques de collation (qūbila) Kalīla wa Dimna. La première est attribuée à al-Rabīʿ b.
figurent à intervalles réguliers, une pratique sou- Ziyād al-Ḥārathī (f.33), la seconde à Sawwār al-Ghanawī
vent présente dans les manuscrits chrétiens55. (f.35).
59 Cette citation provient du Tārīkh madīnat Dimashq
Acéphale, le volume débute au folio copte 4
d’Ibn ʿAsākir.
au milieu de l’histoire d’Alexandre56 ; il se ter- 60 « Ô lecteur, dans ce livre je te demande par la vérité
mine peu avant la fin du « Fils du roi et ses com- de la création des cieux et de la terre de prier pour le
pagnons » dont ne manquent que les dernières pardon des péchés du pauvre et impuissant copiste. »
lignes et le colophon. Écrites par plusieurs com- (trad. Loïc Lesvignes).
61 Gérard Troupeau, « Les colophons des manuscrits
mentateurs, les gloses marginales mêlent cor- chrétiens », in Scribes et manuscrits du Moyen-Orient,
rections du texte et citations poétiques souvent éd. François Déroche et Francis Richard (Paris :
Bibliothèque nationale de France, 1997), 223-231.
62 Marianne Barrucand, « Le Kalila wa Dimna de la
Bibliothèque royale de Rabat : un manuscrit illustré
53 « 3 feuillets entre f.7v et f.8 » ; « 3 entre f.11v et f.12 » ; « 2 il-khanide », Revue des études islamiques, numéro spé-
entre f.24v et f.25 » ; « 4 entre f.56v et f.57 » ; « 6 entre cial en l’honneur de Dominique Sourdel, n°54 (1986) :
f.67v et f.68 » ; « 2 entre f.80v et f.81 » ; mention en bas 17-51 ; Marianne Barrucand, « Un manuscrit arabe
du f.118v : « Il manque 25 feuillets et la fin du dernier illustré de Kalila wa-Dimna du XIIIe siècle et sa copie
chapitre ». ottomane », Archéologie islamique, no 2 (1991) : 81-95.
54 L’une commence au début, l’autre à la fin du volume. Une maîtrise a été soutenue sous sa direction par
55 Déroche, Manuel de codicologie des manuscrits en écri- Claire Levasseur, « Kalila wa Dimna : la relation entre
� ن ن ف ال � ن � ن � ا � ا ت ف
ture arabe, 103, ill. 149. le manuscrit illustré MS 3465 et sa copie ottomane MS
Incipit �كا ����ه
�� م�� �������س�ه ا ح��س�ا � م����ح�ه ل�ل�ه
م
56 3470 » (Mémoire de maîtrise, Université Paris IV, 1991).
Histoires de transmissions 145
tableau 5.2 Ordre reconstitué et lacunes du manuscrit Arabe 3470 (Paris, Bibliothèque nationale de France)
et ornent la table des chapitres, les dix-huit ban- lacunaires et celles en désordre permet, en la com-
deaux de titre mais aussi l’encadrement dans parant avec A3465, de rétablir les miniatures qui
lequel s’inscrivent les miniatures. ont disparues (annexe 2). La palette de couleurs
Ces dernières, au nombre de soixante-dix-huit, est décrite dans un article du même volume63.
occupent une surface de huit à treize lignes de
texte et débordent sur les marges latérales. Elles
contiennent une légende rubriquée d’une autre
main que l’ensemble du texte, ou parfois seule-
ment le nom du personnage ou de l’animal. La 63 Voir chapitre 9, Buisson et Vernay-Nouri, « Étude de la
reconstitution du manuscrit incluant les pages matière picturale ».
146 Vernay-Nouri
figure 5.10 Duel entre Poros et Alexandre, Oxford, Bodleian Library, E.D. Clarke Or. 9, f. 3v
Histoires de transmissions 147
figure 5.11 Duel entre Poros et Alexandre, Manchester, John Rylands Library, Arabic 487, f. 4v
7.2 Oxford, Bodleian Library, E.D. Clarke Clarke (1769-1822), un ecclésiastique anglais dont
Or. 0964 l’ex-libris figure sur le contreplat inférieur. Dans
Un deuxième manuscrit conservé à Oxford à la une lettre adressée en 1802 au R.P. Robert Malthus,
Bodleian Library sous la cote E.D. Clarke Or. 0965, il énumère les différents volumes qu’il a acquis et
présente lui aussi des caractéristiques très proches l’on peut reconnaître notre manuscrit, malgré l’er-
d’A3465. Ce volume est acheté entre 1799 et 1802 lors reur de langue, sous la mention « In Persian ……
d’un des voyages en Orient du R.P. Edward Daniel the work of Bidfai, or Pilpay ; containing the Fables
and Apologues known afterwards to the Greeks on
64 De Blois, Burzōy’s Voyage to India and the Origin of the the name of Aesop ». Bien que la lettre soit écrite
Book of Kalīlah Wa Dimnah, 72 ; Ernst J. Grube, éd., A
à Constantinople, il semble bien que les manus-
Mirror for Princes from India: Illustrated Versions of the
Kalilah Wa Dimnah, Anvar-i Suhayli, Iyar-i Danish, and crits arabes aient été acquis lors de son séjour en
Humayun Nameh (Bombay : Marg Publications, 1991), Égypte66. Il vendra quelques années plus tard sa
168 ; O’Kane, Early Persian Painting, 218.
65 Anciennes cotes : 18420 ; cod. CIV. Notice dans
Alexander Nicoll, éd., Catalogus sive notitia manus- 66 William Otter, The Life and Remains of the Rev. Edward
criptorum qui a cel. E.D. Clarke comparati in bibliotheca Daniel Clarke, LL.D. Professor of Mineralogy in the
bodleiana adservantur, (Oxonii : e typ. Clarendoniano, University of Cambridge (Londres : Printed for George
1815), vol. 2, 5-7. Cowie and Co. in the Poultry, 1824), 177.
148 Vernay-Nouri
figure 5.12 Fanzah refusant de revenir vers le roi, Oxford, Bodleian Library, E.D. Clarke Or. 9, f. 103
Histoires de transmissions 149
figure 5.13 Les chiens que la sobriété a empêchés de fuir leurs maîtres et d’être dévorés, Oxford, Bodleian Library,
E.D. Clarke Or. 9, f. 103v
collection à la Bodleian Library pour une somme manquants aux f.1 et 5-6, lacunes confirmées par
de mille livres67. l’observation des réclames ; la partie finale manque
Recouvert d’une reliure occidentale en cuir, le ainsi que le colophon.
manuscrit se trouve dans un état fort dégradé. Les Le texte, en naskhī noir, est copié sans trace de
décharges d’écriture et de peinture sur un papier réglure dans un cadre rouge de 237 × 147 mm pour
endommagé par de nombreuses taches d’eau en dix-sept lignes par page. Chaque feuillet mesure
rendent la lecture très difficile. D’une taille de 297 × 207 mm. Le papier porte un filigrane (une
300 × 210 mm, le volume comporte cent vingt-cinq couronne portée par deux lièvres (?) dressés, sur-
folios auxquels s’additionnent les pages de garde montée d’une étoile à six branches et à croissant)
en parchemin ajoutées en Angleterre. Due à un non identifié mais dont le motif de couronne se
seul scribe, la copie ne comporte aucune indi- rattache à la seconde moitié du xviie siècle. Le
cation. Antérieure à la foliotation occidentale manuscrit ne possède pratiquement pas de correc-
notée au crayon à papier, une autre numérotation tions ni de mentions marginales, hormis quelques
en chiffres indiens fait apparaître des feuillets marques de plume. Seules apparaissent quelques
annotations au crayon à papier en anglais (ff.27v,
67 En 1809, voir Falconer Madan et Herbert H.E. Craster, 28, 40). La ponctuation est notée par un triangle
A Summary Catalogue of Western Manuscripts in the formé de trois petits points rouges.
Bodleian Library at Oxford which have not hitherto
been catalogued in the quarto series, with references
Le manuscrit possède au f.1v un sarlohw com-
to the Oriental and other manuscripts (Oxford : At the posé d’un triangle central doré bordé de deux
Clarendon Press, 1895), 298-312. demi-triangles latéraux dans lesquels est écrit Kitāb
150 Vernay-Nouri
figure 5.14 Ibn Ẓafar, Sulwān al-muṭāʿ, Paris, BnF, Arabe 3511, détail f. 24v
Kalīla wa-Dimna. Sous la basmala copiée en noir de rouge, de bleu et de jaune verdâtre. De l’or et
se déclinent en rouge les quatre premières lignes du gris rehaussent malhabilement les images.
du texte, encadrées par de légers motifs végétaux Visages, pieds et mains sont directement dessinés
aux rosettes dorées. La table des chapitres (f.27v- en rouge. Tous les éléments architecturaux sont
28) et les dix-huit titres de chapitre enluminés68 tracés à la règle et au compas. Bien que n’utili-
tentent d’imiter ceux d’A3465. L’utilisation de l’or, sant aucun procédé de reproduction directe, ces
réservé d’ordinaire aux manuscrits plus précieux, illustrations se rattachent sans nul doute à A3465
contraste avec l’aspect peu soigné de l’enluminure ou à un exemplaire très proche. Pour Bernard
et de l’écriture. O’Kane, elles ont été copiées sur le premier, anté-
Quatre-vingt-seize miniatures ornent le volume, rieurement à 169869, date du départ d’Égypte de
accompagnées de légendes rubriquées que la Benoît de Maillet. En raison de la ressemblance
détérioration du manuscrit a rendues illisibles. Un entre les illustrations de l’« Histoire d’Alexandre »
double trait fin à l’alignement imparfait encadre dans les deux manuscrits, il attribue également au
chacune d’elles. Exécutés de manière assez gros- copiste d’Or.9 les miniatures refaites en début et
sière, les dessins à l’encre noire sont rehaussés fin d’A3465 remplaçant celles alors disparues70.
d’une peinture très délayée dans de pâles tonalités
68 Ff.14, 20, 28v, 37, 61, 71, 80, 93v, 97, 99, 102, 105v, 109v, 118, 69 Date erronée donnée par Bernard O’Kane.
119v, 120v, 122v. 70 Voir note 3.
Histoires de transmissions 151
71 Ff.3v, 8, 103r et v.
72 Manchester, John Rylands Library, Arabic MSS 487[537].
Le manuscrit dont le texte ne s’apparente à aucune
autre version présente un programme iconographique
exceptionnellement long. Ses miniatures ont été en
figure 5.18a Détails de Paris, BnF, Arabe 3465, f. 20v
Histoires de transmissions 155
seul à offrir quatre peintures consacrées à l’his- de la robe et celui du carrelage en perspective sont
toire d’Alexandre (Fig. 5.11). La seconde miniature peut-être dus à Clarke lui-même, après l’achat du
« Bidpay en prison » (f.8) est attribuable à l’artiste manuscrit.
qui a peint le registre inférieur du f.3v, en raison du L’élaboration complexe de ces quatre peintures
rendu très proche du visage et des mains. grâce à des calques ou à des poncifs suppose l’exis-
C’est un nouveau peintre qui intervient plus tence de plusieurs modèles que nous ne possé-
tardivement pour les deux illustrations inache- dons plus. Les ressemblances iconographiques ou
vées du « Roi et l’oiseau Qubbira » (transcrit ici stylistiques mises en lumière témoignent en outre
Fatra) au recto et verso du f.103. Copié d’une autre de la diversité des manuscrits à peintures qui cir-
main, ce feuillet porte des traces très perceptibles culaient aux xviie et xviiie siècle. Ce milieu était
de poudre blanche prouvant l’utilisation de pon- largement ouvert aux influences de la peinture
cifs pour reproduire le dessin. La première image occidentale par le biais, entre autres, des artistes
« Fanzah refusant de revenir vers le roi » se rat- chrétiens.
tache clairement au cycle iconographique de la
famille d’A346573, mais les nombreuses différences
qui existent entre les deux miniatures excluent 8 Des changements dans le style et
toute possibilité de copie directe sur A3465. l’iconographie ?
La seconde image (Fig. 5.13) « Les chiens que
la sobriété a empêchés de fuir leurs maîtres et Les similitudes profondes entre les cycles icono-
d’être dévorés » est étonnante, car son dessin ne graphiques des trois manuscrits confirment dans
se retrouve dans aucune des traditions iconogra- un premier temps le postulat selon lequel les
phiques de cette fable. La position des deux chiens peintures d’A3470 et d’Or.9 ont été copiées à partir
dressés sur leurs pattes arrière évoque pourtant d’A3465.
celle du chat et du lapin dans une autre histoire, Des changements significatifs apparaissent-ils
celle du « Rossignol, le lièvre et le chat »74 dans entre les cycles iconographiques d’A3465 et de
un manuscrit non pas arabe mais persan75 du ses copies ? Le plus notable concerne la dispa-
xive siècle. Par ailleurs, l’attitude de l’homme rition du frontispice (voir Fig. 5.2). Celui-ci s’ins-
n’est pas sans rappeler celle d’un personnage d’un crivait dans la tradition encore très prégnante au
Sulwān al-muṭāʿ76 non daté (voir Fig. 5.14). Très xiiie siècle des portraits de commanditaires, et le
européen, le tracé à l’encre sépia des plis sur le bas nom de celui-ci figurait peut-être sur la page ini-
tiale disparue. L’absence de frontispice au sein des
manuscrits peints post-mamelouks est révélatrice
partie copiées sur le volume persan (Adab Farsi 61).
Daté de 10 shaʿbān 83, son colophon pourrait, selon
des changements advenus dans les modes de pro-
O’Kane, être interprété non pas comme 1083H/1672 duction. Elle montre l’émergence à l’époque otto-
mais 1183H/1769 : voir O’Kane, Early Persian Painting, mane d’un nouveau public, moins fortuné mais
271. demandeur de manuscrits peints dont le nombre
73 La famille de manuscrits mamelouks n’illustre pas les
de volumes illustrés de Kalīla est le reflet. Hormis
mêmes épisodes de cette fable.
74 La miniature est reproduite dans : O’Kane, « Der Kreis la maladresse des miniatures, les transformations
schliesst sich: Illustrationen aus Kalīla wa-Dimna von restent surtout d’ordre stylistique. La végétation
Indien zum Mittleren Osten und wieder zurück », in peu variée présente des formes simplifiées, l’axia-
Von listigen Schakalen und törichten Kamelen : die Fabel lité qui structurait les compositions dans la copie
in Orient und Okzident, éd. Mamoun Fansa et Eckhard
Grunewald (Wiesbaden : Reichert, 2008), 138.
ancienne a désormais disparu, les motifs floraux
75 Le Caire, Dār al-kutub, Adab Farsi 61 : O’Kane, Early jouant désormais un simple rôle de remplissage
Persian Painting, 261-274. du fond. Repris dans la plupart des scènes, le cadre
76 Ibn Ẓafar, Sulwān al-muṭāʿ, Paris, BnF, Arabe 3511, f.24v. architectural, toujours à base de colonnettes et
Histoires de transmissions 157
d’arcs, s’est simplifié mais sert de support à une La petite calotte qui les coiffait a été remplacée par
ornementation envahissante, les formes structu- un couvre-chef tout en hauteur rehaussé parfois
relles s’alourdissant d’écoinçons, de dômes et de d’un plumet et qui pourrait s’apparenter à la coif-
frises ornées de petits motifs ou de rondeaux, par- fure des janissaires (voir Figs. 5.18A, 5.18B et 5.18.C).
ticulièrement dans A3470 (Figs. 5.15, 5.16 et 5.17). Revenons maintenant sur l’hypothèse d’une
Cette surcharge décorative contamine les auréoles copie directe des deux manuscrits tardifs sur
dorées ceignant la tête de plusieurs personnages, l’exemplaire le plus ancien. La présence de lacunes
transformées en une sorte de casque recouvert de mutilant A3470 et de pages refaites dans A3465
motifs. rend cette comparaison ardue, particulièrement
Le traitement des scènes humaines subit très pour les parties introductives où l’imbrication
peu de modifications, hormis là aussi une simpli- dans A3465 de peintures et de feuillets datant de
fication des corps et des postures. La figure de l’as- différentes époques se conjugue avec la dispari-
cète a gardé sa silhouette allongée et son manteau tion de pages dans les deux autres copies.
à capuche d’influence grecque. L’iconographie ves- Quand on les confronte attentivement, les
timentaire ne reflète que très peu le changement trois cycles iconographiques révèlent, au-delà
d’époque. Le drapé, imité de la période byzantine, des similitudes, des disparités qui ne sont pas
et le dégradé de couleurs ont disparu au profit toujours explicables (voir Tableau 5.3). La dispa-
de motifs en bandes de teintes contrastées. Ces rition de nombreux feuillets dans A3470 explique
bandes s’agrémentent de petits motifs et d’entre- qu’un nombre important d’images d’A3465 ne se
lacs (P3 et A3470). Les turbans sont remplacés par retrouve plus dans A3470. Près de vingt-cinq épi-
des coiffures composées elles aussi de bandes ou sodes illustrés manquent, situés dans les quatre
de petits ronds juxtaposés (P2), chaque peintre introductions de l’ouvrage ainsi que dans les cha-
ayant une légère spécificité. La seule part d’« otto- pitres « Le Lion et le taureau », « La Colombe au
manéité » réside dans le dessin des jeunes gardes. collier » et dans l’épisode « L’Ermite trompé ».
tableau 5.3 Comparaison des cycles iconographiques des manuscrits Arabe 3465, Arabe 3470 (Paris, Bibliothèque
nationale de France) et E.D. Clarke Or.09 (Oxford, Bodleian Library)
Légendes
tableau 5.3 Comparaison des cycles iconographiques des manuscrits Arabe 3465, Arabe 3470 (cont.)
Légende des chapitres et des fables Arabe 3465 Arabe 3470 Or.09
Frontispice 1v 1v
Alexandre et les astrologues 3
Bataille entre Alexandre et Pōros 3v 3
3v
L’alouette et l’éléphant 6 2v
Dabshalīm et Bidpaï 4v
[f.9 espace vide 6v
et légende]
10 5v
11v 7 8
12v 8v
14v 10v
15v 8 11v
17v 10 13
tableau 5.3 Comparaison des cycles iconographiques des manuscrits Arabe 3465, Arabe 3470 (cont.)
Légende des chapitres et des fables Arabe 3465 Arabe 3470 Or.09
4. Vie de Burzūya
5. Le lion et le taureau
tableau 5.3 Comparaison des cycles iconographiques des manuscrits Arabe 3465, Arabe 3470 (cont.)
Légende des chapitres et des fables Arabe 3465 Arabe 3470 Or.09
6. Le procès de Dimna
7. La colombe au collier
tableau 5.3 Comparaison des cycles iconographiques des manuscrits Arabe 3465, Arabe 3470 (cont.)
Légende des chapitres et des fables Arabe 3465 Arabe 3470 Or.09
9. Le singe et la tortue
tableau 5.3 Comparaison des cycles iconographiques des manuscrits Arabe 3465, Arabe 3470 (cont.)
Légende des chapitres et des fables Arabe 3465 Arabe 3470 Or.09
Néanmoins, ces lacunes n’expliquent pas toutes les A3465, mais sur un exemplaire très proche anté-
différences entre les trois cycles iconographiques. rieur ou postérieur à celui-ci ?
Ainsi, certains épisodes illustrés dans A3465 ne Deux exemples étayent cette hypothèse. Dans
sont pas présents dans l’une ou l’autre copie tar- le premier, « Le lièvre et l’éléphant », l’illustration
dive77. Il existe également des cas de figure plus d’A3465 a disparu et a été remplacée à l’époque
complexes, où l’un ou l’autre manuscrit tardif médiévale par un feuillet non illustré. Or, il y a
présente des illustrations absentes du plus ancien pourtant une peinture dans les deux copies tar-
manuscrit. Ainsi trois fables78 n’apparaissent que dives. Celle-ci n’a pu s’inspirer de l’image médié-
dans A3470, une quatrième79 n’est présente que vale disparue bien avant, la réfection se trouvant
dans les deux copies tardives. Est-ce à dire que sur un papier oriental antérieur aux copies plus
ces deux manuscrits ont été copiés, non pas sur récentes faites sur papier filigrané. Leurs com-
positions très similaires, l’éléphant à gauche, le
lièvre à droite, l’eau au centre dans laquelle la
lune se reflète, laissent à penser qu’elle s’inspire
77 Par exemple, Or.9 se différencie des deux autres
volumes dans « Le procès de Dimna » en montrant d’un modèle commun. La même silhouette de
deux fois « Le lion et sa mère », tandis que dans « Le l’éléphant se retrouve dans l’un et l’autre manus-
chat et le rat », il en omet la troisième peinture, pré- crit ainsi que dans d’autres pages. A-t-on utilisé
sente ailleurs. un moyen de reproduction pour le dessiner là ?
78 « L’homme et son trésor » (f.16) ; le stratagème de
la matrone dans « Le moine, le voleur, le renard et
Chaque artiste a-t-il indépendamment « inventé »
l’épouse du cordonnier » (f.26), « Le corbeau et la per- sa peinture ?
drix » (f.117). Dans le second exemple, le statut réservé à
79 « Le lièvre et l’éléphant ». l’une des histoires de la fable « Le moine, le voleur,
Histoires de transmissions 163
figure 5.19 Le moine, le voleur, le renard et la femme du cordonnier, Paris, BnF, A3470, f. 26
164 Vernay-Nouri
le renard et la femme du cordonnier » est plus gine, comme le prouvent l’inclusion d’une pein-
intrigant encore (Fig. 5.19). Rappelons l’anecdote, ture ancienne (« L’éléphant et l’alouette », f.6, Fig.
assez scabreuse : une prostituée se plaint qu’une 5.21) au milieu des pages refaites et les projections
de ses pensionnaires délaisse ses clients pour son de rouge et de bleu (f.4 et 4v) en face de pages
amant. La matrone imagine alors un stratagème remplacées, couleurs absentes dans les peintures
pour s’en débarrasser. À l’aide d’un roseau, elle refaites82. Elle comprenait très certainement plu-
insuffle du poison dans le derrière de l’homme. sieurs peintures de l’histoire d’Alexandre. En l’ab-
Mais ce dernier lâche alors un pet et c’est elle qui sence de toute peinture arabe médiévale connue
meurt, empoisonnée par le retour de la substance illustrant la légende d’Alexandre83, il est impos-
toxique ! Cette scène, qui figure dans plusieurs sible de savoir à quoi ressemblaient les illustra-
manuscrits mamelouks80, est absente d’A3465 et tions originales.
d’Or.9 mais bien présente dans A3470. Son icono-
graphie diffère pourtant de celle du xive siècle et 9.2 Deuxième étape : avant le xvie siècle
n’a pu s’en inspirer. Dans ces volumes, le couple est Des feuillets manquants (dont certains conte-
dépeint en pleins ébats sexuels tandis qu’A3470 naient probablement des illustrations) sont rem-
ne représente plus que l’homme allongé, nu. Le placés par des feuillets non illustrés sur papier
texte de cet épisode apparaît bien dans A3465 et oriental (f.5, 7-8, 68, 75 et 98). Le manuscrit est
son autre copie mais ces deux manuscrits sont alors en Égypte et la première foliotation copte
restés vierges de toute illustration. Leurs peintres est introduite lors de ces réfections. Le début et la
auraient-ils supprimé la scène à cause de sa cru- fin du manuscrit sont toujours présents puisqu’ils
dité ? Mais où alors l’artiste d’A3470 a-t-il puisé sa ne sont pas remplacés par des pages sur papier
source ? Est-ce à partir d’un autre modèle, ou bien oriental.
la peinture est-elle issue de sa propre imagination ?
9.3 Troisième étape : xviie siècle
1. Or.9 est copié sur les peintures originales
9 Une chronologie des réfections d’A3465 d’A3465 avant la disparition du début et de la
fin84 ; seules manquent les peintures rempla-
En l’absence d’éléments probants confirmant ou cées plus tôt dans A3465 par du texte seul.
infirmant le postulat d’une copie directe des deux 2. A3470 est copié sur les peintures originales
manuscrits tardifs sur A3465, on ne peut formuler d’A3465 et, à ce stade, les pages du début et
qu’une série d’hypothèses quant aux différentes celles de la fin de A3465 sont soit très dété-
étapes de la réalisation du manuscrit. riorées soit manquantes. Le copiste remplace
les pages finales par des peintures semblables
9.1 Première étape : xiiie siècle à celles qu’il a exécutées dans A3470. Il est
Le manuscrit est copié autour de 1222 en Égypte ou possible que le peintre ait utilisé un second
en Syrie. Il comporte une centaine de miniatures. manuscrit illustré, très proche d’A3465 pour
La préface d’ʿAlī b. al-Shāh81 est illustrée dès l’ori- compléter ces feuillets finaux. Cette hypo-
thèse est confirmée par la dissemblance
80 Paris, BnF, Arabe 3467, f.16 ; Oxford, Bodleian Library,
Pococke 400, f.46v ; Cambridge, Corpus Christi College,
MS 578, f.37 Riyadh, King Saud Library, 2536, f.30v. Tous 82 O’Kane, Early Persian Painting, 218.
ces manuscrits sont datés ou datables du xive siècle. 83 Seul un manuscrit du xviie siècle illustre la légende
81 La préface d’Alī b. al-Shāh ne figure pas dans toutes d’Alexandre, Sīrat al-Iskandar Dhū al-Qarnayn, Berlin,
les versions du texte, elle est absente des manuscrits Staatbibliothek, Ms Or Fol 2195.
mamelouks et de leurs dérivés ; voir De Blois, Burzōy’s 84 Les peintures finales d’Or.9 ont une composition diffé-
Voyage, 66-73. rente que celles refaites dans A3465 et A3470.
Histoires de transmissions 165
entre les peintures finales d’Or.9 et celles Cela a sans doute été le cas des modèles dont s’est
d’A3470. inspiré A3465. Malgré l’absence de témoignages
matériels en amont, notre manuscrit s’inscrit clai-
9.4 Quatrième étape : avant 1708 rement dans une suite de transmissions comme
Ibn Abū al-Minā, le peintre copiste égyptien des en témoignent les nombreux contours d’animaux
Mille et Une Nuits de Manchester ajoute les pages soulignés d’un épais trait noir88. Ces marques
manquantes et les peintures du début d’A3465. Il ne semblent pas avoir servi de calque au sein du
n’est probablement pas l’auteur des illustrations volume comme on pourrait le supposer. En effet,
d’Or.9, la similitude entre les illustrations de l’épi- dans le cas de figures doubles (les deux bouque-
sode d’Alexandre sur laquelle se base O’Kane étant tins (f.53v) ou les deux chacals (f.48) (Fig. 5.3), où
assez relative85. En outre, ses deux autres peintures seules sont soulignées les figures de gauche, celles
(f.22 et 25v) illustrant « La mission de Burzūya en de droite, qui devraient avoir été copiées à partir
Inde » n’apparaissent dans aucune des copies tar- d’elles, ne correspondent qu’imparfaitement. Ces
dives, preuve qu’Ibn Abū al-Minā est intervenu en contours ont donc servi à des fins de reproduction
dernier dans les réfections d’A3465 alors qu’A3470 dans d’autres manuscrits que l’on ne possède plus.
et Or.9 étaient déjà terminés. Elles présentent En effet, ceux-ci ne peuvent être A3470 et Or.9 car
par contre une grande ressemblance avec celles les différences de taille entre les personnages et
mettant en scène le couple roi/conseiller dans le les postures corporelles dissemblables excluent
manuscrit des Mille et Une Nuits. tout procédé de reproduction directe et plaident
pour un tracé à main levée. Seule la silhouette de
l’éléphant (f.6) qui se superpose parfaitement à
10 Une longue chaîne de transmission celles d’A3470 (ff.2v, 77v) suppose l’utilisation d’un
calque ou d’un pochoir. Il en est probablement de
Dans toutes ces hypothèses, ce schéma intro- même pour Or.9 (ff.3, 124v)89. On voit que la même
duit obligatoirement le recours à un ou plusieurs forme est réutilisée plusieurs fois pour illustrer des
manuscrits nécessaires à la chaîne de transmis- scènes différentes. (Figs. 5.20, 5.21 et 5.22)
sion. On sait que le texte de Kalīla wa Dimna s’ancre
dans une longue tradition picturale. La présence
d’une illustration clairement inspirée de l’icono- 11 Conclusion : un renouveau de la
graphie arabe86 dans les feuillets du xe ou xie siècle peinture arabe de manuscrits dans
d’une traduction grecque de Kalīla est la preuve les provinces ottomanes ?
flagrante que circulaient des manuscrits arabes
illustrés bien avant le xiiie siècle. Le faible nombre Au jeu des attributions communes, la compa-
de copies anciennes conservées87 laisse à penser raison stylistique peine à différencier les artistes
qu’une quantité importante d’entre eux a disparu. tardifs intervenus dans les trois volumes, les illus-
trations de P2 et P3 d’A3465, d’A3470 et d’Or.9 ainsi
85 La première peinture d’A3465, « Alexandre consultant que celles des Mille et Une Nuits de Manchester
les astrologues », n’est pas reprise dans Or.9. Dans la
et de Tübingen se caractérisant par de nombreux
seconde, « Bataille entre Poros et Alexandre », la com-
position générale est différente et, contrairement aux points communs et des différences très minimes.
personnages et aux chevaux qui sont très proches, les
éléphants ne sont pas exécutés sur le même modèle. de cent-trente ans (Beyrouth, Deir al-Shir, 6 Rajab
86 Voir note 1. 739H/18 Jan. 1339).
87 La plus ancienne copie possède un colophon daté de 88 C’est le cas par exemple du taureau (ff.45v, 62), du lion
jumāda II 618H/juillet 1221 (Istanbul, Süleymaniye (ff.49v, 52, 71v), du chien (f.40v) et des tortues (f.67).
Library, Ayasofia 4095). Le manuscrit sur lequel 89 La mise à l’échelle exacte n’a pas pu être faite en raison
Louis Cheikho a fait son édition lui est postérieur de la déformation des photos.
166 Vernay-Nouri
figure 5.21 Bataille entre Alexandre et Poros, Oxford, E.D. Clarke Or. 9, f. 3
168 Vernay-Nouri
La maladresse qui caractérise le dessin y est iden- lors de son séjour au Caire entre 1672 et 167396.
tique : on retrouve les mêmes corps déformés et de Enfin, Sūryāl, un troisième membre de la famille
semblables yeux globuleux ainsi que des procédés est connu comme peintre d’icônes et de manus-
analogues de remplissage des surfaces. Ces simili- crits au xviiie siècle.
tudes confortent l’hypothèse de peintres travaillant La place qu’occupe cette famille dans le paysage
au sein de mêmes scriptoria à défaut de peintres de la copie manuscrite au xviie siècle témoigne
identiques comme c’est le cas pour P3 et A347090. ainsi du rôle joué par le milieu chrétien dans la
Ces exemplaires, ainsi que la vingtaine de Kalīla production de manuscrits à peintures. Plusieurs
illustrés recensés à la période post-mamelouke91, travaux récents soulignent le renouveau que
attestent d’un renouveau de l’activité d’illustration connaît la peinture copte97 dès la seconde moi-
manuscrite dans les provinces ottomanes, essen- tié du siècle. Jouissant d’une certaine autonomie
tiellement en Syrie et en Égypte. par rapport au pouvoir central ottoman, la ville du
Dans le milieu encore largement inconnu Caire, l’une des plus grandes métropoles multicul-
des copistes, émerge dans l’Égypte des xviie et turelles du temps, abrite alors une nouvelle classe
xviiie siècles le patronyme copte d’Abū al-Minā92. moyenne urbaine et éduquée, composée d’arti-
Le premier de la famille, Nasīm b. Yuḥannā b. sans, de marchands et d’érudits. Il est tout à fait
Abū al-Minā est l’auteur des peintures refaites au concevable qu’un petit nombre de copistes et de
début d’A3465, de celles des Mille et Une Nuits de peintres en milieu chrétien ait travaillé à la réalisa-
Manchester et peut-être de celles de Tübingen. Le tion de manuscrits religieux mais aussi profanes98,
second, Abū al-Minā b. Nasīm al-Naqqāsh a laissé répondant à la demande d’une clientèle cultivée
son nom dans le colophon d’un Kalīla, achevé chrétienne mais aussi musulmane. On illustre
le 14 septembre 167193. Ce manuscrit, dont huit alors des textes anciens tels Kalīla wa Dimna ou
reproductions figurent dans la notice de vente l’on en ornemente de nouveaux à caractère dis-
chez Christie’s, se rapproche fortement de l’ico- trayant comme les Mille et Une Nuits ou la Légende
nographie d’A3465. C’est vraisemblement à ce d’Alexandre. Dans ce cas, les peintres ont recours
même peintre qu’il faut attribuer un autre Kalīla94 au même répertoire d’images qu’ils adaptent indif-
conservé à Munich, au style très similaire95, et féremment au texte copié ou qu’ils puisent éven-
dans lequel on peut reconnaître « le copiste copte tuellement dans d’autres traditions culturelles.
Abulmène » avec qui le père Vansleb est en contact Dans ce nouveau contexte historique, le manus-
crit A3465 a-t-il joué un rôle capital ? On peut ima-
giner que notre manuscrit, sorti des circuits de
90 Hypothèse confirmée par les analyses physico- production durant toute la période mamelouke,
chimiques. ait resurgi au cours du xviie siècle. Il supplante
91 Il faut y ajouter les six manuscrits dont les peintures
n’ont pas été exécutées, chiffre sans doute inférieur à
alors la tradition textuelle et iconographique qui
la réalité, vu la difficulté rencontrée pour repérer ces circulait alors, plus courte et représentée par les
exemplaires dans les catalogues.
92 Cette famille de copistes-peintres fera l’objet d’un pro-
chain article. 96 Omont, Missions archéologiques françaises en Orient
93 Vendu chez Christie’s en 2003, ce manuscrit daté aux XVIIe et XVIIIe siècles.
du 10 jumādā I 1082H/14 septembre 1671 comprend 97 Magdi Guirguis, An Armenian Artist in Ottoman Egypt:
quatre-vingt-dix-huit illustrations dont les titres Yuhanna al-Armani and his Coptic Icons (Le Caire :
figurent dans la notice. American University in Cairo Press, 2008) ; Nelly
94 Munich, Staatsbibliothek, Arab 615, manuscrit consul- Hanna, In Praise of Books a Cultural History of Cairo’s
table en ligne https://daten.digitale-sammlungen.de/ Middle Class, Sixteenth to the Eighteenth Century
~db/0007/bsb00077202/images/. (Syracuse, N.Y. : Syracuse University Press, 2003).
95 On peut attribuer à ce peintre au moins cinq autres 98 C’est le cas du peintre de Munich qui illustre la Bible et
manuscrits. des sujets profanes.
170 Vernay-Nouri
Kalīla illustrés mamelouks et leurs copies99. Les Édité par Robert Hillenbrand et Sylvia Auld, 179-194.
pages manquantes complétées par de nouvelles Londres : Altajir Trust, 2009.
peintures, il servira de modèle, non seulement Dehérain, Henri. « Asselin de Cherville, drogman
à A3470 et Or.9, mais aussi à toute une nouvelle du consulat de France en Égypte et orientaliste ».
génération de manuscrits peints dont les manus- Journal des savants 14, no 4 (1916) : 176-187.
crits de Munich et de Christie’s sont le reflet. Déroche, François, éd. Manuel de codicologie des manus-
crits en écriture arabe. Paris : Bibliothèque nationale
de France, 2000.
Bibliographie Grube, Ernst J., éd. A Mirror for Princes from India:
Illustrated Versions of the Kalilah Wa Dimnah,
Akel, Ibrahim. « Ahmad al-Rabbât al-Halabî : sa biblio- Anvar-i Suhayli, Iyar-i Danish, and Humayun Nameh.
thèque et son rôle dans la réception, diffusion et Bombay : Marg Publications, 1991.
enrichissement des Mille et une Nuits. » Thèse de Guirguis, Magdi. An Armenian Artist in Ottoman Egypt:
doctorat, Université Paris-Sorbonne, 2016. Yuhanna al-Armani and his Coptic Icons. Le Caire :
Avery, Myrtilla. « Miniatures of the Fables of Bidpai and American University in Cairo Press, 2008.
the Life of Aesop in the Pierpont Morgan Library ». Hanna, Nelly. In Praise of Books a Cultural History of
Art Bulletin 23, n°2 (1941) : 103-116. Cairo’s Middle Class, Sixteenth to the Eighteenth Cen-
Barrucand, Marianne. « Le Kalila wa-Dimna de la tury. Syracuse, N.Y. : Syracuse University Press, 2003.
Bibliothèque royale de Rabat : un manuscrit illustré Humbert, Geneviève. « Le manuscrit arabe et ses
il-khanide ». Revue des études islamiques, numéro papiers ». Revue des mondes musulmans et de la
spécial en l’honneur de Dominique Sourdel, n°54 Méditerranée, no 99-100 (2002) : 55-77.
(1986) : 17-51. Hunt, Lucy-Anne. « Christian-Muslim Relations in
Barrucand, Marianne. « Un manuscrit arabe illustré Painting in Egypt of the xiith to Mid xiiith Cen-
de Kalila wa Dimna du xiiie siècle et sa copie otto- turies: Sources of Wall-Painting at Deir Es-Suriani
mane ». Archéologie islamique, no 2 (1991) : 81-95. and the Illustration of the New Testament MS Paris,
Benouniche, Leïla. Le « Kalila et Dimna » de Genève : his- Copte-Arabe 1/Cairo, Bibl. 94 ». Cahiers archéolo-
toire d’un recueil de fables illustré. Genève : Slatkine, giques, no 33 (1985) : 111-55.
1995. Levasseur, Claire. « Kalila wa Dimna : la relation entre
Blochet, Edgar. Les peintures des manuscrits orientaux le manuscrit illustré MS 3465 et sa copie ottomane
de la Bibliothèque nationale. Paris : Société française MS 3470 ». Mémoire de maîtrise, Université Paris IV,
de reproductions de manuscrits à peintures, 1920. 1991.
Blois, François de. Burzōy’s Voyage to India and the Mac Guckin de Slane, William. Catalogue des manuscrits
Origin of the Book of Kalīlah Wa Dimnah. Londres : arabes / Bibliothèque nationale. Paris : Imprimerie
Royal Asiatic Society, 1990. nationale, 1883.
Buchthal, Hugo. « “Hellenistic” Miniatures in Early Madan, Falconer et Herbert H.E. Craster. A Summary
Islamic Manuscripts ». Ars Islamica 7, no 2 (1940) : Catalogue of Western Manuscripts in the Bodleian
125-150. Library at Oxford which have not hitherto been cata-
Contadini, Anna. « Ayyubid Illustrated Manuscripts and logued in the quarto series, with references to the
their North Jaziran and ʿAbbasid Neighbours ». In Oriental and other manuscripts. 8 vol. Oxford : At the
Ayyubid Jerusalem. The Holy City in Context 1187-1250. Clarendon Press, 1895.
Mézin, Anne. Les consuls de France au siècle des
99 Voir l’article dans ce même volume Éloïse Brac de la Lumières : 1715-1792. Paris : Ministère des Affaires
Perrière, « Les tribulations d’un manuscrit à peintures étrangères, 1997.
ou comment fabriquer une illustration à partir de plu-
sieurs modèles ».
Histoires de transmissions 171
Mingana, Alphonse. Catalogue of the Arabic Manus- Otter, William. The Life and Remains of the Rev. Edward
cripts in the John Rylands Library, Manchester. Daniel Clarke, LL.D. Professor of Mineralogy in the
Manchester : Manchester University Press, 1934. University of Cambridge. Londres : Printed for George
Nicoll, Alexander, éd. Catalogus sive notitia manus- Cowie and Co. in the Poultry, 1824.
criptorum qui a cel. E.D. Clarke comparati in biblio- Pelliot, Paul. Les arts de l’Iran, l’ancienne Perse et Bagdad.
theca bodleiana adservantur. 2 vol. Oxonii : e typ. Paris : Bibliothèque nationale de France, 1938.
Clarendoniano, 1815. Silvestre de Sacy, Antoine-Isaac. Calila et Dimna, ou
O’Kane, Bernard. « Der Kreis schliesst sich : Illustra- Fables de Bidpaī, en arabe, précédées d’un mémoire
tionen aus Kalīla wa-Dimna von Indien zum Mitt- sur l’origine de ce livre, et sur les diverses traduc-
leren Osten und wieder zurück ». In Von listigen tions qui en ont été faites dans l’Orient, et suivies de
Schakalen und törichten Kamelen : die Fabel in Orient la « Moalloka » de Lébid, en arabe et en français, par
und Okzident. Édité par Mamoun Fansa et Eckhard M. Silvestre de Sacy. Paris : Imprimerie royale, 1816.
Grunewald, 127-148. Wiesbaden : Reichert, 2008. Troupeau, Gérard. « Les colophons des manuscrits
O’Kane, Bernard. Early Persian Painting: Kalila and chrétiens ». In Scribes et manuscrits du Moyen-Orient.
Dimna Manuscripts of the Late Fourteenth Century. Édité par François Déroche et Francis Richard, 223-
Londres : I.B. Tauris, 2003. 231. Paris : Bibliothèque nationale de France, 1997.
O’Kane, Bernard. « The Uses of Captions in Medieval Villefroy, Guillaume de. Catalogus codicum manuscrip-
Literary Arabic Manuscripts ». In Arab Painting: Text torum Bibliothecae regiae, 7 vol. Parisiis : e typografia
and Image in Illustrated Arabic Manuscripts, édité regia, 1739.
par Anna Contadini, 135-144. Leiden : Brill, 2007.
Omont, Henri. Missions archéologiques françaises
en Orient aux XVIIe et XVIIIe siècles, 2 vol. Paris :
Imprimerie nationale, 1902.
6
Confluence of Cultures: The Rabat Kalīla and Dimna and Its Copy
Bernard O’Kane
Baghdad owed its foundation to several factors population ensured that, as it does to this day, Iran
connected with the Iranian world, the large num- continued to play an important part in Baghdad’s
bers of Iranian supporters of the ʿAbbasid cause political, social, and cultural affairs.
and the proximity of the previous Sasanian capi- In 1986, Marianne Barrucand was the first to
tal of al-Madaʾin among them. Al-Ṭabarī even publish the Rabat Kalila and Dimna, at a time
reports that al-Manṣūr’s Persian minister, Khālid when its folios had not yet been rebound in the cor-
ibn Barmak, laid out the city for him and gave him rect order (as detailed below).4 The paintings are
advice on its planning.1 The city remained a space unique in Mongol and Arab figures with an Arabic
where Arab and Iranian cultural influences inter- text and an Arab painting style. She analyzed its
mingled; with its various Iranian, Turkish, and salient features and, on stylistic grounds, attributed
Mongol rulers (Buyid, Saljuq, Ilkhanid, Jalayirid, it to Baghdad c. 1265–80, an astute evaluation that
Turcoman,2 and Safavid), who were as much at has stood the test of time. In 1990, she discussed
home in the Persian as in the Arab sphere, this is it further in connection with a copy, dated 1761, in
not surprising. Only after the final Ottoman recon- the Bibliothèque nationale de France in Paris, not-
quest of the city from the Safavids in 16383 did ing correctly that the closeness of this copy might
the borders harden a little, but the majority Shiʿi permit the restitution of the original iconographic
program of the Rabat Kalila and Dimna.5 In 1999,
1 Muḥammad ibn Jarīr al-Ṭabarī, The History of al-Ṭabarī:
she further analyzed the Rabat manuscript and
An Annotated Translation, Volume 29: al-Mansur and al- its copy, and proposed preliminary identifications
Mahdi, A.D. 763–786; A.H. 146–169, trans. Hugh Kennedy for their paintings.6 Thanks to efforts of the edi-
(Albany NY: State University of New York, 1990), 4; see tors of this volume, it has been possible for me to
also Bernard O’Kane, “Islamic Art and Architecture in
work on the codicology of these two manuscripts
Pre-Mongol Baghdad,” in Baghdad – Space of Knowledge,
ed. Isabel Toral-Niehoff and Jens Scheiner, in press. and provide the desired restitution of the original
2 Masterpieces of Persian painting such as the Khamsa of iconographic program of the Rabat manuscript.
Khvaju Kirmani (British Library Add. 18113) or the Topkapi The tabulation presented below provides addi-
Saray Khamsa of Nizami for Pir Budaq (H.761), both com- tions and corrections to the previous publications,
pleted in Baghdad, spring to mind. For the British Library
manuscript, see Ursula Sims-Williams, “An Illustrated
14th Century Khamsah by Khvaju Kirmani,” Asian and 4 Marianne Barrucand, “Le Kalila wa Dimna de la Biblio-
African studies blog (blog), August 1, 2013, http://british thèque royale de Rabat : un manuscrit illustré il-khanide,”
library.typepad.co.uk/asian-and-african/2013/07/an-illus Revue des Études Islamiques, numéro spécial en l’honneur
trated-14th-century-khamsah-by-khvaju-kirmani.html, de Dominique Sourdel 54 (1986): 17–51.
accessed July 8, 2016. For Pir Budaq’s patronage in Baghdad, 5 Marianne Barrucand, “Un manuscrit arabe illustré de
see David J. Roxburgh, “‘Many a Wish Has Turned to Dust’: Kalila Wa Dimna du XIIIe siècle et sa copie ottomane,”
Pir Budaq and the Formation of Turkmen Arts of the Archéologie Islamique, no. 2 (1991): 81–95.
Book,” in Envisioning Islamic Art and Architecture: Essays 6 Marianne Barrucand, “Kopie: Nachempfindung oder
in Honor of Renata Holod, ed. David J. Roxburgh (Leiden: Umgestaltung, am Beispiel Mittelalter Arabischer Bilder-
Brill, 2014), 175–222, esp. 212–213. handschriften und ihrer Osmanischen Kopien,” in Rezep-
3 Ebubekir Ceylan, “Baghdad, from 1500 to Iraqi Indepen- tion in der Islamischen Kunst, ed. Barbara Finster, Christa
dence,” in Encyclopaedia of Islam, third edition, ed. Kate Fragner, and Herta Hafenricher (Beyrouth; Stuttgart: Bei-
Fleet et al. (Leiden: Brill, 2014), accessed June 22, 2016. ruter Texte und Studien; F. Steiner, 1999), 19–41.
and displays it in a way that – I hope – makes the elongated headdress characteristic of Mongol
relationships between the original and the copy women of the ruling strata.9 The first conclusion
easier to understand at a glance. In addition, I fur- to be drawn from this is that the unknown patron
ther analyze the links between text and paintings was almost certainly from the Ilkhanid elite.
in the manuscripts. How consistent is the Rabat painter in differ-
Sadly, the Paris copy is not quite complete. It entiating Mongols and Arabs in other subjects?
also has several peculiarities, among them the fre- There are certainly examples of both together
quent omission of paintings in the spaces left for in many scenes that do not involve rulers. There
them in the text, and the insertion of folios, with- seems a consistent attempt to differentiate them
out text, containing one or more paintings adja- on physiognomical grounds, with most Arabs,
cent to the blank spaces. Its copies of the Rabat especially men, given an elongated face, and most
painting are crude, sometimes extremely so, but Mongols, male and female, having a rounder face
for the most part, even when simplified, they are with more narrowed eyes, and sometimes a bul-
sufficient to permit a general sense of the original bous nose (Fig. 6.4). Most Arab men have a beard
composition.7 The correspondence between the (Figs. 6.1–6.2, 6.3b), while Mongols are either
two can be seen more easily in the following com- beardless (Figs. 6.3a, 6.8) or have a mustache and
parative table. goatee (Fig. 6.3b). The non-royal Mongol women
Although there is not quite a one-to-one cor- usually have a distinctive costume: a narrow scarf
respondence between the Rabat manuscript and knotted at the head that flies out in two ribbons
its Paris copy, it can be seen from the table that behind; an upper cape (usually blue) that covers
Rabat is probably missing around sixty-five of its the shoulders; a wide-sleeved dress below, usually
original paintings.8 Its original total of around 187 red (sometimes the red and blue are reversed);
paintings would have made it the most heavily and, where visible, pants, frequently with a check-
illustrated Kalila and Dimna manuscript known. ered or striped pattern, that narrow at the ankles
The most unique feature of the manuscript is, of (Figs. 6.3a, 6.4).10
course, its pictorial derivation from the Arab man- However, one would expect consistency within
uscript tradition, with its incorporation of figures the substories in the decision whether to depict
in Mongol costume, the key to locating its produc- the main actors as Mongols or Arabs. The first
tion in Ilkhanid-ruled Baghdad. It also incorpo- indication of a lack of this is in “The Imprudent
rates Arab figures, but it is worth dwelling on the Lovers,” in which a wife has a subterranean pas-
artist’s choices of Arab or Mongol garb. Kings fea- sage dug to aid her lover’s visits to her. On hear-
ture prominently in the cycle, and in every paint- ing her husband entering the house, she urges her
ing where a king is depicted, he is a Mongol. The lover to escape by the passage, the entry to which
King’s son in “The Snake and the Frog King: The
Snake Bites the Ascetic’s Son” (Rabat f. 85b) is 9 Thomas T. Allsen, Commodity and Exchange in the
also a Mongol. Similarly, when a queen is shown, Mongol Empire: A Cultural History of Islamic Textiles
as in the chapter “The King and the Brahmans,” (Cambridge: Cambridge University Press, 1997), 17–18.
Many illustrations in the Diez albums in Berlin show
she is wearing the red silk-wrapped boghta, the
Ilkhanid women of the court in such headgear, e.g.
Linda Komaroff and Stefano Carboni, eds., The Legacy
7 Eight Rabat paintings and their Paris copies are repro- of Genghis Khan: Courtly Art and Culture in Western
duced in Barrucand, “Un manuscrit arabe illustré de Kalila Asia, 1256–1353 (New York: Metropolitan Museum of
Wa Dimna,” figs. 1–16, with accompanying analysis; one Art, 2002), figs. 84 and 80. In the one case where the
further painting is illustrated in Barrucand, “Kopie,” 5–6. queen is not wearing the boghta, the exception, Rabat
8 As noted above, ff. 4–5 in the copy may have replaced f. 99b, the text explicitly refers to her picking a crown as
three folios with two paintings, hence the total in the copy headgear.
may well have been as high as 187. 10 E.g. ff. 9b, 10b, 15b, and 69a.
174 O ’ Kane
a Barrucand, “Kopie,” 37, interprets this as the hitherto unknown subject of Kisra’s palace; see the discussion below.
b I include in the total the blank spaces that were both not painted in place and whose subjects were not covered in an added
folio with no text. Hence, from this point on, my tabulation differs from that of Barrucand, “Kopie,” 37–41.
Confluence of Cultures 175
2) 2b “Kisra Orders the Treasury to be Thrown 10) 18a blank space for “Kisra Orders the Treasury to be
Open to Burzuya” (Fig. 6.1b) Thrown Open to Burzuya”
3) 3a “Kisra Orders Buzurjmihr to Write Burzuya’s 11) 19a blank space for “Kisra Orders Buzurjmihr to
Biography” Write Burzuya’s Biography”
113 (out of order, catchword matches 19a) folio with
no text: 113a “Kisra Orders Buzurjmihr to Write
Burzuya’s Biography”; 113b “Buzurjmihr Reads
His Biography of Burzuya to Kisra”
4) 3b “Buzurjmihr Reads His Biography of Burzuya 12) 19b blank space for “Buzurjmihr Reads His
to Kisra” Biography of Burzuya to Kisra”
15) 15b “The Condition in Which Man is Born” 23) 31 folio with no text: 31b blank; 31b “The Condition
(Fig. 6.4) in Which Man is Born”
32a blank space for “The Condition in Which Man
is Born”
16) 17b “The Perils of Life” 24) 34 folio with no text: 34a “The Perils of Life”;
34b blank
35a blank space for “The Perils of Life”
c A55/3 refers to ʿAbd Allāh Ibn al-Muqaffaʿ, Kalila Wa Dimna, ed. ʿAbd al-Wahhâb ʿAzzâm, 2nd edition (Cairo, 1981), 55, line 3
and line number; subsequent citations are on the same pattern. I would like to thank my research assistant, Imane Abaza, for
her work in matching the foliation of the Rabat manuscript with the ʿAzzam edition.
Confluence of Cultures 177
Gap
28b: ends A94/14
29a: begins A97/7
38) 50a: “The Ascetic and the Robe: The Greedy Fox”
39) 50b: “The Ascetic and the Robe: The Madam’s
Stratagem Backfires” (Fig. 6.5a)
40) 51a “The Ascetic and the Robe: The Carpenter Cuts
Off the Nose of the Barber’s Wife”
29) 29a “The Ascetic and the Robe: The Barber and 41) 52a “The Ascetic and the Robe: The Barber and His
His Wife before the Judge” Wife before the Judge”
30) 30a “The Crow and the Snake: The Jackal Talks to 42) 53b “The Crow and the Snake: The Jackal Talks to
the Crow” the Crow”
31) 31a “The Crow and the Snake: The Tortoise and the 43) 54a “The Crow and the Snake: The Tortoise and the
Crab” (Fig. 6.6a) Crab” (Fig. 6.6b)
32) 32a “The Crow and the Snake: The Crab Kills the 44) 55a “The Crow and the Snake: The Crab Kills the
Tortoise” Tortoise”
33) 32b “The Crow and the Snake: The Men Kill the 45) 55b: blank space for “The Crow and the Snake: The
Snake” Men Kill the Snake”
56: folio with no text: 56a “The Crow and the Snake:
The Men Kill the Snake”; 56b blank
34) 33b “The Hare and the Lion: The Hare and the 46) 57b: blank space for “The Hare and the Lion:
Lion at the Well” (Fig. 6.7) The Hare and the Lion at the Well”
58 folio with no text: 58a “The Hare and the Lion:
The Hare and the Lion at the Well”; 58b “Dimna
Tries to Turn the Lion against Shanzaba”
35) 34a “Dimna Tries to Turn the Lion against 47) 59a blank space for “Dimna Tries to Turn the Lion
Shanzaba” against Shanzaba”
36) 35a “The Fishermen and the Three Fish: 48) 60a “The Fishermen and the Three Fish: The
The Fishermen Lower Their Nets” Fishermen Lower Their Nets”
49) 61b space in text for “Dimna Warns the Lion about
Gap Shanzaba,” but the crude drawing, in ink outline, is
35b ends A109/11 of a jackal with its leg tied to an object
35a begins A111/2 50) 62a space in text for “The Man Kills the Louse
because of the Flea’s Bite,” but illustrated by what
seems to be a female rather than a male, and, inex-
plicably, with a jackal
51) 63b space in text for “Dimna Conversing with
Shanzaba,” and caption for “The Bull and Dimna,”
but only Dimna drawn (in crude ink outline)
52) 64a “Dimna Converses with Shanzaba”
53) 64b “The Duck Mistakes the Star’s Reflection
for Fish”
178 O ’ Kane
37) 37a “The Horse Whose Good Qualities Were Its 54) 66a “The Horse Whose Good Qualities Were Its
Ruin” (Fig. 6.8a) Ruin”
38) 38a “The Bee Imprisoned in the Waterlily” 55) 67a “The Bee Imprisoned in the Waterlily”
(Fig. 6.9a) (Fig. 6.9b)
39) 39a “The Lion, the Crow, the Jackal, 56) 67b “The Lion, the Crow, the Jackal, the Wolf, and
the Wolf, and the Camel: The Lion Invites the Camel: The Lion Invites the Camel to Remain
the Camel to Remain with Him” with Him” (only a lion, not a copy of the original, is
depicted)
40) 39b “The Lion, the Crow, the Jackal, 57) 68b “The Lion, the Crow, the Jackal,
the Wolf, and the Camel: The Crow Tries the Wolf, and the Camel: The Crow Tries
to Persuade the Lion to Kill and Eat the to Persuade the Lion to Kill and Eat the
Camel” Camel” (only a lion, not a copy of the original, is
depicted)
41) 41a “The Lion, the Crow, the Jackal, the Wolf, and 58) 69b: “The Lion, the Crow, the Jackal, the Wolf, and
the Camel: The Camel Offers Himself as Food for the Camel: The Camel Offers Himself as Food for
His Friends” His Friends”
42) 41b “The Lion, the Crow, the Jackal, the Wolf, 59) 70b blank space for “The Lion, the Crow, the
and the Camel: The Attack on the Camel” Jackal, the Wolf, and the Camel: The Attack on the
Camel”d
71 folio with no text: 71a “The Lion, the Crow, the
Jackal, the Wolf, and the Camel: The Attack on the
Camel” (oriented sideways); 71b blank
43) 42b “The Waterfowl and the Spirit of the Sea: 60) 72a “The Waterfowl and the Spirit of the Sea: The
The Two Waterfowl” Two Waterfowl” (composition reversed)
61) 72b “The Talkative Tortoise: The Tortoise and the
Gap Two Ducks”
42b ends A118/3 62) 73a “The Talkative Tortoise: The Flight of the
43a begins A120/19 Tortoise”
63) 74 folio with no text: 74a blank, 74b “The Waterfowl
and the Spirit of the Sea: The Assembly of the
Birds” (oriented sideways) (Fig. 6.10)
75a blank space for “The Waterfowl and the Spirit
of the Sea: The Assembly of the Birds”
44) 44a “Kalila and Dimna Converse” 64) 76a “Kalila and Dimna Converse” (outlined in black
ink only)
45) 46a “The Monkeys, the Firefly, and the Bird” 65) 78a “The Monkeys, the Firefly, and the Bird” (sim-
plified composition)
d The text of the attack on the camel finishes five lines from the bottom of f. 69b, but the blank space at the top of 70b is cap-
tioned mahall al-jamal al-maʾkūl (the place for the eaten camel).
Confluence of Cultures 179
46) 46b “The Simpleton and the Rogue: The Two Men 66) 78b blank space for “The Simpleton and the Rogue:
Find the Gold”e The Two Men Find the Gold”
79 folio with no text: 79a “The Simpleton and the
Rogue: The Two Men Find the Gold”; 79b blank
47) 47a “The Simpleton and the Rogue: The Simpleton 67) 80a “The Simpleton and the Rogue: The Simpleton
Finds the Treasure Missing” Finds the Treasure Missing”
48) 47b “The Simpleton and the Rogue: The Two Men 68) 80b “The Simpleton and the Rogue: The Two Men
before the Judge” (Fig. 6.11) before the Judge” (only two figures shown, both
outlined in black ink only)
49) 48a “The Crane, the Snake, the Crab, and the 69) 81a “The Crane, the Snake, the Crab, and the
Weasel: The Crab Advises the Crane” Weasel: The Crab Advises the Crane”
50) 48b “The Crane, the Snake, the Crab, and the 70) 81b “The Crane, the Snake, the Crab, and the
Weasel: The Weasel Kills the Crane and Her Young” Weasel: The Weasel Kills the Crane and Her Young”
51) 49a “The Simpleton and the Rogue: The Rogue’s 71) 82a “The Simpleton and the Rogue: The Rogue’s
Stratagem is Discovered” Stratagem is Discovered”
52) 49b “Kalila and Dimna Converse” 72) 82b “Kalila and Dimna Converse”
53) 50a “The Dishonest Merchant and His Colleague: 73) 83a “The Dishonest Merchant and His Colleague:
The Merchant Explains How Mice Ate His The Merchant Explains How Mice Ate His
Colleague’s Iron” Colleague’s Iron”
74) 84b “Dimna Placates the Lion after Shanzaba’s
Death”
e Earlier (Bernard O’Kane, Early Persian Painting: Kalila and Dimna Manuscripts of the Late Fourteenth Century (London:
I.B. Tauris; Cairo: AUC Press, 2003), 230, n. 13.), I had interpreted the painting as follows: “One of the partners seems to be
hitting the other (which occurs later in the story), although the second is not responding but rather is engaged in burying the
treasure, which does accord with the text at this point.” Now that a clearer reproduction is available, it seems that the Mongol
is pulling the beard of the Arab (although again, this does not follow the text), who is using a scale to divide the proceeds.
Fig. 6.3b also shows the husband pulling the beard of the lover.
180 O ’ Kane
Gap 79) 89b blank space for “The Painter and His Cloak:
52b ends A139/11 The Merchant’s Wife and the Slave Make Love”
53a begins 140/14 90 folio with no text: 90a “The Painter and His
Cloak: The Merchant’s Wife and the Slave Make
Love” (Fig. 6.5b); 90b blank
55) 53b “Dimna Pleads His Innocence before the Lion 80) 91a “Dimna Pleads His Innocence before the Lion
and His Mother” and His Mother” (only one crudely outlined jackal
shown)
56) 54b “The Lion’s Mother Warns the Lion Against 81) 92a “The Lion’s Mother Warns the Lion Against
Dimna” Dimna” (only one lion painted)
57) 55a “Kalila Visits Dimna in Prison” 82) 92b “Kalila Visits Dimna in Prison”
58) 56a “Dimna’s Trial” 83) 93b “Dimna’s Trial”
59) 57b “The Ignorant Doctor Administers Poison to 84) 95a “The Ignorant Doctor Administers Poison to
the King’s Daughter” the King’s Daughter”
60) 59a “Dimna Reveals the King of the Boars’ Faults” 85) 96b “Dimna Reveals the King of the Boars’ Faults”
61) 60a “Fayruz Visits Dimna in Prison” 86) 97b “Fayruz Visits Dimna in Prison”
62) 61a “The Lion and His Mother” 87) 98b “The Lion and His Mother” (only one lion
painted)
88 (out of order, but catchword matches 99a)
folio with no text: 88a “The Lion and His Mother”
(oriented sideways); 88b blank
63) 62a “The Judge Rebukes Dimna” 88) 99b blank space for “The Judge Rebukes Dimna”
113a (out of order) folio with no text: 113a “The
Judge Rebukes Dimna”; 113b
89) 114b (out of order) blank space for “The Parrots
and the Governor’s Wife: The Parrots Accuse the
Governor’s Wife”
64) 63a “The Parrots and the Governor’s Wife: 90) 110 folio with no text: 110a “The Parrots and the
The Parrots Accuse the Governor’s Wife” Governor’s Wife: The Parrots Accuse the Governor’s
Wife”; 110b blank
65) 63b “The Parrots and the Governor’s Wife: 91) 111b folio with no text: 111a blank; 111b “The Parrots
The Falcon Attacks the Falconer” and the Governor’s Wife: The Falcon Attacks the
Falconer” (as in the original, the falcon is repre-
sented as a parrot)
109a (out of order) blank space for “The Parrots
and the Governor’s Wife: The Falcon Attacks the
Falconer”
66) 64a “Dimna’s Death” 92) 109b “Dimna’s Death” (out of order)
Confluence of Cultures 181
f Although this is indicated by the text (it is the point where the leader of the pigeons asks the rat to free his companions
before him), the rat has not been illustrated. It seems as if the blank space to the right of the miniature was meant to contain
this illustration.
g This is not, as might first be thought, a stray from the story of “The Wolf, the Crow, the Lion, the Jackal, and the Camel”
(in which the lion is first injured by an elephant), but an illustration of a metaphor for mutual enmity in the conversation
between the rat and the crow.
182 O ’ Kane
108) 121 folio with no text: 120a blank; 120b “The Crow
and the Gazelle Divert the Hunter” (oriented
sideways)
80) 76b “The Crow and the Gazelle Divert the Hunter” 122a blank space for “The Crow and the Gazelle
Divert the Hunter”
Chapter: The Owls and the Crows
81) 77b “The Crows Consider Their Response to the 109) 123a “The Crows Consider Their Response to the
Owls” Owls”
82) 78b “The Advice of the Fifth Crow” 110) 124b “The Advice of the Fifth Crow” (composition
reversed)
83) 79b “The Origin of the Enmity Between the Crows 111) 125a “The Origin of the Enmity Between the Crows
and the Owls” and the Owls”
84) 80a “The Hares and the Elephants: The Council of 112) 125b “The Hares and the Elephants: The Council of
the Hares” the Hares”
85) 80b “The Hares and the Elephants: Fayruz Warns 113) 127a blank space for “The Hares and the Elephants:
the Elephants of the Moon’s Displeasure” Fayruz Warns the Elephants of the Moon’s
Displeasure”
86) 81a “The Hares and the Elephants: Fayruz Tricks 114–115) 126 folio with no text: 126a “The Hares and
the Elephant” the Elephants: Fayruz Warns the Elephants of
the Moon’s Displeasure” and “The Hares and the
Elephants: Fayruz Tricks the Elephant” (composi-
tion reversed, without the pool or reflected moon);
126b “The Hares and the Elephants: Fayruz Tricks
the Elephant” (composition reversed and without
the pool or reflected moon)
127a blank space for “The Hares and the Elephants:
Fayruz Warns the Elephants of the Moon’s
Displeasure” and for “The Hares and the Elephants:
Fayruz Tricks the Elephant”
87) 81b “The Crow Watches the Hare Take the 116) 127b “The Crow Watches the Hare Take the
Woodcock’s Lair/The Hare and the Partridge Woodcock’s Lair/The Hare and the Partridge
Argue”h Argue”
88) 82b “The Hypocritical Cat: The Cat Gives Advice 117) 128b “The Hypocritical Cat: The Cat Gives Advice
to the Hare and the Partridge” to the Hare and the Partridge”
h The artist has drawn a crow rather than a partridge (ṣafrid). In the text, the crow observes the scene where the hare and par-
tridge argue over the rights to the lair. As the next scene shows the cat with the hare and the crow, it could well be that the
artist has interpreted ṣafrid as crow.
Confluence of Cultures 183
89) 83a “The Ascetic and the Thief” 118) 129 folio with no text: 129a blank; 129b “The Ascetic
and the Thief”
130a blank space for “The Ascetic and the Thief”
90) 84a “The Crow Spy” 119) 130b “The Crow Spy”
120) 131b “The Ugly Merchant and His Wife”
Gap 121) 132a blank space for “The Ascetic, the Thief and the
84 b ends A200/25 Demon: The Thief and the Demon Quarrel”
85a begins A211/2 122) 133 folio with no text: 133a blank, 133b “The
Deluded Carpenter: The Carpenter, His Unfaithful
Wife, and Her Lover”
134a blank space for “The Deluded Carpenter: The
Carpenter, His Unfaithful Wife, and Her Lover”
123) 135 folio with no text: 135a blank; 135b “The Rat Girl:
A Girl Holding a Rat” (misinterpretation of the
following)
136b blank space for “The Rat Girl: The Sun
Explains that a Cloud is More Powerful than
Himself”
91) 85a “The Snake and the Frog King: The Snake 124) 138b “The Snake and the Frog King: The Snake
Tells the Frog of His Plight” Tells the Frog of His Plight” (but the artist has
drawn a tortoise instead of a frog)
92) 85b “The Snake and the Frog King: The Snake Bites 125) 139a “The Snake and the Frog King: The Snake Bites
the Ascetic’s Son” the Ascetic’s Son” (composition reversed)
93) 86a “The Snake and the Frog King: The King of the 126) 139a “The Snake and the Frog King: The King of the
Frogs Rides the Snake” Frogs Rides the Snake”
94) 86b “The King of the Crows Thanks the Crow Spy 127) 140 folio with no text: 140a “The King of the Crows
for Having Saved Them” Thanks the Crow Spy for Having Saved Them”; 140b
blank
Chapter: The Cat and the Rat
95) 87a “The Cat in the Trap, the Weasel, and the Owl” 128) 142b “The Cat in the Trap, the Weasel, and the Owl”
96) 88a “The Rat Makes a Pact with the Cat” 129) 143a “The Rat Makes a Pact with the Cat”
97) 89b “The Cat Tries to Befriend the Rat” 130) 144b “The Cat Tries to Befriend the Rat”
131) 145b “The Rat Refuses to Make Friends with the
Gap Cat”
89b ends 278/1 chapter ends 278/17 132) 146a “The Monkey Makes Friends with the
Tortoise”
Chapter: The Monkey and the Tortoise 133) 146b “The Tortoise’s Wife Talks with Her Friend”
(begins A219) 134) 147a “The Tortoise Learns of the Remedy for His
90a begins 224/1 Wife’s Illness from Her Friend”
135) 147b blank space for “The Monkey Converses with
the Tortoise”
136) 148b “The Monkey Rides the Tortoise”
184 O ’ Kane
98) 90b “Kardana’s Escape”i 137) 149b blank space for “Kardana’s Escape”
99) 91b “The Ass Escapes from the Lion” 138) 150 folio with no text: 151a blank; 150 b “The Lion
Kills the Ass” (oriented sideways)
151a blank space for “The Ass Escapes from the
Lion”
100) 92a “The Lion Kills the Ass” 139) 151b copy of “The Ass Escapes from the Lion” (the
text shows it should be “The Lion Kills the Ass”)
Chapter: The Ascetic and the Weasel
101) 93a “The Devout Man and the Pot: The Devout 140) 153a “The Devout Man and the Pot: The Devout
Man Dreams of Riches”j Man Dreams of Riches”
102) 94a “The Devout Man and the Pot: The Devout 141) 153b “The Devout Man and the Pot: The Devout
Man Knocks a Pot of Oil on His Head” Man Knocks a Pot of Oil on His Head”
103) 94b “The Ascetic, His Wife, and Their Child” 142) 154a “The Ascetic, His Wife, and Their Child”
104) 95a “The Ascetic Kills the Faithful Weasel” 143) 154b blank space for “The Ascetic Kills the Faithful
Weasel”
95b ends A234/7 (end of chapter “The Ascetic and the 155 folio with no text: 155a “The Ascetic Kills the
Weasel”) Faithful Weasel”; 155b blank
Chapter: The King and Fanza 144) 156 folio with no text: 156a blank; 156b “The Prince
95b ends A282/3 Kills Fanza’s Chick”
157a blank space for “The Prince Kills Fanza’s
Gap Chick”
145) 157b: “Fanza Blinds the Prince”
146) 158 folio with no text: 158a “Fanza Refuses to Return
to the King”; 158b blank
159a blank space for “Fanza Refuses to Return to
the King”
147) 160 folio with no text: 160a blank; 160b “The Wild
Elephant That is Captured with the Aid of a Tame
One” (oriented sideways) (Fig. 6.12b)
161a space for “The Wild Elephant That is Captured
with the Aid of a Tame One”
Chapter: The King and the Brahmans 148) 161 bis (unnumbered folio between 161 and 162) –
96a begins A238/18 blank space for “The King Tells His Dream to the
Brahmans”
149) 162 folio with no text: 162a “The King Tells His
Dream to the Brahmans”
i This is not Kardana dropping figs in the water (Barrucand, “Kopie,” 36), but the later episode in the chapter.
j The iconography of this is identical to that of the following illustration, except that the scene is outdoors. The text at the bot-
tom of the painting begins the substory.
Confluence of Cultures 185
105) 97a “The King Discusses His Dream with Irakht” 150) 165a blank space for “The King Discusses His
Dream with Irakht”
106) 97b “The Sage Kibayrun Interprets the King’s 151) 165b blank space for “The Sage Kibayrun Interprets
Dreams” the King’s Dreams”
107) 98b “The King in His Audience Hall” 152) 166b blank space for “The King in His Audience
Hall”
167 folio with no text: 167a “The King in His
Audience Hall”; 167b blank
108) 99b “The King Rebukes Irakht for Having Chosen 153) 168b “The King Rebukes Irakht for Having Chosen
the Crown” the Crown”
109) 100b “Bilad Pretends to Have Executed Irakht” 154) 169b “Bilad Pretends to Have Executed Irakht”
110) 101a “The Pigeon Who Unjustly Kills His Mate” 155) 170a “The Pigeon Who Unjustly Kills His Mate”
111) 101b “The Man, the Monkey and the Lentils” 156) 170b “The Man, the Monkey, and the Lentils”
157) 171b “The King and Bilad”
Gap 158) 173a “The King and Bilad”
102b ends A248/9
103a begins A252/11
112) 104a “The King and Bilad”k 159) 174b space for “The King and Bilad”
175 folio with no text: 175a “The King and Bilad”;
175b blank
113) 105a “The King and Bilad” 160) 176b “The King and Bilad”
114) 106a “The King and Bilad” 161) 177b “The King and Bilad”
115) 107b “The King and Bilad” 162) 179a “The King and Bilad”
116) 108a “The King and Bilad” 163) 179b blank space for “The King and Bilad”
180 folio with no text: 180a “The King and Bilad”;
180b “Bilad Restores Irakht to the King”
117) 108b “Bilad Restores Irakht to the King” 164) 181a blank space for “Bilad Restores Irakht to the
King”
Chapter: The Hunter and the Lioness
118) 109a “The Hunter Kills the Cubs” 165) 182 folio with no text: 182a blank; 182b “The Hunter
folio has been bound flipped horizontally; 109b Kills the Cubs” (oriented sideways)
has the caption of the opening of the chapter, and 183a blank space for “The Hunter Kills the Cubs”
should precede 109a) (Fig. 8b)
119) 110a “The Jackal Asks the Cause of the Lion’s 166) 182b “The Jackal Asks the Cause of the Lion’s
Lament” Lament”
k This and the following four paintings have been numbered 4–8 by a later hand, perhaps that of the artist of the Paris copy,
who may have been trying to keep track of so many almost identical scenes.
186 O ’ Kane
Chapter: The Lion and the Jackal 169) 186b “The Lion and the Jackal”
122) 113b “The Lion and the Jackal” 170–71) 187b top and bottom, identical, “The Lion
Tries to Persuade the Ascetic Jackal to Become His
Gap Minister”; in each the jackal is outlined only in
From here to the end of the manuscript black ink, and the caption is “The Lion and Dimna”
172) 188b “The Animals Accuse the Jackal” (according
to the text); captioned “The Lion and His Mother,”
the painting, unusually, covers part of the text, and
shows only the lion
173) 189a “The Lion’s Mother Intervenes” (according to
the text); captioned “The Lion and His Mother,”
the painting, unusually, covers part of the text, and
shows only the lion
174) 189b “The Vindication of the Jackal” (according to
the text); captioned “The Lion and Dimna”
175) 190a “The Vindication of the Jackal” (according to
the text); captioned “The Lion and Dimna”
l A village called Sanjalaf is known in the Egyptian province of Minufiyya. The manuscript was acquired by the French consul
in Egypt from 1772 to 1822 (Barrucand, “Un manuscrit arabe illustré de Kalila Wa Dimna,” 83), so there is a possibility that the
copy was made in Egypt, presumably in Cairo.
is marked by a jar. He finds the passage but not the similar ink and handwriting to the main text, so
jar, and returns to remonstrate with the wife con- they are presumably by the same calligrapher.
cerning her false directions. They remain arguing The first caption reads “Image of the man enter-
until the husband finds them and beats the lover. ing upon his wife [and] finding a man whom she
The story is illustrated by two paintings, the first of has hidden in the passageway;”12 the second reads
the wife and her lover arguing (Fig. 6.3a), the sec- “Image of the house owner who has come upon his
ond of the husband beating the lover (Fig. 6.3b). wife’s lover and beats him.”13
In the first the lover is a Mongol, but in the sec- The substory of “The Simpleton and the Rogue”
ond a Mongol is shown beating an Arab. This is is illustrated in four paintings. In the first two the
evidence first of all that the painter, unlike the simpleton is Arab and the rogue Mongol, but in
celebrated illustrator of the Maqāmāt of al-Wāsiṭī, the third and fourth (Fig. 6.11), in the company of
for instance,11 was not the calligrapher of the text. the judge, both men are dressed in Arab garb. In
Second, it calls into question whether he even “The Rat, the Ascetic, and His Guest” both men in
read the captions that the calligrapher occasion- the first illustration are Arab, but in the two subse-
ally provided for the paintings. The captions, usu- quent paintings the ascetic is a Mongol.14
ally arranged vertically to one side, are written
in smaller script than the main text, but in very 12 Ṣurat rajul dakhala ʿala zawjatihi fa-wajada ʿandaha
rajul wa-qad khabatahu fi-l-jubb.
13 Ṣurat ṣāḥib al-bayt wa-qad dakhala ʿalayhi ḥarīf zaw-
11 David James, A Masterpiece of Arab Painting: The jatihi fa-qāma ilayhi yaḍribahu (I would like to thank
“Schefer” Maqāmāt Manuscript in Context (London: Dina Montasser for her help with the captions).
East & West Publishing, 2013); Bernard O’Kane, “Text 14 Another such potential lack of continuity appears in
and Paintings in the al-Wāsiṭī ‘Maqāmāt,’” Ars Orientalis “The Ascetic and the Weasel.” In the first illustration
42 (2012): 41–55. (f. 94b), the ascetic seems to be Arab, but the face has
188 O ’ Kane
figure 6.1a
Architectural scene. Paris,
BnF, Arabe 3475, f. 10b
figure 6.1b
“Kisra Orders the Treasury
to be Thrown Open to
Burzuya”. Rabat, BRR, ms.
3655, f. 2b
Confluence of Cultures 189
figure 6.2a “The Thief and the Moonbeam: The Thief Approaches the Bed”. Rabat, BRR, ms. 3655, f.
9a
figure 6.2b “The Thief and the Moonbeam: The Thief is Beaten as He Falls Through the Roof”.
Rabat, BRR, ms. 3655, f. 9b
190 O ’ Kane
figure 6.3a “The Imprudent Lovers: The Wife and Her Lover Arguing”. Rabat, BRR, ms. 3655, f. 10a
figure 6.3b “The Imprudent Lovers: The Husband Beats the Lover”. Rabat, BRR, ms. 3655, f. 10b
Confluence of Cultures 191
figure 6.4 “The Condition in Which Man is Born”. Rabat, BRR, ms. 3655, f. 15b
Further evidence that the painter was not the The first illustration in the story of “The Thief
calligrapher is underlined by several choices as and the Moonbeam” has the thief in the bedroom
to what is or is not illustrated in the paintings. while the couple are sleeping (Fig. 6.2a). According
The point of the story mentioned above, “The to the text, the thieves should still be on the roof
Imprudent Lovers,” is that the jar is not to be waiting for the householder and his wife to fall
found where the wife said it was, confusing the asleep at this point. The painter seems strongly
lover (Fig. 6.3a). Yet the jar and its stand, as tall as reluctant to show architecture and one could
the man, are the only other furniture visible in the argue that he chose the easier option of having the
painting (apart from the cushion upon which the thief already in the bedroom, rather than on the
woman is sitting). This could be thought of as a roof of the house, as he should be, even though
mistake made by the artist, although it is also pos- this sequence is contradicted by the next painting,
sible to view it as the artist’s visualization of the which shows him being beaten while falling down
main subject of the dispute between the wife and the moonbeam (Fig. 6.2b). However, one could
her lover. also blame the caption writer, who describes the
scene as an “Image of the family asleep while the
thief proceeds towards them.” One would have
been damaged and repainted and it is therefore impos- thought that the calligrapher would not make such
sible to be sure of this (in the copy, f. 154b, the render-
ing seems to be of an Arab). In the second painting, at
a mistake, but perhaps he added the captions after
the dénouement where he finds the dead snake beside copying the whole text, or even after the paintings
the cot (f. 95a), he is clearly Mongol. were done, rather than before, with only a cursory
192 O ’ Kane
figure 6.5a
“The Ascetic and the Robe:
The Madam’s Stratagem
Backfires”. Paris, BnF, Arabe
3475, f. 50b
figure 6.5b
“The Merchant’s Wife and
the Slave Make Love”. Paris,
BnF, Arabe 3475, f. 90a
glance at the text to remind himself of the main Rabat painting (Fig. 6.7) is the earliest surviving
actors. In any event, he soon tired of this task, manuscript illustration, and exhibits some confu-
with, other than single names written beside the sion in the separation of the depiction of the edge
animals, no captions appearing after “The Monkey of the well and the water within the well. In many
and the Carpenter” (f. 20a). versions of the text,15 although not in this one, the
The most frequently illustrated episode in the hare asks the lion to lift him up to his breast to look
substory of “The Hares and the Lion” comes at the down at the other lion: this, and its reflection, is
climax, when the hare and the lion look into the
well, and are both reflected in the water below. The 15 A102/15: min ṣidrihi.
Confluence of Cultures 193
figure 6.6a
“The Crow and the Snake:
The Tortoise and the Crab”.
Rabat, BRR, ms. 3655, f. 31a
figure 6.6b
“The Crow and the Snake:
The Tortoise and the Crab”.
Paris, BnF, Arabe 3475 f. 54a
what is shown on the right side of the painting. In artist conflated paintings from two earlier, and
the Mamluk-illustrated versions, the well is also lost, manuscript traditions.
shown as a hole in the ground, but in these the One strange error is the substitution of a tor-
rabbit is not between the lion’s paws but oppo- toise for a crane (ʿuljūm) in the substory of “The
site it on the other side of the well.16 This is also, Crane, the Crab, and the Fish” (ff. 31a and 32a,
erroneously, shown in Rabat; it seems that the Fig. 6.6a). The story involves the flight of the crane,
so the substitution of the tortoise makes no sense
here. The term used, ʿuljūm, is admittedly one that
16 O’Kane, Early Persian Painting, figs. 36–37, 87 (Oxford has a variety of meanings, “frog” being an alterna-
and Cambridge mss.); Hans-Caspar Graf von Bothmer, tive, but the appearance of the tortoise, found in
Kalila Und Dimna: Ibn al-Muqaffaʿ’s Fabelbuch in Einer
Mittelalterlichen Bilderhandschrift, Cod. Arab. 616 Der
no other illustrated manuscript, is inexplicable.
Bayerischen Staatsbibliothek München (Wiesbaden: This is all the more so in that in the later story of
Reichert Verlag, 1981), 124. “The Crane, the Snake, the Crab, and the Weasel”
194 O ’ Kane
figure 6.7
“The Hare and the Lion:
The Hare and the Lion
at the Well”. Rabat, BRR,
ms. 3655, f. 33b
figure 6.8a
“The Horse Whose Good
Qualities Were Its Ruin”.
Rabat, BRR, ms. 3655, f. 37a
figure 6.8b
“The Hunter Kills the Cubs”.
Rabat, BRR, ms. 3655, f. 109a
Confluence of Cultures 195
figure 6.9a
“The Bee Imprisoned in the
Waterlily”. Rabat, BRR, ms.
3655, 38a
figure 6.9b
“The Bee Imprisoned in the
Waterlily”. Paris, BnF, Arabe
3475, 67a
figure 6.10
“The Waterfowl and the Spirit of
the Sea”. Paris, BnF, Arabe 3475
f. 75a
196 O ’ Kane
figure 6.11a
“The Simpleton and the
Rogue: The Two Men before
the Judge”. Rabat, BRR,
ms. 3655, 47b
figure 6.11b
“The Simpleton and
the Rogue: The Rogue’s
Stratagem is Discovered”.
Rabat, BRR, ms. 3655, 49a
(ff. 48a–b), where the animal is also termed ʿuljūm, rogue, in response to the judge’s question, “Was
it is drawn as a bird.17 there any witness to the transaction?”, replies:
Returning to the story of “The Simpleton and “the tree”; but the actual visit to the tree where
the Rogue,” the third illustration, according to the rogue has concealed his father is the climax of
the text, should show the two men making their the story (Fig. 6.11b). At first sight, since Fig. 6.11a
first appearance before the judge (Fig. 6.11a). The shows the qadi mounted on a mule pointing to the
tree, this is a defective illustration of the climax
17 The ambiguity of the term is reflected in the illustra-
of the story; it lacks the fire and the father in the
tions in the Persian cycle, which are of a frog rather tree. It is essential to the plot that the judge not
than a crane. be near the tree at this point, so this is probably
Confluence of Cultures 197
figure 6.12a
“The Mutual Enmity
Between the Lion and the
Elephant”. Rabat, BRR, ms.
3655, f. 67b
figure 6.12b
“The Wild Elephant That
is Captured with the Aid
of a Tame One”. Paris, BnF,
Arabe 3475, f. 160b
a misunderstanding, perhaps based on an illus- The substory of “The Hypocritical Cat” con-
tration the artist half-remembered from another cerns a partridge, a rabbit, and a cat. The artist
Kalila and Dimna manuscript. It is also possible has made the bird (ṣafrid) a crow instead of a
that, as in the case of “The Imprudent Lovers,” the partridge (strangely, as the Arabic name comes
artist decided to gloss the painting with what he from the singing qualities of the bird, hardly the
thought was most important – the tree – although crow’s strong point) (Rabat ff. 81b, 82b). The chap-
the incorporation of the mounted judge makes ter in which the story is nested, “The Owls and
this less likely.18 the Crows,” already provided multiple options
18 One bizarre feature of the copy is the addition, as in the to then all should have been similarly inscribed. Some
two paintings in Fig. 11, of the words laʿanahum Allah faces are rubbed out, as on ff. 47b, 63a, 74a (the deer),
(God has cursed them: Qurʾan 33:57) scrawled above but this could be simply accidental damage, as is the
the figures. The same words are also found beside case with the elephant on f. 81b. The head of the snake
the human figures in ff. 46a, 47a–6, 49a, 50a, 58b, 62a, on ff. 85a–b seems to have been defaced deliberately, as
63b, and 68b. It is puzzling that only these scenes do those of the owls on ff. 87a and 88a. But even though
with human figures (and not all of them incorporate owls are considered unlucky animals in the Middle
immoral characters) and not others, or indeed that any East, there was no systematic attempt to deface them
scenes, were so inscribed. If figures were being objected in the manuscript.
198 O ’ Kane
for the depiction of crows, so this could just have 1 Unique Episodes
been a case of choosing an easier option. Another
example of indifference – or more likely, misin- As mentioned above, the original total of around
formation – is the birds that attack the falconer 187 paintings makes Rabat the most extensively
in the substory of “The Parrots and the Governor’s illustrated Kalila and Dimna manuscript known. It
Wife” (Rabat f. 126b).19 They should be the falcons can then be no surprise that so many of its subjects
trained by their owner, the falconer, but have been are unique to the illustrative cycle. One of these is
represented as parrots instead. “The Bee Imprisoned in the Waterlily” (Fig. 6.9a).
In the chapter of “The Owls and the Crows,” a The text, as with “The Perils of Life,” refers to a bee
crow spy volunteers to infiltrate the owls. One that enjoys the perfume of the waterlily so much
would have thought that it was essential to the that it forgets to fly away and is trapped when the
story that only one crow be painted in the scene flower closes at night. The painting in the manu-
(and such is the case in all other illustrated ver- script is of a pool in which the flowers are growing;
sions), but here (Rabat f. 84a) the artist has opted while the waterlilies are given prominence, the
for symmetry, not usually a strong concern of his, artist has omitted the bee. This, it should be noted,
with an owl on the ground and an owl in the tree is thus the first landscape painting extant in any
being balanced by a crow on each level. example of Islamic book painting. However, it is
The reluctance of the artist to depict architec- probably not a coincidence that it was produced
tural surroundings has been noted already,20 and under Ilkhanid patronage, when the importation
for some reason this continued to be a frequent of Chinese landscape-dominated paintings would
failing in later cycles.21 The most detailed archi- have begun. Pure landscapes that owe a debt to
tectural interior is that in “The Thief is Beaten as the contemporary Chinese style appear shortly
He Falls Through the Roof” (Fig. 6.2a), helped by after in two paintings of the early fourteenth cen-
the desire to show the moonbeam on which the tury Jāmiʿ al-tawārīkh.22 In the Paris manuscript
thief was sliding and the lantern in the muqarnas (Fig. 6.9b), the landscape of trees is the sole one
roof though which it shone. The lengths to which in the copy that is clearly by a different, presum-
the artist would go to avoid showing architecture ably later, artist, who declined to reproduce the
is seen in the two illustrations of “The Devout Man original and added a three dimensionality not
and the Pot,” in which a jug hanging from the ceil- present in other paintings. He inexplicably added
ing starts the man daydreaming. In the first, “The a lamp with a stand hovering above the ground to
Devout Man Dreams of Riches” (Rabat f. 93b), the right.
arched corners in red are the only concession to Other original subjects, such as “The Qadi of
the embellishment of the plain blue frame that Marv Who Decided in Favor of Whichever Plaintiff
supposedly delineates the room; in the second, He Heard Last,” (f. 14a)23 demanded little in the
“The Devout Man Knocks a Pot of Oil on His Head” way of innovation, showing as it does a simple
(Rabat f. 94a), the scene is now outdoors, and the grouping of three people. More original is the birth
ceiling almost resembles a grassy hillock. scene of “The Condition in Which Man is Born”
(Fig. 6.4). It is not clear, as has been suggested,24 misplaced, since they are invariably rather puny,
that the artist based this on the well-known scene much smaller than they should be in real life.29
in al-Wāsiṭī’s Maqāmāt.25 In both, the woman sits
upon a birthing stool and is helped by handmaid-
ens on either side,26 but this method of delivery 2 The Copy
was the standard for pre-modern childbirth in the
Middle East, and the poses of the principal figures What do the paintings in the copy tell us of the
are quite different in each case. missing originals? It should be noted first of all
Another original subject, “The Horse Whose that the artist was looking at the Rabat manu-
Good Qualities Were Its Ruin” (Fig. 6.8a), enabled script for models and not at any others, since he
the artist to depict not only the horse but also its repeats the mistakes of the Rabat artist, such as
dashing Mongol rider bearing down with a spear showing an extra hare in “The Hare and the Lion
upon an invisible enemy. As Marianne Barrucand at the Well,” depicting parrots instead of a falcon
noticed,27 “The Hunter Kills the Cubs” (Fig. 6.8b) in “The Falcons Attack the Falconer,” and showing
shows not just the skinned cubs, as is usual in a tortoise instead of a crane in the substory of “The
other illustrations of this scene, but a similar deb- Crane, the Crab and the Fish” (Fig. 6.6b). He also
onair Mongol with a spear, riding a horse whose had trouble understanding what a tortoise should
elaborately knotted tail echoes the elegance of look like, since he uses the flower-like tri-petal feet
its rider. of the tortoise for the tortoise’s head as well.30 It
Other original episodes involved elephants, is quite possible that some of the other irregulari-
such as “The Mutual Enmity Between the Lion and ties, such as the inclusion of a jackal in the paint-
the Elephant”28 (Fig. 6.12a) and, in the copy only, ing of “The Louse and the Flea” (Paris f. 62a), his
“The Wild Elephant That is Captured with the Aid depiction of a female instead of a male in the same
of a Tame One” (Fig. 6.12b). The calligrapher made painting, or the painting of a girl holding a rat in
the choice of where to leave spaces, although the “The Sun Explains that a Cloud is More Powerful
mistakes in interpretation noted above suggest he than Himself” (Paris f. 135b), are the copyist’s
was not also the artist. His faith in the ability of the mistakes.
artist to capture elephants faithfully was slightly Occasionally the copyist captures the spirit
of what must have been a majestic original,
notably in “The Assembly of the Birds” from the
tale of “The Waterfowl and the Spirit of the Sea”
(Fig. 6.10). The king of the birds, the ʿanqā, is rep-
24 Barrucand, 25. resented as a double-headed eagle.31 It takes up
25 Richard Ettinghausen, Arab Painting (Cleveland OH:
Albert Skira, 1962), 121.
all of the page, with four smaller birds arranged
26 The woman on the right is holding a rectangular fan
and is reminiscent of the figure in the nearly contem- 29 E.g. those in the substory of “The Hares and the
porary (1287) frontispiece of the Rasāʾil Ikhwān al-Ṣafā: Elephants: The Council of the Hares”: ff. 80a-b and
Ettinghausen, 98. Although wearing a turban, she has a 81a. More realistic elephants within the Arab paint-
physiognomy more Mongol than Arab. ing tradition are to be found in the Manāfiʿ-i Ḥayavān,
27 Barrucand, “Le Kalila wa Dimna de la Bibliothèque Pierpont Morgan Library, Ms 500, f. 13b, illustrated in
royale de Rabat,” 25. Ettinghausen, Arab Painting, 134.
28 This is not, as might first be thought, a stray from the 30 Examples are found on ff. 106b, 118a, 119a, 146b,
story of “The Wolf, the Crow, the Lion, the Jackal and and 147a.
the Camel” (in which the lion is first injured by an 31 It was not invariably represented thus; for instance, in
elephant), but an illustration of a metaphor for mutual the Ilkhanid London Qazvīnī, it is painted as a large
enmity in the conversation between the rat and the cock: Stefano Carboni, The ‘Wonders of Creation’ and
crow. the Singularities of Painting: A Study of the Ilkhanid
200 O ’ Kane
almost symmetrically above. This heraldic image One puzzling painting in the copy is on f. 10b
is paralleled in just one other manuscript, the (Fig. 6.1a). A blank space was left in the copy at the
Cambridge manuscript,32 where three birds are bottom of f. 9b, which corresponds in the text to
instead arranged vertically to either side of the “Kisra Orders Burzuya to Obtain the Book of Kalila
double-headed eagle. and Dimna.” At the top of f. 11a, where the text con-
Another surprising image in the copy is that tinues, another blank space for a painting was left.
illustrating the substory of “The Painter and His The inserted f. 10 depicts “Kisra Orders Burzuya
Cloak.” In the second of the two illustrations of this to Obtain the Book of Kalila and Dimna” on f. 10a,
episode, “The Merchant’s Wife and the Slave Make and a rendering of architecture on f. 10b, which
Love” (Fig. 6.5b), the depiction of intercourse is presumably also copied an original painting.
explicit. It has been suggested that this might not However, it is difficult to grasp what the subject
have been the case in the Ilkhanid original,33 but was, as it consists of fragments of mostly archi-
in the copy the naked buttocks of the couple in tecture in separate compartments: a metal grille,
“The Madam’s Stratagem Backfires” are visible (as a hanging lamp, imitation marble panels, blind
they always are in the later Mamluk examples) niches and a half arch, spear points, and circles
(Fig. 6.5a). There are also earlier parallels: in one and semi-circles. Given the aversion of the origi-
of al-Wāsiṭī’s paintings in the Maqāmāt, Abu Zayd nal artist and the artist of the copy to depicting
is shown displaying his penis.34 Admittedly the architecture, it is difficult to believe that it was an
image of Abu Zayd is not within an erotic con- invention of the copyist, although the text at this
text, but such erotic imagery may have been part point does not indicate any obvious architectural
of the appeal of Kalila and Dimna manuscripts, reference. The structure that it most resembles is
with their several risqué episodes, for many of the treasury that is shown on f. 3b of the original,
their readers.35 Such explicit imagery was not by “Kisra Orders the Treasury to be Thrown Open to
any means limited to non-royal manuscripts, the Burzuya” (Fig. 6.1b) – in the Paris manuscript the
nature of manuscript collections making them an space for copy of the painting (f. 18a) is blank.
ideal vehicle for the pleasures of private perusal.36
in Baghdad have been well documented. The art- Bilderhandschrift, Cod. Arab. 616 Der Bayerischen
ist of the Rabat manuscript, if not as daring or Staatsbibliothek München. Wiesbaden: Reichert
well-rounded39 in his repertoire as the celebrated Verlag, 1981.
painters of the great Maqāmāt manuscripts earlier Carboni, Stefano. The ‘Wonders of Creation’ and the
in the 13th century (al-Wāṣiṭī and St Petersburg), Singularities of Painting: A Study of the Ilkhanid
was nevertheless well acquainted with their land- ‘London Qazvīnī’. Edinburgh: Edinburgh University
scape idiom and was a sympathetic and accom- Press, 2015.
plished painter of the bulk of the subjects in the Ceylan, Ebubekir. “Baghdad, from 1500 to Iraqi Inde-
manuscript: animals. The manuscript is the earli- pendence.” In Encyclopaedia of Islam, third edi-
est known to incorporate depictions of Mongols in tion, edited by Kate Fleet, Gudrun Krämer, Denis
Islamic painting. It also has the largest number of Matringe, John Nawas, and Everett Rowson. Leiden:
illustrations of any known Kalila and Dimna man- Brill, 2014.
uscript. Now that we have a better idea of its origi- Ettinghausen, Richard. Arab Painting. Cleveland OH:
nal cycle, we can appreciate both the novelty and Albert Skira, 1962.
variety that the painter brought to the formidable Hartmann, Angelika. An-Nāṣir li-Dīn Allāh (1180–1225):
task of illustrating the spaces left for 187 paintings. Politik, Religion, Kultur in der späten ʿAbbāsidenzeit.
Berlin ; New York: De Gruyter, 1975.
Ibn al-Muqaffaʿ, ʿAbd Allāh. Kalila wa Dimna. Edited by
Bibliography ʿAbd al-Wahhâb ʿAzzâm. 2nd edition. Cairo, 1981.
James, David. A Masterpiece of Arab Painting: The
Adamova, Adel T. Persian Painting and Drawing of the “Schefer” Maqāmāt Manuscript in Context. London:
15th to 19th Centuries from the Hermitage Museum. East & West Publishing, 2013.
St Petersburg: AO “Slavii͡a͡,” 1996. Komaroff, Linda, and Stefano Carboni, eds. The Legacy
Allsen, Thomas T. Commodity and Exchange in the of Genghis Khan: Courtly Art and Culture in Western
Mongol Empire: A Cultural History of Islamic Textiles. Asia, 1256–1353. New York: Metropolitan Museum of
Cambridge: Cambridge University Press, 1997. Art, 2002.
Barrucand, Marianne. “Kopie: Nachempfindung oder Leoni, Francesca, and Mika Natif, eds. Eros and
Umgestaltung, am Beispiel Mittelalter Arabi- Sexuality in Islamic Art. Farnham, Surrey ; Burlington
scher Bilderhandschriften und ihrer Osmanischen VT: Ashgate Publishing, 2013.
Kopien.” In Rezeption in der Islamischen Kunst, O’Kane, Bernard. Early Persian Painting: Kalila and
edited by Barbara Finster, Christa Fragner, and Herta Dimna Manuscripts of the Late Fourteenth Century.
Hafenricher, 19–41. Beyrouth ; Stuttgart: Beiruter London: I.B. Tauris ; Cairo: AUC Press, 2003.
Texte und Studien ; F. Steiner, 1999. O’Kane, Bernard. “Islamic Art and Architecture in
Barrucand, Marianne. “Le Kalila wa Dimna de la Pre-Mongol Baghdad.” In Baghdad – Space of
Bibliothèque royale de Rabat: un manuscrit illustré Knowledge, edited by Isabel Toral-Niehoff and Jens
il-khanide.” Revue des Études Islamiques, numéro Scheiner, in press.
spécial en l’honneur de Dominique Sourdel 54 (1986): O’Kane, Bernard. “Text and Paintings in the al-Wāsiṭī
17–51. ‘Maqāmāt.’” Ars Orientalis 42 (2012): 41–55.
Barrucand, Marianne. “Un manuscrit arabe illustré de O’Kane, Bernard. “The Arboreal Aesthetic: Landscape,
Kalila wa Dimna du XIIIe siècle et sa copie otto- Painting and Architecture from Mongol Iran to
mane.” Archéologie Islamique, no. 2 (1991): 81–95. Mamluk Egypt.” In The Iconography of Islamic Art:
Bothmer, Hans-Caspar Graf von. Kalila und Dimna: Ibn Studies in Honour of Robert Hillenbrand, edited by
al-Muqaffaʿ’s Fabelbuch in Einer Mittelalterlichen Bernard O’Kane, 223–51. Edinburgh: Edinburgh
University Press, 2007.
39 Particularly in his aversion to architecture.
202 O ’ Kane
Roxburgh, David J. “‘Many a Wish Has Turned to Surieu, Robert. Sarv-é naz: An Essay on Love and the
Dust’: Pir Budaq and the Formation of Turkmen Representation of Erotic Themes in Ancient Iran.
Arts of the Book.” In Envisioning Islamic Art and Geneva: Nagel, 1967.
Architecture: Essays in Honor of Renata Holod, edited Ṭabarī, Muḥammad ibn Jarīr al-. The History of al-Ṭabarī:
by David J. Roxburgh, 175–222. Leiden: Brill, 2014. An Annotated Translation, Volume 29: al-Mansur and
Sims-Williams, Ursula. “An Illustrated 14th Century al-Mahdi, A.D. 763–786; A.H. 146–169. Translated by
Khamsah by Khvaju Kirmani.” Asian and African Hugh Kennedy. Albany NY: State University of New
studies blog (blog), August 1, 2013. http://british York Press, 1990.
library.typepad.co.uk/asian-and-african/2013/07/an
-illustrated-14th-century-khamsah-by-khvaju-kirma
ni.html.
partie 3
Construire l’image / Building the Image
∵
7
Perles de sagesse : peintures d’une copie mamelouke du
Sulwān al-muṭāʿ fī ʿudwān al-atbāʿ
Mounia Chekhab-Abudaya
Le manuscrit MS.27.1999 du Musée d’Art islamique cours de sa vie, séjourné entre autres en Égypte,
de Doha (Qatar) est une copie mamelouke du en Ifrīqiyya, et en Syrie, à Alep notamment, mais
Sulwān al-muṭāʿ fi ʿudwān al-atbāʿ (« Consolation également à Hama où il aurait fini ses jours. Il est
du souverain face à l’hostilité de ses sujets »), l’auteur d’exégèses du Coran, d’ouvrages de théo-
texte écrit par Abū ʿAbdallāh Muḥammad b. Abī logie et de fiqh, de philosophie et de grammaire4.
Muḥammad b. Ẓafar al-Siqillī (m. 565H/1170)1. Le Sulwān al-muṭāʿ fi ʿudwān al-atbāʿ est un
Cette copie a fait l’objet d’une monographie en 1985 texte qui appartient au genre littéraire médiéval du
dans laquelle on retrouve un facsimilé du manus- « miroir des princes », soit l’énoncé, par un auteur,
crit, une réédition de la traduction anglaise de 1852 d’un ensemble de règles d’éthique de bon gouver-
et une étude réalisée par A.S. Melikian-Chirvani nement (adab). Le premier miroir des princes fut
qui offre des éléments intéressants sur le contexte composé dans un contexte persan, mais l’établis-
de production de ce Sulwān2, au travers notam- sement de l’adab en tant que genre littéraire dans
ment de comparaisons avec plusieurs manuscrits le monde arabe s’est accompagné du développe-
médiévaux3. Ibn Ẓafar al-Siqillī est né en Sicile en ment d’une multitude de « miroirs », dont les plus
497H/1104 et a été élevé à La Mecque. Il aurait, au célèbres sont ceux de ʿAbd al-Ḥamīd b. Yaḥyā (m.
132H/750) (Risāla ilā al-kuttāb « Épître aux secré-
1 Note de l’auteur : je souhaiterais remercier chaleureu- taires ») et d’Ibn al-Muqaffaʿ (m. 139H/757) (Kalīla
sement Amandine Adwan pour son aide précieuse dans wa-Dimna), écrits durant la période omeyyade5.
la compréhension du texte du Sulwān al-muṭāʿ fi ʿudwān
al-atbāʿ et du genre des miroirs des princes. Son exper-
tise m’a été d’une grande utilité dans l’élaboration de cet 4 D’autres sources relatent qu’il est probablement mort en
article. Un grand merci également à Moez Dridi pour son 567H/1171 ou en 598H/1201. L’œuvre du Sulwān, probable-
aide dans la confirmation du déchiffrage de certaines ins- ment la plus célèbre de cet auteur, a été lithographiée en
criptions, et merci à Mélisande Bizoirre d’avoir accepté 1861 au Caire, puis imprimée par la suite à Tunis en 1862 et à
de relire ce texte. Je souhaiterais également exprimer ma Beyrouth en 1882. Elle a été traduite en italien par Michele
gratitude au Musée d’Art islamique de Doha et en parti- Amari en 1851 à Florence dont le travail a été ensuite
culier au département Multimédia pour avoir entièrement retranscrit en anglais en 1852. Une autre traduction du
numérisé ce manuscrit, initiative qui je l’espère pourra texte existe en turc et a été réalisée par Kara Khalilzade en
encourager d’autres études sur cette fabuleuse copie. Mes 1868. Umberto Rizzitano, « Ibn Ẓafar », in Encyclopaedia of
remerciements vont également à l’Aga Khan Museum à Islam, Second Edition, éd. Peri Bearman et al., 3 (Leiden :
Toronto, à la Freer Gallery of Art de Washington D.C. et Brill, 1960-2007), 970. Il existe une édition critique du texte
à la Dar al-Athar al Islamiyyah au Koweït, pour m’avoir plus récente : Muhammad ibn Ẓafar al-Siqilli, The Just
transmis des informations sur les folios conservés dans Prince, A Manual of Leadership, Including an Authoritative
leurs collections. English Translation of the Sulwan al-Mutaʿ fi ʿUdwan
2 Nous avons choisi tout au long de ce travail de nous référer al-Atbaʿ (Consolation for the Ruler During the Hostility of
au texte avec le terme de Sulwān pour abréger le titre et Subjects), trad. Joseph A. Kechichian et R. Hrair Dekmejian
éviter de le répéter entièrement à chaque occurrence. (Londres : Saqi Books, 2003).
3 Assadullah S. Melikian-Chirvani, Sulwān al-Muṭāʿ fī ʿudwān 5 Pour plus d’informations sur le genre littéraire du
al-atbāʿ, An Art Historical Essay (Koweït : TRI Publishing, miroir, voir Clifford E. Bosworth, « Naṣīḥat al-mulūk » in
1985). Encyclopaedia of Islam, Second Edition, éd. Peri Bearman
D’après les différents miroirs des princes connus, Ibn al-Muqaffaʿ souligne, dans son introduction
la volonté de Dieu prévaut sur les impératifs de aux fables des deux chacals, l’intérêt de l’illustra-
gouvernement dans l’exercice du pouvoir. Ibn tion des fables pour le plaisir des souverains8. Il
Ẓafar décrit lui-même son texte comme une est donc particulièrement intéressant d’analyser
source de « réconfort » pour les « souverains », notre manuscrit en relation avec la production des
mais on peut le concevoir comme une fiction lit- copies illustrées de Kalīla wa-Dimna, qui peuvent
téraire constituée d’une succession d’intrigues avoir servi de source d’inspiration.
atemporelles où les personnages principaux sont À l’instar du Pañcatantra et du Kalīla wa-Dimna,
des archétypes tirés de l’histoire byzantine, sassa- la structure du Sulwān al-muṭāʿ fi ʿudwān al-atbāʿ
nide et islamique. suit une division en cinq chapitres (sulwāna), qui
Le texte du Sulwān a été composé en 545H/1150 ne sont pas organisés autour des fables principales,
et revu en 554H/1159 par Ibn Ẓafar qui le dédia mais autour de préceptes moraux ou spirituels :
cette fois au qāʾid de Sicile, Abū ʿAbd Allāh – Sulwānat al-tafwīḍ : s’en remettre à Dieu ; réflé-
Muḥammad Ibn Abū al-Qāsim Ibn ʿAlī al-Qurashī, chir soigneusement face à divers ennemis pour
connu sous le nom de Ibn Ḥajar. L’auteur, qui avait choisir celui qu’il faut combattre.
déjà réalisé plusieurs commentaires des Maqāmāt – Sulwānat al-taʿassā : la force et le courage
d’al-Ḥarīrī (445-515H/1054-1122), l’aurait rédigé face à l’adversité ; pourquoi un dirigeant doit
dans l’intention de produire un ouvrage proche de apprendre à tenir compte des bons conseils.
celui d’Ibn al-Muqaffaʿ, mais qui, en plus des fables – Sulwānat al-ṣabr : la patience.
animalières, comprendrait des anecdotes mettant – Sulwānat al-riḍā : le contentement et la satis-
en scène des personnages historiques6. La compo- faction de l’homme face à la volonté de Dieu ;
sition du texte, qui compte des histoires prenant savoir développer un penchant pour la ruse et
place en contexte persan ou chrétien, témoigne les stratagèmes.
aussi de cette influence de Kalīla wa-Dimna, sur- – Sulwānat al-zuhd : l’abnégation et l’ascétisme ;
tout par le biais de sa traduction en syriaque7. savoir se détourner des richesses du monde
pour se concentrer sur l’élévation de l’âme9.
L’auteur explique dans cet ouvrage que la sulwāna
et al., 3 (Leiden : Brill, 1960-2007), 984-988 ; Jocelyne est, selon les croyances populaires arabes, une
Dakhlia, « Les miroirs des princes islamiques : une moder-
perle ayant le pouvoir de consoler un être malheu-
nité sourde ? », Annales. Histoire, Sciences sociales 57, n° 5
(2002) : 1191-1206 ; Ann K.S. Lambton, « Islamic Mirrors for reux, lorsque celui-ci boit l’eau versée sur elle10.
Prince », in La Persia nel medioevo : Atti del Convegno inter- La narration fait office à la fois de distraction, de
nazionale (Rome : Accademia Nazionale Dei Lincei, 1971), guide pour un souverain idéal et de diatribe contre
419-442 ; Louise Marlow, « Advice and Advice Literature »,
in Encyclopaedia of Islam, Third Edition, éd. Kate Fleet
et al., 3 (Leiden : Brill, 1960-2007) ; Marlow, « Surveying 8 Voir introduction de ʿAbdallāh b. al-Muqaffaʿ, Kalilah
Recent Literature on the Arabic and Persian Mirrors for and Dimnah, trad. Jallad (Le Caire : Bulāq, 2002), 36.
Princes Genre », History Compass 7, n°2 (2009) : 523-238. Jallad base sa traduction sur l’édition imprimée en
6 Carl Brockelmann, « Kalīla wa-Dimna », in Encyclopaedia arabe de 1817 réalisée par Silvestre de Sacy.
of Islam, Second Edition, éd. Peri Bearman et al. 4 (Leiden : 9 Voir pour plus de détails sur le texte : Melikian-Chirvani,
Brill, 1960-2007), 503-6. La structure des récits enchâssés Sulwān, 51-64. R. Hrair Dekmejian et Adel Fathy Thabit,
est très similaire mais les éléments de narration sont en « Machiavelli’s Arab Precursor: Ibn Ẓafar al-Siqillī, »
revanche différents. British Journal of Middle Eastern Studies 27, n° 2 (2000) :
7 François de Blois, « The Pañcatantra from India to the 125-137.
West – and Back, a Mirror for Princes from India », in A 10 Le terme sulwān dérive du verbe salā/yaslū signifiant
Mirror for Princes from India: Illustrated Versions of the « consoler après la perte de quelque chose ». La 5e
Kalilah wa Dimnah, Anvar-i Suhayli, Iyar-i Danish, and forme du verbe, tasallā, signifie « consoler, divertir ».
Humayun Nameh, éd. Ernst J. Grube (Bombay : Marg Albert de Biberstein Kazimirski, Dictionnaire Arabe-
Publications, 1991), 10-15. Français, (Paris : Maisonneuve et Cie, 1860) : 1132.
Perles de sagesse 207
Cet article tentera de saisir la relation entre la couleurs : rouge, blanc, bleu, vert et jaune. Elles
narration et les peintures, au travers de compa- présentent des caractéristiques communes : des
raisons avec des copies arabes des xiiie et xive écoinçons colorés aux angles dans les scènes
siècles de Kalīla wa-Dimna, source d’inspiration d’intérieur lesquelles sont coiffées de dômes ;
d’Ibn Ẓafar, mais aussi œuvre littéraire parmi les une végétation stylisée de plusieurs couleurs
plus illustrées à cette époque ou en tout cas dont le plus souvent placée au centre des composi-
il a survécu de nombreux exemplaires. Si l’étude tions par effet de symétrie, avec entre autres la
détaillée d’A.S. Melikian-Chirvani permet de représentation de la fleur de lotus et d’un ciel en
replacer la réalisation de cette copie vers 1330- demi-cercle présentant plusieurs dégradés de bleu
1340, peut-être en Syrie, il s’agit dans ce travail de et d’arabesques blanches symbolisant les nuages
déterminer quels éléments de l’iconographie cor- dans un style proche du motif chinois tchi. La
respondent à l’utilisation de modèles et de codes représentation des personnages met en scène des
de représentations diffusés dès le xiiie siècle dans caractéristiques de la peinture mamelouke dans le
la peinture mamelouke, à l’inverse de ceux qui traitement des costumes et des traits du visage sur
répondent à l’illustration de ce texte en particu- lesquels nous nous attarderons ci-après17.
lier. Nous tenterons enfin d’offrir quelques pistes L’analyse du contenu des peintures permet de
de réflexion autour de l’attribution de cette copie distinguer trois ensembles ainsi qu’une illustra-
à une production syrienne et de la mettre en pers- tion isolée18 :
pective avec le contexte socio-politique15. 1. douze peintures mettent en scène des per-
Les peintures du Sulwān sont toutes exécutées sonnages, et en particulier des scènes de
sur fond d’or16 et présentent la même palette de cour (AKM12, ff.1, 31, 37, 41, 56, 58, 77, 86v, 97,
100v et 107v) ;
Naṣūḥ Pasha. La partie droite comprend deux lignes de 2. douze peintures présentent des scènes ani-
poésie. Les marges des ff.109 et 110, ainsi que l’inscrip- malières (ff.12, 12v, 42v, 44, 61v, 62, 63v, 76, 78,
tion, comportent un texte écrit en turc ottoman, dont
l’encre et la graphie sont similaires à l’inscription men-
F1954.2, LNS104MS, et 104) ;
tionnant Naṣūḥ Pasha au premier folio du manuscrit. 3. cinq peintures combinent figures humaines
Son nom apparaît une nouvelle fois dans la partie basse et animales (ff.32v, 39v, F1954.1, 90r, 104v) ;
du f.110r. Voir également à propos de Naṣūḥ Pasha et la 4. une peinture ne met en scène aucune figure
ʿĀdiliyya : Çiǧdem Kafescioǧlu, « In the Image of Rum:
(f.15v).
Ottoman Patronage in Sixteenth-Century Aleppo and
Damascus », Muqarnas 16, n° 1 (1999) : 77-97 et Bruce
Masters, « Janbulad Ali Pasha » in Encyclopedia of the
Ottoman Empire, éd. Gábor Ágoston et Bruce Masters
(New York : Facts on File, 2009) : 295-296. ultérieure ou simplement d’une phase de dorure inter-
15 Pour plus d’informations détaillées sur l’art de la pein- venant peu de temps après l’exécution des images.
ture mamelouke, voir entre autres Duncan Haldane, 17 Ainsi le personnage vêtu d’une robe bleue qui accom-
Mamluk Painting (Warminster : Aris and Phillips, pagne les deux animaux au folio 39v est très similaire
1978) ; Esin Atıl, Renaissance in Islam: Art of the à la représentation de ʿAyn Ahlīh au folio 37. Le peintre
Mamluks (Washington D.C. : Smithsonian Institution semble n’individualiser que quelques traits des per-
Press, 1981) ; Anna Contadini, éd., Arab Painting: Text sonnages qu’il représente, souvent des éléments du
and Image in Illustrated Arabic Manuscripts (Leiden, visage, qui permettent d’identifier certaines figures
Boston : Brill, 2007). d’une anecdote à l’autre ou dans la même anecdote.
16 L’or semble avoir été appliqué à une période ultérieure Dans l’histoire de Shāpūr au chapitre 2, les membres
à l’exécution des dessins et l’application des couleurs. de la cour de l’empereur byzantin sont représentés
Cela est visible notamment dans quelques peintures de façon individuelle dans les folios 31 et 32v de sorte
où l’or recouvre certains éléments peints (par exemple qu’ils soient reconnaissables d’une scène à l’autre, et ce
dans le folio F1954.2 de la Freer Gallery of Art). Après afin aussi de suivre le récit.
observation physique du manuscrit, il est difficile 18 La liste de l’ensemble des peintures et des anecdotes
cependant de savoir s’il s’agit d’une phase nettement qui leurs sont liées est dressée dans le tableau 7.1.
Perles de sagesse 209
figure 7.1 Prince à sa cour tenant une coupe contenant une perle, f.1r (MS.27.1999, MIA)
Certaines peintures font directement référence au l’effet de symétrie que l’on peut retrouver dans les
texte et d’autres correspondent en revanche à un modèles conventionnels de peintures animalières
parti pris de l’artiste d’employer des modèles déjà d’autres manuscrits illustrés de la même période
connus afin de simplifier la scène représentée. (Fig. 7.3).
Ainsi, pour ne citer que quelques exemples, Une des premières questions concernant le
le folio 1 du frontispice comprend l’image d’un rapport entre texte et image est de savoir quel
prince nimbé, sur son trône, qui tient une coupe processus a mené à la création de cette copie
dans sa main droite et regarde une perle dans sa illustrée et au choix des histoires à représenter
main gauche (Fig. 7.1). La transcription visuelle de parmi toutes celles qui constituent la narration. La
la narration permet de reconnaître la métaphore répartition des illustrations au sein du manuscrit
de la perle de sagesse contenue dans le titre et semble être tout à fait aléatoire. Comme l’a souli-
dans l’introduction de l’ouvrage, à travers le geste gné R. Hillenbrand, les peintures ne se succèdent
de boire de l’eau versée sur une perle. En revanche, pas nécessairement à intervalles réguliers19, ce
au folio 12 le serpent qui confronte Ẓālim n’est pas qui implique que le copiste a décidé ou savait à
représenté et seule la lecture de la fable laisse sup- l’avance ce qui devait être peint. Mais cela peut
poser la présence du serpent. Le texte vient dans ce également sous-entendre un travail collaboratif
cas compléter l’illustration. Le folio 15 comprend le entre le copiste et le peintre20. Ces derniers se
paysage sans les renards mais avec l’esquisse d’un
oiseau recouvert ultérieurement par de l’or qui ne 19 Robert Hillenbrand, « Mamlūk and Īlkhānid Bestiaries:
s’apparente pas à une quelconque partie du récit Convention and Experiment », in Studies in the Islamic
Arts of the Book, éd. Robert Hillenbrand (Londres : The
(Fig. 7.2). Au folio 62, le deuxième rat est simple- Pindar Press, 2012), 215.
ment suggéré par une deuxième queue peinte 20 Une copie mamelouke du Kitāb al-Ḥiyal, traité sur
derrière le premier, sans doute pour respecter les automates des frères Banū Mūsā, est conservée
210 Chekhab-Abudaya
figure 7.2 Scène abstraite illustrant le moment où Mufawwaḍ brûle Ẓālim vivant, f.15v (MS.27.1999, MIA)
figure 7.3 Les rats en conversation puis rencontrent la gerboise, ff.61v et 62r (MS.27.1999, MIA)
Perles de sagesse 211
sont peut-être concertés pour planifier l’ouvrage illustre les éléments du répertoire qui font écho
et ont décidé d’un commun accord des anecdotes à la maîtrise de motifs et de modèles spécifiques
à illustrer et du nombre de peintures pour cha- par les artistes. C’est cet académisme qui est le
cune d’entre elles. Nous avons pu d’ailleurs remar- reflet de l’unité esthétique mamelouke, et non
quer que certaines fables comportent plusieurs nécessairement la proximité entre les textes22. Les
peintures, très rapprochées les unes des autres sources littéraires comme les Maqāmāt ou Kalīla
tandis que d’autres épisodes ne sont pas illustrés wa-Dimna ont permis la création de ces images
du tout. Ces choix peuvent aussi répondre à une archétypales, ce qui suggère une certaine récipro-
commande particulière, le mécène ayant précisé cité ou une interconnexion entre les sources d’ins-
quelles anecdotes il souhaitait ou non voir illus- piration de ces différents ateliers.
trées. Le choix des peintures dans la copie du Une des grandes différences de notre manuscrit
Sulwān est aussi en partie motivé par les images avec deux copies tardives du Sulwān est l’impor-
archétypales déjà employées depuis le xiiie siècle, tance donnée à la représentation d’animaux, équi-
comme en atteste également le tableau de com- valente en nombre aux peintures représentant des
paraison avec les copies du Kalīla wa-Dimna des figures humaines23. Cet élément participe d’un
xiiie et xive siècles (Tableau 7.2). Ces illustrations
standardisées répondent à un contexte artistique 22 Melikian-Chirvani voit les illustrations archétypales
plus global dans lequel certaines scènes ou cer- comme des variations d’une scène à l’autre et d’un
taines thématiques étaient illustrées de la même texte à l’autre. Il souligne le fait que l’artiste ne s’inté-
façon, quel que soit le support. resse pas à la narration ou à l’action et les images n’ap-
portent aucun élément supplémentaire au texte. Les
Les images archétypales sont ici adaptées au images sont ainsi figées sur une représentation prévi-
récit et permettent de reconnaître certaines thé- sible et l’artiste préfère utiliser ces archétypes préexis-
matiques représentées ailleurs. Elles sont des tants plutôt que de créer de nouvelles compositions
indications visuelles d’éléments spatio-temporels (Sulwān, 150). Pourtant, si l’on compare les composi-
tions de manuscrits illustrés à l’époque mamelouke, le
récurrents, liés au contexte artistique de l’époque pourcentage de ces illustrations archétypales n’est pas
mamelouke. Mais la création de cette copie nécessairement révélateur du manque d’intérêt par
unique, dont on ne connaît pas d’autres exemples l’artiste pour l’illustration du texte en particulier. Dans
à l’époque mamelouke, questionne l’idée d’une la peinture persane, certaines scènes sont représentées
de la même manière (scène de couronnement, scène
unité esthétique inhérente à cette époque, telle
de chasse ou de bataille), mais certaines compositions
que conçue par Haldane21. Il existe certes une mettent en évidence la relation étroite entre texte et
certaine unité dans les éléments artistiques au image, et donc entre le texte et le peintre.
sein de groupes de copies d’un même texte (par 23 Le manuscrit Ms. arabe 528 de la Bibliothèque de
ex les Maqāmāt ou Kalīla wa-Dimna), mais sans l’Escorial est une copie maghrébine probablement
du xviiie siècle, qui comprend quarante-sept pein-
autre exemple de copies du Sulwān, on ne peut tures dont six seulement sur le thème animalier. La
qu’en modérer la portée ou le cadre. Le tableau 7.2 copie de la Bibliothèque nationale de France (Arabe
3511) provient sans doute d’Égypte ou de Syrie otto-
manes, du xviie ou du xviiie siècle. Elle est inache-
au Musée d’art islamique de Doha (MS.683.2001). vée et le manuscrit mériterait une investigation plus
Dans le cas de cette copie, il est curieux d’observer la approfondie car certains feuillets semblent être à la
non-concordance entre peintures et texte. Le peintre mauvaise place. Elle présente cinquante-huit pein-
ne lit peut-être pas l’arabe et copie des illustrations tures ou emplacements de peintures dont neuf sur le
dont il semble prendre modèle dans d’autres copies thème animalier. Ces copies tardives rendent compte
illustrées contemporaines du même texte. Seulement, de l’influence de l’iconographie des manuscrits occi-
il insère ces schémas de manière aléatoire dans les dentaux et se concentrent sur la représentation de la
espaces laissés vides par le peintre et ne suit pas le figure humaine, mais la seule constante entre les trois
texte. manuscrits est le nombre de peintures et l’illustration
21 Haldane, Mamluk painting, 13-31. des mêmes éléments de narration issus de la fable des
212 Chekhab-Abudaya
Cycle des peintures du Numéro de folio Numéro de page Résumé du texte accompagnant
ms.27.1999 ou d’inventaire dans le fac-similé l’illustration
et la publication
de 1985
Fable des deux renards : f.12v FS.26 Ẓālim, en quête d’un nouveau terrier, est
scène symétrique où les accueilli par un autre renard, Mufawwaḍ
deux renards se font face (« celui qui s’en remet à Dieu ») qui décide
de l’héberger et de l’aider à se débarrasser
du serpent.
Fable des deux renards : f.15v FS.32 Ẓālim, qui convoite le repaire de
paysage similaire au cadre Mufawwaḍ, se cache en son absence der-
des deux peintures précé- rière un tas de bois. Mufawwaḍ allume le
dentes mais sans aucun bois dans le terrier puis part à la recherche
élément de figuration de son ami, mais le brûle vivant sans le
savoir.
Shāpūr II à un banquet f.31 FS.63 Shāpūr II (r. 309-379) explore l’Empire
à la cour de l’empereur byzantin déguisé et accompagné d’un de
byzantin à Constantinople ses vizirs, lui-même vêtu d’une tenue de
moine chrétien. Il est invité à un banquet
au cours duquel il est démasqué et empri-
sonné.
Shāpūr II emprisonné f.32v FS.66 Démasqué au cours du banquet, Shāpūr est
emprisonné dans un taureau en cuir. L’em-
pereur décide d’envoyer son armée attaquer
l’Iran en emportant son prisonnier.
Perles de sagesse 213
Cycle des peintures du Numéro de folio Numéro de page Résumé du texte accompagnant
ms.27.1999 ou d’inventaire dans le fac-similé l’illustration
et la publication
de 1985
ʿAyn Ahlīh attaché à un f.37 FS.75 Le vizir est emmené en tant que médecin
pilier sous la garde de la avec l’armée et le patriarche de l’Église
vieille femme à la main byzantine. Il lui raconte l’histoire de ʿAyn
coupée Ahlīh, métaphore de l’histoire de Shāpūr.
ʿAyn Ahlīh (« l’œil de son peuple », Shāpūr),
marié à Sayyidat al-Nār (« la maîtresse du
feu », la Perse), rencontre et tombe amou-
reux de Sayyidat al-Dhahab (« la femme en
or », l’Empire byzantin), mariée à al-Dhiʾb
(« le loup », l’empereur byzantin). Capturé
par al-Dhiʾb, ʿAyn Ahlīh est gardé par une
vieille femme à la main coupée (le vizir) à
qui il relate ses mésaventures.
Fable du cheval et du F1954.1 (Freer FS.78 La vieille femme relate à ʿAyn Ahlīh la fable
sanglier : rencontre entre Gallery of Art) enchâssée du cheval et du sanglier. Un che-
le cheval entravé et le val s’enfuit loin de son maître et se blesse
sanglier en tentant de se libérer de son harnache-
ment. Il demande à un sanglier de l’aider.
Fable du cheval et du f.39v FS.82 Le sanglier lui fait d’abord admettre son
sanglier : cheval libéré de erreur d’avoir voulu se libérer de son joug
son harnachement sans aucune expérience dans la nature.
Ayant reconnu cela, le cheval finit par être
détaché.
La vieille femme parle à f.41 FS.84 La vieille femme relate à ʿAyn Ahlīh sa
ʿAyn Ahlīh, toujours atta- propre histoire : vendue au souverain
ché à un poteau comme esclave, il lui a fait couper le nez,
retirer un œil, puis couper la main gauche
à chacune de ses tentatives de s’échapper.
ʿAyn Ahlīh lui propose plus tard dans le
récit de s’enfuir ensemble.
Fable de la gazelle et de f.42v FS.86 La vieille femme relate ensuite à ʿAyn Ahlīh
l’antilope : la gazelle et une autre fable, celle de la gazelle et de l’an-
l’antilope en conversation tilope, qui deviennent amies pendant leur
pendant leur captivité captivité et décident de s’enfuir ensemble.
214 Chekhab-Abudaya
Cycle des peintures du Numéro de folio Numéro de page Résumé du texte accompagnant
ms.27.1999 ou d’inventaire dans le fac-similé l’illustration
et la publication
de 1985
Fable de la gazelle et f.44 FS.90 L’antilope est prise dans le piège d’un chas-
de l’antilope : la gazelle seur et renvoyée en captivité, tandis que la
fuit l’antilope, se sentant gazelle est également capturée. Cette der-
trahie. nière reproche à l’antilope de l’avoir aban-
donnée dans le désert. L’antilope explique
qu’elle a aussi été prise au piège et elles
se réconcilient. Ce récit est une nouvelle
métaphore mettant en scène Shāpūr (la
gazelle) et le vizir (l’antilope) qui fait certes
preuve de loyauté à l’égard de son souverain
mais est impuissant à le libérer.
Khusraw I et le souverain f.56 FS.115 Le prince indien donne audience à l’am-
indien : le prince indien bassadeur de Khusraw. Il le fait surveiller
sur son trône donnant par un espion, que l’ambassadeur réussit à
audience à l’ambassadeur tromper en jouant à l’idiot.
de Khusraw
Khusraw I et le souverain f.58 FS.119 À son retour, l’ambassadeur décrit à Khus-
indien : le prince indien et raw l’Inde comme un pays faible. Le souve-
ses conseillers rain met en place un plan visant à inciter
le peuple indien à se rebeller, ce afin de
conquérir plus aisément le pays. Le prince
indien convoque alors un conseil composé
de quatre de ses vizirs et du responsable du
culte zoroastrien.
Fable des deux rats et de f.61v FS.126 Durant l’audience, le quatrième vizir prend
la gerboise : les deux rats à la parole et relate au souverain la fable des
la recherche d’un abri deux rats et de la gerboise : un couple de
rats explore un wādī, à la recherche d’un
endroit où vivre. Ils décident de se creuser
un abri au pied d’un monticule entouré
d’eau.
Fable des deux rats et de f.62 FS.127 Les deux rats rencontrent une gerboise qui
la gerboise : les rats (dont sort de son terrier pour s’installer sur un
un seul est représenté) monticule situé en hauteur de la zone inon-
rencontrent la gerboise. dable. Elle les met en garde et leur conseille
de se déplacer ailleurs.
Perles de sagesse 215
Cycle des peintures du Numéro de folio Numéro de page Résumé du texte accompagnant
ms.27.1999 ou d’inventaire dans le fac-similé l’illustration
et la publication
de 1985
Fable des deux rats et de f.63v FS.130 Le rat mâle revient un jour dans son
la gerboise : la gerboise repaire et le trouve inondé et sa famille
fait face au rat mâle du noyée. N’ayant pas écouté les conseils de la
haut du monticule. gerboise, il la voit assise en haut du
monticule, saine et sauve.
Fable de l’ours et du singe : f.76 FS.155 Bahrām Gūr est confié par son père, le roi
les singes approchent sassanide Yazdagard, à la cour de Nuʿman le
l’ours. Lakhmide. Son tuteur arabe Ḥils lui raconte
plusieurs fables dont celle de l’ours et des
singes. Plusieurs singes s’approchent d’un
ours qui feint d’être mort.
Fable du moine et du f.77 FS.157 Le récit est interrompu par une fable
voleur : le voleur tombe enchâssée, celle du moine et du voleur :
dans le piège du moine. un moine s’enrichit avec l’aumône de ses
fidèles. Un voleur entre un jour dans sa
chapelle et le moine feint de ne pas l’avoir
vu. Le voleur se précipite vers lui et tombe
dans un piège à travers une trappe dissimu-
lée dans le sol. Il est exécuté le lendemain
par les autorités, et le moine continue à
dissimuler son vrai dessein.
Fable de l’ours et du singe : f.78 FS.159 Le récit revient à la fable de l’ours et du
l’ours attrape le singe. singe : l’ours réussit à attraper un des singes
et l’assujettit.
Fable de l’ours et du singe : F1954.2 FS.166 Le singe feint de perdre la vue et prétend
le singe consulte le singe (Freer Gallery ne plus pouvoir cueillir de fruits pour l’ours.
médecin pendant que of Art) L’ours le laisse consulter un singe médecin,
l’ours attend au pied de qui lui recommande de prétendre que sa
l’arbre. vue est meilleure pendant la nuit, pour
réussir à s’échapper.
Fable de l’ours et du singe : LNS 104 MS (Dar FS.172 Pressentant la ruse, l’ours surveille le singe
l’ours tue le singe en le al-Athar al Isla- et finit par le tuer lorsqu’il tente de s’échap-
faisant chuter de l’arbre. miyyah) per.
216 Chekhab-Abudaya
Cycle des peintures du Numéro de folio Numéro de page Résumé du texte accompagnant
ms.27.1999 ou d’inventaire dans le fac-similé l’illustration
et la publication
de 1985
Cycle des peintures du Numéro de folio Numéro de page Résumé du texte accompagnant
ms.27.1999 ou d’inventaire dans le fac-similé l’illustration
et la publication
de 1985
La fable des deux f.104 FS.215 Bābak tente de convaincre son père, le roi
éléphants : les deux Ardashīr, premier roi sassanide de Perse,
éléphants se rencontrent. de l’intérêt d’être ascète. Il lui raconte la
fable des deux éléphants : les gardes d’un
roi chassent un éléphant sauvage et tentent
de l’apprivoiser sans succès. L’éléphant déjà
dompté conseille à son congénère d’ac-
cepter d’être apprivoisé car il bénéficiera
de nombreux privilèges (nourriture, soins,
riches tissus).
La fable des deux f.104v FS.216 L’éléphant sauvage accepte le compromis
éléphants : l’éléphant et informe l’autre éléphant des contraintes :
sauvage dans une parade porter de lourdes charges, prendre part
de guerre à des guerres et perdre sa liberté. Tous
deux feignent alors d’être malades et leurs
maîtres les relâchent dans la nature.
Le moine et le berger f.107 FS.221 Ardashīr tente de persuader son fils de
en conversation l’intérêt des richesses. Bābak relate une
dernière anecdote à son père pour le
dissuader de leur intérêt : un berger se
plaint à un moine chrétien du travail dont
il a la charge. Le moine lui raconte alors
l’histoire d’un moine qui découvre un
jour un ermitage délabré et décide d’en
travailler la terre. À mesure que ses revenus
augmentent, de plus en plus de moines le
rejoignent. Il distribue les richesses aux
pauvres et les autres moines protestent. Il
ordonne pendant la nuit de faire couper les
arbres et les vignes et les moines furieux le
chassent de l’ermitage. Le moine chrétien
du début du récit conseille alors au berger
de ne plus donner à son troupeau autant
de soin.
218 Chekhab-Abudaya
tableau 7.2 Table de comparaison entre MS.27.1999 (MIA) et des copies des xiiie et xive siècles de Kalīla wa-Dimna
Éléments Pococke Cod. Arab. Ms Ar. 578 T-S Ar.51.60 Ms 3655 Arabe 3465 Arabe 3467 King Fayṣal
principaux 400 (1354, 616 (vers (xive siècle, (peut-être (Fin xiiie (xiiie siècle, (xive siècle, 2536 (1346, 65
de similitude 78 illustra- 1340-50, 120 illustra- xive siècle, siècle, 98 illustra- 47 illustra- illustrations),
tions), OBL 73 illustra- tions), Parker 1 folio), 113 illustra- tions dont tions), BnF KFC
tions), BSB Library Cambridge tions), BRR 8 tardives),
University BnF
Library
Structures f.13v, archi- ff.22r, 23v, ff.27v, 30v, ff.9v, 16r, 26v, ff.4v, 6r, 7r,
architectu- tecture et 28r, 28v, 29r, 31v, 36v, 39r, 28v, 51r, 76r, 30v, 77v, et
rales pour lampe ; f.16r, 32v, 33r, 36r, 40r, 80v, 90v, 88r et 93r 117v, struc-
les scènes lampe et 36v, 37r, 37v, 102v, 104r, tures archi-
d’intérieur trône ; f.21v, 49r, 67r, 70v, 115v, 130v, tecturales
architecture 79v, 102r, 131v, 134v, 137r, similaires
102v, 128v, architecture mais simpli-
structures et/ou drapés fiées ; ff.14v et
architectu- (ou tentures) 115v, tenture
rales f.121r, archi-
tecture simi-
laire au f.77r
du Sulwān
Souverain f.10r, mêmes ff.19r, 24v, 50v, f.3r, trône f.29v, trône ff.10r, 12v, ff.3v, 93r, 94v, ff.52v, 117v,
sur son trône couleurs, pos- 104r, 107v, 111v, uniquement simplifié par 14v, 20v, 19r, 98v, 106v, 110r 142r, trône
donnant ture, motifs trône (souverain rapport au 23v, et 133r, mais repré-
audience vestimen- de face et Sulwān structures sentation
taires. Pas simplifié) ; architectu- simplifiée
de couronne ff.4r, 9r, 9v, rales, trône
dans le 11v, 28v, 38r, et souve-
MS.27.1999 et 55r, 102v, 105v, rain côté
ici souverain 106v, 107r, gauche mais
à droite ; 108r, 114r traitement
ff.48v, 123v, différent
125v et 129v,
trône et
souverain
côté gauche
(faisant
face à son
audience) ;
ff.133r, 143r,
146r, trône
à gauche,
lampe mais
pas l’architec-
ture (inversé
mais similaire
au f.136v)
Perles de sagesse 219
tableau 7.2 Table de comparaison entre MS.27.1999 (MIA) et des copies des xiiie et xive siècles de Kalīla wa-Dimna (cont.)
Éléments Pococke Cod. Arab. Ms Ar. 578 T-S Ar.51.60 Ms 3655 Arabe 3465 Arabe 3467 King Fayṣal
principaux 400 (1354, 616 (vers (xive siècle, (peut-être (Fin xiiie (xiiie siècle, (xive siècle, 2536 (1346, 65
de similitude 78 illustra- 1340-50, 120 illustra- xive siècle, siècle, 98 illustra- 47 illustra- illustrations),
tions), OBL 73 illustra- tions), Parker 1 folio), 113 illustra- tions dont tions), BnF KFC
tions), BSB Library Cambridge tions), BRR 8 tardives),
University BnF
Library
Exemples f.40v, Kalīla f.44v, Kalīla et f.31v, Kalīla et f.44r, 49v f.48r, Kalīla et ff.21r, 41v ff.24v, 55r,
de symétrie et Dimna ; Dimna Dimna Dimna Kalīla et
animalière f.58v, lion et f73v, traite- Dimna
éléphant ment des f.127v, lion
lions très et chacal
différent mais (posture
symétrie également)
Représenta- ff.41v, 45v, ff.45r, 47v, 68r, ff.30r, 32v, 36r, Lion et sa ff.23v, 39v, ff.3r, 22r, 46r, ff.43r, posture,
tion du lion 70r, 138v, 142v, 72r, 118r, 127v 47v, 49r, 57v, mère, struc- 53v, 54v, 114v, 100r mais style un
postures (et 58v, 59v, 62v, ture similaire postures, peu différent
symétrie) 65v, 99v, 123r, au f.90r du symétrie, type
124v, 125r, 127r, Sulwān de représen-
134v tation
Représen- ff.58v, 99r. ; f.89r f.84v, élé- f.6r f.71r
tation de f.148r roi sur phant assez
l’éléphant l’éléphant similaire mais
(structure simplifié
similaire f.106r, élé-
au f.104v phant blanc
du Sulwān mais modèles
et éléphant assez simi-
dompté simi- laire
laire)
Représenta- ff.65v, 111v, f.64r et 98v f.54v f.42v et 89r f.101v, 113v,
tion du singe 114v, et 144v 146r, singes
similaires
mais simpli-
fiés
Représenta- ff.82v, 92v, ff.77v, 79v et ff.69v, 71r, 72r, f.119v ff.63r, 64r et v, f.107v
tion du rat 132v 84r 76r, 77r, 115v 109r
Représen- f.92v f.92v f.153r, simi-
tation de laire au f.44r
l’antilope/ du Sulwān
gazelle
Représen- f.61v, canards f.52r, 60v f.55r f.13v, phœnix f.84v f.83v, cor-
tation des représentés et quatre beaux, repré-
oiseaux à partir de oiseaux simi- sentation
modèles laires mais archétypale
d’oiseaux simplifiés d’oiseaux
(similaire mais simpli-
au f.90r du fiés
MS.27.1999)
220 Chekhab-Abudaya
tableau 7.2 Table de comparaison entre MS.27.1999 (MIA) et des copies des xiiie et xive siècles de Kalīla wa-Dimna (cont.)
Éléments Pococke Cod. Arab. Ms Ar. 578 T-S Ar.51.60 Ms 3655 Arabe 3465 Arabe 3467 King Fayṣal
principaux 400 (1354, 616 (vers (xive siècle, (peut-être (Fin xiiie (xiiie siècle, (xive siècle, 2536 (1346, 65
de similitude 78 illustra- 1340-50, 120 illustra- xive siècle, siècle, 98 illustra- 47 illustra- illustrations),
tions), OBL 73 illustra- tions), Parker 1 folio), 113 illustra- tions dont tions), BnF KFC
tions), BSB Library Cambridge tions), BRR 8 tardives),
University BnF
Library
Éléments ff.20v et 107r, f.27v, rochers ff.43r, 51r, 55v, ff.39v, 44r, f.6r, plante ff.41v, herbe ; f.24v, monti-
naturels : par- rochers bleus bleus avec 69v, 92r, 97r, 52r, 69v, très stylisée. 71r, mon- cule, eau
terre d’herbe, stylisés, herbe composition monticule monticule en ticule ; 89r
monticule, en tressage ; identique à demi-cercle herbe et eau
eau, rochers 40v, herbe ; Pococke f.20v
53r, repré- et Sulwān
sentation de ff.54v, 83r, 98v,
l’eau et plante monticule
au centre + scènes
comme sépa- d’extérieur
ration ; f.58v, avec parterre
végétation ; d’herbe en
f.90v, monti- tressage
cule similaire
au f.62r du
Sulwān
Ciel en arc de ff.31r, 99r, ff.36r, 89r f.71r, ciel
cercle demi-cercle inversé
dans l’autre comme dans
sens le Sulwān
art proprement mamelouk, et est probablement Les folios (90 et 42v) représentant les deux lions
issu aussi de la tradition fatimide, en incluant se faisant face et celle de la gazelle et de l’anti-
également des modèles sassanides préexistants24. lope, font écho à de nombreuses copies du Kalīla
wa-Dimna et à d’autres illustrations de bestiaires
deux renards. Voir également comme comparaison notamment25. Ces paires d’animaux sont adaptées
un manuscrit hispanique du xve siècle issu de la tra-
dition castillane du Kalīla wa-Dimna dont la structure
et le traitement des figures sont similaires à ce qui des animaux de profil avec une plante stylisée en fond
est présenté dans les copies du xviie siècle qui nous dans la céramique lustrée). Mais ces similitudes se
intéressent ici. Hugo O. Bizzarri, « L’image enchâs- rencontrent en dehors du milieu arabe et nous avons
sée dans le Calila e Dimna », in D’Orient en Occident. déjà souligné le même phénomène en milieu persan
Les recueils de fables enchâssées avant les Mille et une (par exemple les similitudes entre vaisselles haft rang
Nuits de Galland (Barlaam et Josaphat, Calila et Dimna, et copies du Shāhnāma). Voir aussi Melikian-Chirvani,
Disciplina clericalis, Roman des Sept Sages), éd. Marion Sulwān, 129-168 ; Hillenbrand, « Mamlūk and Īlkhānid
Uhlig et Yasmina Foehr-Janssens (Turnhout : Brepols, Bestiaries », fig. 7 ; Contadini, A World of Beasts: A
2014), 309-323. Pour une analyse du cycle de peintures Thirteenth-Century Illustrated Arabic Book on Animals
de la copie de l’Escorial, voir Rachel Arié, Miniatures (the Kitāb Naʿt al-Ḥayawān) in the Ibn Bakhtīshūʿ
hispano-musulmanes (Leiden : Brill, 1969). Tradition (Leiden : Brill, 2011), 85-89.
24 Notons la proximité avec les peintures de céramiques 25 Contadini a fait plusieurs rapprochements entre la
lustrées et les manuscrits à l’époque fatimide (notam- copie syrienne du Kitāb Manāfiʿ al-Ḥayawān datée
ment les scènes de chasse ou encore la représentation de 1354 conservée à l’Escorial (Ar. 898) et la copie du
Perles de sagesse 221
à des textes et à des éléments de narration diffé- également qu’aucune des peintures n’est annon-
rents (Figs. 7.4, 7. 5 et 7.6). Mais dans le cas des cée par un titre comme c’est généralement le cas
lions symbolisant la lutte du pouvoir sur la Perse dans d’autres manuscrits et dans les copies tar-
dans le Sulwān, c’est la couronne représentée au dives du même texte. Cela suppose de suivre le
centre de la composition qui vient apporter un déroulement des illustrations au fil de la lecture
élément propre à ce texte en particulier et donc du texte.
en dehors de l’image archétypale. On remarque L’effet de symétrie est utilisé de manière systé-
matique et la répartition des différentes couleurs
Sulwān, présentée comme probablement réalisée crée un rythme dans les compositions. La repré-
dans le même atelier et à la même époque. Ces copies sentation est plane, sans échelle, même si l’on
du xive siècle montrent clairement la présence de remarque une ébauche de perspective dans la
modèles déjà préexistants dans des manuscrits du
siècle précédent comme la copie du Naʿt al-Ḥayawān
représentation des composants de certaines illus-
de la British Library (Or. 2784). Contadini rapproche trations sur plusieurs plans. Comme le souligne
notamment les folios représentant la gazelle et l’an- A.S. Melikian-Chirvani, toutes les compositions
tilope qui sont effectivement presqu’identiques, tant ont comme point commun de présenter un fond
par leur composition que par le traitement du sujet. La
monochrome or, un élément dont l’abstraction et
seule différence est la présence d’éléments de végéta-
tion dans le Sulwān. Contadini, A World of Beasts, 84, la neutralité rendent la lecture de l’image et de la
116-117. Melikian-Chirvani rapproche également cette scène plus simple26.
composition avec la mosaïque de Khirbat al-Mafjar
et donne d’autres exemples de similitudes (Sulwān,
109-121 et 129-168). Voir également Sofie Walzer,
« An Illustrated Leaf from a Lost Mamlūk Kalīlah 26 Ces critères de l’art mamelouk sont détaillés par
Wa-Dimnah Manuscript », Ars Orientalis, n° 2 (1957) : Haldane, Mamluk painting. L’effet de perspective
503-505. se retrouve aux folios 32v, 39v, 56r, 94r, 97r, 100v,
222 Chekhab-Abudaya
figure 7.6 La couronne de Perse est placée entre Khusraw l’Usurpateur et Bahrām Gūr, f.90r (MS.27.1999, MIA)
Perles de sagesse 223
figure 7.7 Le cheval en joug libéré par le sanglier, f.39v (MS.27.1999, MIA)
Dans les scènes d’extérieur, souvent animalières, la mécénat princier27 qui n’apparaît pas dans les
végétation semble déconnectée de l’histoire repré- deux copies tardives. Le frontispice avec le souve-
sentée. Elle apparaît souvent disproportionnée et rain sur son trône attenant à une scène de la vie
semble pousser de nulle part lorsqu’elle n’est pas de cour (montrant des chasseurs ou des musiciens
reliée à un parterre d’herbe tressée (Figs. 7.7, 7.8 par exemple) est courant dans les manuscrits du
et 7.9). Cette représentation symbolique peut être xive siècle28. Les personnages de l’ensemble des
reliée directement à celle de l’architecture qui est peintures présentent des têtes dont la taille est
aussi schématique et dont les éléments sommaires disproportionnée par rapport au corps et sont
laissent penser aux décors du théâtre d’ombres. représentés de face ou de trois-quarts. Le visage
Ces éléments agissent comme une synecdoque rond avec une barbe (noire pour la plupart), les
visuelle qui suggère l’espace plus vaste de la nar- quelques nimbes, les yeux en amande, sont autant
ration. La simplicité de la composition permet au d’éléments qui rappellent la tradition mésopota-
fil de la lecture d’aller à l’essentiel dans la visualisa- mienne du xiiie siècle29. Le traitement des cos-
tion des protagonistes. Ainsi, le trône ou encore la tumes et des turbans est standardisé, mais les
bipartition de la représentation du souverain fai-
sant face à sa cour définissent l’intrigue du palais.
27 Contadini, Arab Painting, 21.
Les scènes comprenant des personnages 28 Contadini, A World of Beasts, 56.
mettent en évidence une continuité du style déve- 29 Hillenbrand, « Mamlūk and Īlkhānid Bestiaries », 221-
loppé au xiiie siècle. On retrouve notamment 234. Le nimbe n’est pas un élément religieux mais est
l’adoption du frontispice honorant l’idée d’un un élément emprunté au contexte romain et byzantin,
et permet de distinguer le souverain des autres person-
nages et de mettre en valeur la tête des personnages
104v et F1954.1 de la Freer Gallery of Art. Voir aussi lorsqu’il est utilisé pour tous. Contadini, A World of
Melikian-Chirvani, Sulwān, 104-108. Beasts, 129-130.
224 Chekhab-Abudaya
figure 7.8 L’ours feint d’être mort tandis que les singes s’approchent de lui, f.76r (MS.27.1999, MIA)
figure 7.9 Le voleur tombe à travers la trappe dans la chapelle, f.77r (MS.27.1999, MIA)
Perles de sagesse 225
figure 7.10 ʿAyn Ahlīh attaché en compagnie de la vieille femme mutilée, f.37r (MS.27.1999, MIA)
visages sont personnalisés de façon à ce que l’on mésopotamiennes antérieures, reprises dans la
reconnaisse certains personnages d’une même production ilkhanide puis timouride (Fig. 7. 14).
fable, illustrés sur plusieurs images30 (Fig. 7.10). La répétition de schémas et de motifs permet
Le souverain apparaît surélevé sur un trône ou ainsi de visualiser plus rapidement les person-
sur une estrade, sur le côté gauche des composi- nages et la scène représentée. Le peintre allie
tions et portant toujours le même type de robe ainsi un répertoire simple avec un certain degré
(Figs. 7.11 et 7.12). Le folio 31 représentant Shāpūr de stylisation notamment en ce qui concerne la
à la cour de l’empereur byzantin est l’une des représentation de la végétation (plantes, arbres,
traces de l’influence d’archétypes sur l’artiste : la parterre d’herbe) et d’autres éléments naturels
composition, le traitement des personnages, les (ciel, rochers, rivière, monticules). Ce cadre natu-
éléments du mobilier, la structure architectu- rel, réalisé uniquement pour signifier une scène
rale se rapprochent de la copie des Maqāmāt de en extérieur, contraste complètement avec les
Vienne datée de 1334, et en particulier les folios protagonistes de chaque scène qui semblent pour
42v et 95v31 (Fig. 7.13). D’autres éléments, comme certains flotter en arrière-plan. Le rôle de ce cadre
la tente du f.97r, laissent penser aux traditions est d’agencer la symétrie et la linéarité de la com-
position. Tous les éléments de ce cadre dérivent
30 Haldane décrit dans son ouvrage les illustrations carac- certainement de modèles chinois et l’on retrouve
térisées par leur immobilité. Pourtant le Sulwān met en des éléments presque identiques dans la copie du
scène certains animaux en mouvement et une certaine Manāfiʿ al-Ḥayawān de l’Escorial. L’introduction
gestuelle entre les personnages même s’ils demeurent de la fleur de lotus notamment, parmi les éléments
assez statiques dans l’ensemble. Haldane, Mamluk
painting, 13.
de végétation stylisée dans plusieurs des peintures
31 Atıl, Renaissance in Islam, 258-259 ; Melikian-Chirvani, du Sulwān, laisse penser à une influence de la
Sulwān. peinture arabe du xiiie siècle et de l’art ilkhanide
226 Chekhab-Abudaya
figure 7.11 Le prince indien donne audience à l’ambassadeur de Khusraw, f.56r (MS.27.1999, MIA)
figure 7.14 Nuʿmān b. Imruʾ al-Qays en compagnie de ʿAdī b. Zayd à la shaqīqa, f.97r (MS.27.1999, MIA)
228 Chekhab-Abudaya
sur la production mamelouke32. Le fond doré des dans le frontispice est aussi un élément qui peut
peintures permet en revanche de faire un lien avec être comparé avec l’art byzantin, où l’adoption de
l’art byzantin et l’art médiéval d’Occident dont la ce type de représentation est très courante pour
production de manuscrits rend compte à cette illustrer le mécène du manuscrit34.
époque d’une utilisation presque systématique A.S. Melikian-Chirvani propose d’attribuer ce
de ce procédé33. La mise en exergue du souverain manuscrit à Damas, en raison de sa similitude avec
la copie du Manāfiʿ al-Ḥayawān de l’Escorial35.
32 Haldane, Mamluk painting, 6, 13. Voir également David Cette hypothèse semble renforcée par l’enregistre-
James, Qurʾans of the Mamluks (Londres : Alexandria ment de ce manuscrit en tant que waqf au début
Press, 1988) ; J. Michael Rogers, « Evidence for Mamluk- du xviie siècle à Alep, comme signalé dans l’in-
Mongol Relations 1260-1360 », in Colloque international
sur l’histoire du Caire, 27 mars-5 avril 1969, éd. André troduction. Notons cependant que l’attribution à
Raymond, J. Michael Rogers et Magdi Wahba (Le Caire : l’Égypte ou la Syrie reste délicate en raison du peu
Ministry of Culture of the Arab Republic of Egypt, de copies datées et localisées, et ne permet pas
1972), 385-404. d’être affirmatif dans le cas spécifique du Sulwān.
33 Anna Contadini, « Kitāb naʿt al-ḥayawān (Book on
the Characteristics of Animals, BL Or.2784) and the
Les proximités avec d’autres copies mameloukes
“Ibn Bakhtīshūʿ” Illustrated Bestiaries » (thèse de doc- déjà citées permettent justement de montrer à
torat, University of London, School of Oriental and quel point les centres de production égyptien et
African Studies, 1992), 288. Il existe clairement une syrien sont très proches les uns des autres. Les
tendance dans la structure des compositions à fond artistes pouvaient sans doute avoir voyagé d’un
or dans l’art médiéval d’Occident dès le début du xiiie
siècle. Ces manuscrits, des bibles et autres ouvrages atelier à l’autre et utiliser des modèles de représen-
religieux illustrés, sont eux-mêmes inspirés de l’art tation très similaires, comme c’est le cas plus tard
byzantin. Nous observons notamment des similitudes au xve siècle36.
jusqu’à la seconde moitié du xvie siècle, en prenant
comme exemple la célèbre Bible de la Hague, copie
royale complétée entre 1371 et 1372 (conservée à la travers la tradition de manuscrits syriaques du nord de
Bibliothèque nationale des Pays-Bas, M10 B23). Voir la Mésopotamie. À propos de l’influence notamment
Charles Sterling, La Peinture médiévale à Paris, 1300- des ateliers syriens par l’art byzantin (notamment
1500, (Paris : Bibliothèque des Arts, 1987), 20, 187-192, l’usage de l’or mais également la tenture triangulaire),
194, 197, fig. 107-108. De manière tout à fait similaire, voir Haldane, Mamluk painting 5. Voir également sur
Gülru Necipoğlu avait souligné le lien entre influence les éléments chinois, en particulier la fleur de lotus,
chinoise et influence européenne dans les albums Diez Melikian-Chirvani, Sulwān, 99-104 et 109-121.
à la même époque et les avait donc replacés dans un 34 Eva Hoffman, « The Author Portrait in Thirteenth-
contexte plus global : Gülru Necipoğlu, « Persianate Century Arabic Manuscripts: a New Islamic Context for
Images between Europe and China: The “Frankish a Late-Antique Tradition », Muqarnas 10 (1993) : 6-20.
Manner” in the Diez and Topkapı Albums c. 1350-1450 », 35 Melikian-Chirvani, Sulwān, 76. Chirvani place le Sulwān
in The Diez Albums Contents and Context, éd. Julia dans un groupe de quatre manuscrits mamelouks qui
Gonnella, Friederike Weis et Christoph Rauch (Leiden : partagent un ensemble de caractéristiques avec le
Brill, 2017), 531-592. Voir aussi la comparaison entre manuscrit du Manāfiʿ qui aurait été produit à Damas.
certaines copies des Maqāmāt et de Kalīla wa-Dimna Leur style se caractérise par l’introduction d’éléments
avec des manuscrits byzantins et coptes dans Hugo issus des répertoires iranien et chinois. Voir également
Buchthal, « “Hellenistic” Miniatures in Early Islamic à ce sujet Contadini, « The Kitāb Manāfiʿ al-Ḥayawān in
Manuscripts », Ars Islamica 7, n° 2 (1940) : 125-133. Non the Escorial Library », Islamic Art 3 (1989) : 53.
seulement certains personnages mais aussi les repré- 36 Atıl, « Mamluk Painting in the Late Fifteenth Century »,
sentations architecturales révèlent des similitudes. Muqarnas 2 (1984) : 159-171. La page de droite du fron-
L’auteur fait référence à de possibles contacts avec des tispice avait été identifiée comme une copie égyp-
cours locales seljouks en Asie Mineure mais aussi en tienne lors de son passage chez Drouot en 1973 puis
Syrie, notamment au travers des similitudes architec- dans la publication d’une partie des collections de l’Aga
turales. Il note également l’utilisation d’un répertoire Khan. La page de la Dar al-Athar al Islamiyyah a aussi
chrétien probablement déjà connu à cette époque à été cataloguée comme étant égyptienne, syrienne voire
Perles de sagesse 229
mamelouke40. On peut tout à fait imaginer une être accompagné de destructions mais aussi d’une
destruction massive de copies illustrées dont seu- fuite des artistes syriens vers l’Égypte ou d’autres
lement peu d’exemples ont survécu de nos jours. contrées. La copie mamelouke du Sulwān est ainsi
L’utilisation d’un fond or dans chacune des pein- un manuscrit qui à lui seul illustre l’histoire litté-
tures montre le caractère exceptionnel de cette raire et artistique complexe de la Méditerranée
copie, lorsque l’on sait que l’or s’est justement médiévale, avec un texte rédigé en Sicile, copié
raréfié aux xive et xve siècles dans ces régions41. et illustré au Moyen-Orient puis au Maghreb par
Ce manuscrit rend aussi compte sans doute de la la suite.
présence d’ouvrages byzantins, par exemple des
ouvrages de liturgie illustrés qui faisaient sans
doute partie des collections de manuscrits de cer- Bibliographie
taines bibliothèques de la région42.
La rareté des copies illustrées à la fin du xive Arié, Rachel. Miniatures hispano-musulmanes. Leiden :
siècle a laissé penser à une éventuelle interruption Brill, 1969.
de la peinture arabe quelques décennies après Atıl, Esin. « Mamluk Painting in the Late Fifteenth
l’exécution du Sulwān, phénomène qui pourrait Century ». Muqarnas 2 (1984) : 159-171.
s’expliquer et être nuancé par le contexte histo- Atıl, Esin. Islamic Art and Patronage: Treasures
rique, culturel et social. Il existe en réalité des from Kuwait. New York : Rizzoli International
copies de manuscrits plus tardives, datant du xve Publications, 1990.
siècle et du xvie siècle qui montrent, comme le Atıl, Esin. Renaissance in Islam: Art of the Mamluks.
souligne E. Atıl, l’arrivée au Caire d’artistes entraî- Washington D.C. : Smithsonian Institution Press,
nés dans les cours turkmènes, marquant ainsi une 1981.
transition vers le mécénat de manuscrits illus- Bellino, Francesca. « Animal Fables in Sulwān al-Muṭāʿ
trés à l’époque ottomane43. Un autre événement by Ibn Ẓafar al-Siqillī ». In Islamic Sicily : Philological
d’importance qui pourrait expliquer le manque and Literary Essays. Édité par Mirella Cassarino, 103-
d’exemples est le sac d’Alep et de Damas par les 22. Rome: Istituto per l’Oriente, 2015.
armées de Timur vers 1400-140144. Ce sac a pu Bizzarri, Hugo O. « L’image enchâssée dans le Calila
e Dimna ». In D’Orient en Occident, Les recueils
de fables enchâssées avant les Mille et une Nuits
40 Atıl, Renaissance in Islam, 250-251. Melikian-Chirvani
offre une comparaison avec deux copies de Maqāmāt de Galland (Barlaam et Josaphat, Calila et Dimna,
(celle datée de 1334 et conservée à Vienne – Öster- Disciplina clericalis, Roman des Sept Sages). Édité par
reichische Nationalbibliothek A.F. 9 – et celle de 1337 Marion Uhlig et Yasmina Foehr-Janssens, 309-323.
conservée à Oxford – Bodleian Library, Marsh 458) et la Turnhout : Brepols, 2014.
copie du Manāfiʿ al-Ḥayawān datée de 1354 conservée à
Blois, François de. « The Pancatantra from India to the
l’Escorial – Ar. 898. Melikian-Chirvani, Sulwān, 76-99.
Cette comparaison vient corroborer son identification West – and Back, a Mirror for Princes from India ». In
de la copie du Sulwān comme une copie syrienne qu’il A Mirror for Princes from India: Illustrated Versions of
situe entre 1330 et 1340. the Kalilah wa Dimnah, Anvar-i Suhayli, Iyar-i Danish,
41 Hillenbrand, « Mamlūk and Īlkhānid Bestiaries », 220.
and Humayun Nameh. Édité par Ernst J. Grube, 10-15.
42 Pour certains rapprochements entre les ouvrages
syriaques et la peinture du xiiie siècle, voir Contadini, Bombay : Marg Publications, 1991.
A World of Beasts, 68-71.
43 Atıl, Renaissance in Islam, 250. Elle mentionne une
copie du Kashf al-Asrār réalisée en Égypte au début du par la peste noire vers 1348-1349. Cette épidémie
xve siècle dans une lignée similaire. dévaste aussi l’Europe au même moment et décime de
44 Haldane mentionne le fait que de nombreux artisans nombreuses populations. Peter M. Holt, « Mamlūks »,
syriens suivent Timur à Samarcande. Haldane, Mamluk in Encyclopaedia of Islam, Second Edition, éd. Peri
painting, 1. La Syrie et l’Égypte sont également touchées Bearman et al. 6 (Leiden : Brill, 1960-2007), 321-31.
Perles de sagesse 231
Bolshukov, Oleg G. « The St. Petersburg Manuscript of George, Alain F. « The illustrations of the Maqāmāt and
the Maqāmāt by al-Ḥarīrī and its Place in the History the Shadow Play ». Art History 35, n° 2 (2012), 1-42.
of Arab Painting ». Manuscripta Orientalia 3, n° 4 Grabar, Oleg. « Pictures or Commentaries: The Illus-
(1997) : 59-67. trations of the Maqamat of al-Hariri ». In Islamic
Bosworth, Clifford E. « Naṣīḥat al-mulūk », In Encyclo- Visual Culture, 1100-1800, Constructing the Study of
paedia of Islam, Second Edition. Édité par Peri Bear- Islamic Art. Édité par Oleg Grabar, vol. 2, 188-205.
man et al. vol. 7, 984-988. Leiden : Brill, 1960-2007. Hampshire : Ashgate Publishing Limited, 2006.
Brockelmann, Carl. « Kalīla wa Dimna ». In Encyclopae- Grabar, Oleg. The Illustrations of the Maqamat. Chicago :
dia of Islam, Second Edition. Édité par Peri Bearman University of Chicago Press, 1984.
et al. vol. 4, 503-506. Leiden : Brill, 1960-2007. Gray, Basil. « Fourteenth-Century Illustrations of the
Buchthal, Hugo. « “Hellenistic” Miniatures in Early “Kalilah and Dimnah” », Ars Islamica 7, n° 2 (1940) :
Islamic Manuscripts ». Ars Islamica 7, n° 2 (1940) : 134-140.
125-133. Haldane, Duncan. Mamluk Painting. Warminster : Aris
Contadini, Anna, éd. Arab Painting, Text and Image in and Phillips, 1978.
Illustrated Arabic Manuscripts. Leiden : Brill, 2007. Hillenbrand, Robert. « Mamlūk and Īlkhānid Bestiaries:
Contadini, Anna. « The Kitāb Manāfiʿ al-Ḥayawān in Convention and Experiment ». In Studies in the
the Escorial Library ». Islamic Art 3 (1989) : 33-57. Islamic Arts of the Book. Édité par Robert Hillenbrand,
Contadini, Anna. « Kitāb naʿt al-ḥayawān (Book on the 211-289. Londres : The Pindar Press, 2012.
Characteristics of Animals, BL Or.2784) and the “Ibn Hoffman, Eva. « The Author Portrait in Thirteenth-
Bakhtīshūʿ” Illustrated Bestiaries ». Thèse de doc- Century Arabic Manuscripts: a New Islamic Context
torat, University of London, School of Oriental and for a Late-Antique Tradition ». Muqarnas 10 (1993) :
African Studies, 1992. 6-20.
Contadini, Anna. A World of Beasts: A Thirteenth-Century Holt, Peter M. « Mamlūks ». In Encyclopaedia of Islam,
Illustrated Arabic Book on Animals (the Kitāb Naʿt Second Edition. Édité par Peri Bearman et al. vol. 6,
al-Ḥayawān) in the Ibn Bakhtīshūʿ Tradition. Leiden : 321-331. Leiden : Brill, 1960-2007.
Brill, 2012. Holter, Knut. « Die islamischen Miniaturhandschriften
Daiber, Verena et Junod, Benoît, éd. Treasures of the Aga vor 1350 ». Zentralblatt für Bibliothekswesen, n°54
Khan Museum, Masterpieces of Islamic Art. Berlin : (1937) : 1-34.
Nicolai Verlag, 2010. Ibn al-Muqaffaʿ, ʿAbd Allāh. Kalilāh and Dimnah.
Dakhlia, Jocelyne. « Les miroirs des princes islamiques : Traduit par Jallad. Le Caire : Būlāq, 2002.
une modernité sourde ? ». Annales. Histoire, Sciences Ibn Ẓafar al-Siqilli, Muhammad. The Just Prince, A
sociales 57, n° 5 (2002) : 1191-1206. Manual of Leadership, Including an Authoritative
Dekmejian, R. Hrair et Thabit, Adel Fathy. « Machiavelli’s English Translation of the Sulwan al-Mutaʿ fi ʿUdwan
Arab Precursor: Ibn Ẓafar al-Siqillī ». British Journal al-Atbaʿ (Consolation for the Ruler During the Hostility
of Middle Eastern Studies 27, n° 2 (2000) : 125-137. of Subjects). Traduit par Joseph A. Kechichian et
Falk, Toby. Treasures of Islam. Londres: Philipp Wilson R. Hrair Dekmejian. Londres : Saqi Books, 2003.
Publishers, 1985. Importante collection de miniatures orientales. Vente,
Fullerton, Arlene. « A Brief History of the Jasim Paris, Hôtel Drouot, 26 octobre 1973. Paris : Anne-
al-Homaizi Collection ». In Kuwait: Arts and Marie Kevorkian, 1973.
Architecture: A Collection of Essays. Édité par Arlene James, David. « Space-Forms in the Work of the
Fullerton et Géza Fehérvári, 51-65. Koweït : Oriental Baghdad ‘Maqamat’ Illustrators: 1225-1258 A.D. ».
Press, 1995. Bulletin of the School of Oriental and African Studies
George, Alain F. « Orality, Writing and the Image in the 37, n° 2 (1974) : 305-320.
Maqamat: Arabic Illustrated Books in Context ». Art James, David. Qurʾans of the Mamluks. Londres :
History 35, n° 1 (2012) : 10-37. Alexandria Press, 1988.
232 Chekhab-Abudaya
Kafescioǧlu, Çiǧdem. « In the Image of Rum: Ottoman Contents and Context. Édité par Julia Gonnella,
Patronage in Sixteenth-Century Aleppo and Friederike Weis et Christoph Rauch, 531-592. Leiden :
Damascus ». Muqarnas 16 (1999) : 77-97. Brill, 2017.
Marlow, Louise. « Advice and Advice Literature ». In Rizzitano, Umberto. « Ibn Ẓafar ». in Encyclopaedia of
Encyclopaedia of Islam, Third Edition. Edité par Kate Islam, Second Edition, éd. Peri Bearman et al., vol. 3,
Fleet et al. Leiden : Brill, 1960-2007. 970. Leiden : Brill, 1960-2007.
Marlow, Louise. « Surveying Recent Literature on the Rogers, J. Michael. « Evidence for Mamluk-Mongol
Arabic and Persian Mirrors for Princes Genre ». Relations 1260-1360 ». In Colloque international sur
History Compass 7 (2009) : 523-538. l’histoire du Caire, 27 mars-5 avril 1969. Édité par
Masters, Bruce. « Janbulad Ali Pasha ». In Encyclopedia André Raymond, J. Michael Rogers et Magdi Wahba,
of the Ottoman Empire. Edité par Gábor Ágoston et 385-404. Le Caire : Ministry of Culture of the Arab
Bruce Masters, 295-296. New York : Facts on File, Republic of Egypt, 1972.
2009. Sterling, Charles. La peinture médiévale à Paris, 1300-
Mayer, Leo A. « A Hitherto Unknown Damascene 1500, tome 1. Paris : Bibliothèque des Arts, 1987.
Artist ». Ars Islamica 9 (1942) : 168. Swynford Lambton, Ann K. « Islamic Mirrors for
Melikian-Chirvani, Assadullah S. Sulwān al-Muṭāʿ fī Prince ». In La Persia nel medioevo: Atti del Convegno
ʿudwān al-atbāʿ, An Art Historical Essay. Koweït : TRI internazionale, 419-442. Rome : Accademia Nazionale
Publishing, 1985. Dei Lincei, 1971.
Necipoğlu, Gülru. « Persianate Images Between Europe Walzer, Sofie. « An Illustrated Leaf from a Lost Mamlūk
and China: The “Frankish Manner” in the Diez and Kalīlah Wa-Dimnah Manuscript ». Ars Orientalis 2
Topkapı Albums c. 1350-1450 ». In The Diez albums (1957) : 503-505.
8
Les tribulations d’un manuscrit à peintures, ou comment fabriquer
une illustration à partir de plusieurs modèles
Swietochowski décrivent les pages du manuscrit chasseur (f.101). Il s’agit du quatorzième chapitre
en des termes excessivement élogieux – « the de la table des matières qui en annonce quinze,
most exciting gifts recently offered to the Islamic le dernier chapitre, « L’ermite et son invité », ne
Department and among the most important figure pas parmi les pages conservées au MMA.
Indian manuscripts to have appeared in many D’autre part, la fable du lion et du chacal, qui
years » – et l’attribuent sans ambages à l’Inde des devrait correspondre au chapitre 11, n’apparaît
sultanats3. Trois ans plus tard, en 1985, lors de pas non plus parmi elles, alors que celle du roi de
l’exposition India qui a lieu au musée, un double l’Inde et des brahmanes qui, au contraire, n’est pas
folio et un feuillet isolé sont présentés au public citée dans le sommaire, est bien présente dans ce
parmi les œuvres pré-mogholes4. Durant la décen- manuscrit (f.94v).
nie suivante toutefois, s’appuyant sur certains élé- Les feuillets mesurent 30,5 × 22,4 cm, mais ils
ments iconographiques, Ernst Grube rattache le ont été largement massicotés (par exemple, le
manuscrit à l’Empire ottoman, voire à la Turquie f.12r). Les numéros de folios sont disposés avec
même5. Selon lui, comme pour le manuscrit de la soin dans la marge supérieure gauche des rectos.
collection S. dont il est question ailleurs dans cet Dans la marge inférieure gauche des versos, des
ouvrage6, il pourrait s’agir d’une version unique réclames sont visibles, ce qui facilite la reconsti-
du texte arabe à destination du marché turc7. Cela tution du manuscrit et permet de s’assurer que la
reste à prouver dans le cas du codex de New York foliotation est conforme à la disposition originale
dont le texte n’a pas encore été examiné par des du texte8. Une série de signes variés (losanges,
spécialistes de la littérature arabe. Comme on le fleurettes, cercles ponctués d’un point …) ponctue
verra dans les pages qui suivent, son attribution le manuscrit à plusieurs endroits (ff.12v, 16v, 18r,
est particulièrement ardue. 18v, 19v, 20a, 22v, 23r, 25v, 26r, etc. , Fig. 8.3, Fig. 8.4),
L’ouvrage est démembré et dépourvu de reliure. traces d’un probable remontage du codex à une
C’est une copie composée de quatre-vingt-seize époque ancienne. D’autre part, plusieurs filigranes
folios : quelques feuillets sont manquants, et ont été relevés, mais aucun n’a encore été identifié
le texte s’achève sur la fable de la lionne et du (ff.9, 16, 19, 27, 56, 65, 82).
L’écriture, un petit naskh soigné et de bonne
tenue, se déploie sur un papier teint dans un jaune
3 Stuart C. Welch et Marie L. Swietochowski, « And This
ocre. Une particularité graphique est notable :
is the Picture of the Leopard Relating to the Army What
the Lion Has Ordered », The Metropolitan Museum of Art, le tracé des kāf qui, dans les premiers folios du
Notable Acquisitions 1981-1982 (New York : Metropolitan manuscrit, revêt parfois à l’initiale, en médiane ou
Museum of Art, 1982), 15‑16 ; voir aussi « Art of Asia en finale, la forme singulière d’un kāf minuscule
Acquired by North American Museums 1982 », Archives of porté sur une base de kāf privé de sa barre supé-
Asian Art 36 (1983) : fig. 43, 105.
4 Stuart C. Welch, India: Art and Culture, 1300-1900 (New
rieure (Fig. 8.1).
York : Metropolitan Museum of Art, 1985), no 81, 137-139. Cette forme apparaissant en plusieurs lieux et
5 Ernst J. Grube, « Prolegomena for a Corpus Publication of époques, il est malheureusement impossible d’en
Illustrated Kalilah wa Dimna Manuscripts », Islamic Art 4 tirer quelque conclusion.
(1990-91), 301-481 ; voir aussi Ernst J. Grube, « Some Obser-
L’ouvrage est doté d’une page de titre (Fig. 8.2).
vations Concerning the Ottoman Illustrated Manuscripts
of the Kalilah wa Dimnah: Ali Çelebi’s Humayun-Name », Celui-ci, disposé en haut du folio d’ouverture, a sans
in Milletlerarası Türk Sanatları Kongresi (Ankara : Kültür doute été copié lors de la première phase d’élabo-
Bakanlığı, 1995), 195-205. ration du manuscrit. Une bordure constituée d’un
6 Voir dans ce même ouvrage, le chapitre 11 : El Khiari et al.,
« Entre deux mondes : une anthologie illustrée inédite ins-
pirée de Kalīla wa-Dimna ». 8 Contrairement à ce qu’écrit Grube dans « Some Observa-
7 Grube, « Some Observations Concerning the Ottoman tions Concerning the Ottoman Illustrated Manuscripts of
Illustrated Manuscripts of the Kalilah wa Dimnah », 196. the Kalilah wa Dimnah », 197.
Les tribulations d’un manuscrit à peintures 235
Ce manuscrit est par bien d’autres aspects un jalons qui marquent l’histoire du manuscrit et,
curieux objet, qui n’a pas d’équivalent ailleurs. Il sans doute, celle des interventions de différents
est surprenant que les rares publications qui le peintres, ou groupes de peintres, qui se sont suc-
citent ne relèvent pas cette évidente singularité cédé à chaque stade de la réalisation. Les peintures,
qui réside tout à la fois dans le mélange original à plusieurs endroits, en particulier au niveau des
des styles et des répertoires iconographiques, visages, affichent des traces de repeints, de même,
et dans le fait que la qualité des représentations nombre d’entre elles sont restées inachevées (Fig.
puisse varier énormément d’une page à l’autre. 8.7) et enfin, la disparité des traitements – pose
Certaines peintures montrent un tracé assuré, de la couleur, dilution de la matière, construction
des compositions complexes et inattendues, sans de l’image –, dans l’élaboration des compositions,
aucun doute un travail de peintre professionnel tend à confirmer l’hypothèse d’une construction
(Fig. 8.3). picturale en plusieurs phases distinctes, qui se
D’autres, d’une indéniable naïveté, sont à attri- sont sans doute déroulées sur une longue période.
buer à un amateur, et qui plus est un amateur bien
peu talentueux : la ligne est grossière, les figures
parfois ne sont pas les bonnes (par exemple, un 4 Les sources iconographiques
animal est représenté à la place d’un autre), elles
sont mal achevées ou inachevées, la peinture origi- Trois copies en langue arabe de Kalīla wa-Dimna,
nale n’est de toute évidence pas connue (Fig. 8.4). datant de la période mamelouke, présentent des
Les sources d’inspiration sont multiples ; on en peintures comparables à celles du manuscrit de
distingue au moins cinq dans le Kalīla wa-Dimna du New York, ainsi qu’un feuillet isolé : Arabe 3467 à la
MMA. Premièrement, l’iconographie mamelouke Bibliothèque nationale de France9 (Fig. 8.9), Cod
que l’on peut situer aux alentours de 1350 (Fig. 8.8, Arab 616 à la Staatsbibliothek de Munich10 (Figs.
par exemple). Deuxièmement, la manière safavide
du xviie siècle, ce que je désignerai comme le 9 François de Blois, Burzōy’s Voyage to India and the Ori-
« style safavide 1 » (Fig. 8.3), pour le distinguer du gin of the Book of Kalīlah Wa Dimnah (Londres : Royal
« style safavide 2 », présent dans d’autres peintures Asiatic Society, 1990), 70 ; Richard Ettinghausen, La
du manuscrit, postérieures (Fig. 8.7). Dans ces der- peinture arabe (Genève : Skira, 1962), 153-155 ; Grube,
« Prolegomena », no 3, 374 ; Duncan Haldane, Mamluk
nières, qui constituent le troisième groupe, les per-
Painting (Warminster : Aris & Phillips, 1978), 95-99 ; Ber-
sonnages sont petits, les scènes moins denses, le nard O’Kane, Early Persian Painting: Kalila and Dimna
tracé du pinceau est aussi moins soigné. On peut Manuscripts of the Late Fourteenth Century (Londres :
réunir, dans un quatrième ensemble, les peintures I.B. Tauris, 2003), no 5, 39 et appendix 5, 222 ; Sofie
affichant une iconographie attestant de liens avec Walzer, « An Illustrated Leaf from a Lost Mamluk Kali-
lah wa Dimnah manuscript », Ars Orientalis 2 (1957) :
le monde ottoman : personnages à larges turbans, 503-505 ; Sofie Walzer, « The Mamlûk Illuminated
architectures, éléments végétaux, comme les Manuscript of Kalîla wa-Dimna », in Aus der Welt der
tulipes et les cyprès (Figs. 8.3 et 8.5). islamischen Kunst: Festschrift für Ernst Kühnel zum 75
Enfin sont réunis dans un cinquième groupe Geburstag am 26-10-1957, éd. Richard Ettinghausen
(Berlin : Gebr. Mann, 1959), 195-206.
des éléments plus disparates, figures humaines et
10 Walzer, « An Illustrated Leaf from a Lost Mamluk Kalilah
décors, qui demeurent plus difficiles à identifier wa Dimnah Manuscript » ; Walzer, « The Mamlûk
d’un point de vue stylistique (Fig. 8.6). Illuminated Manuscript of Kalîla wa-Dimna » ; Hal-
Le curieux assemblage que constituent ces dane, Mamluk Painting, 81-82 ; Hans-Caspar Bothmer,
styles au sein des illustrations du manuscrit, où Kalila und Dimna: Ibn al-Muqaffaʿ’s Fabelbuch in
einer mittelalterlichen Bilderhandschrift : Cod. arab
ils apparaissent souvent simultanément, rend 616 der Bayerischen Staatsbibliothek München (Wies-
l’œuvre particulièrement difficile à analyser. Cette baden : Reichert, 1981) ; Karl Dachs, Das Buch im
diversité formelle est l’empreinte de différents Orient : Handschriften und kostbare Drucke aus zwei
238 de la Perrière
8.10 et 8.21), Pococke 400 à la Bodleian Library Bien qu’incomplets, les codex semblent tous avoir
à Oxford11 (Fig. 8.11) et le feuillet T-S Ar.51.60 à été largement illustrés : soixante-dix-huit pein-
la bibliothèque universitaire de Cambridge, au tures pour les manuscrits d’Oxford et de New York
Royaume Uni12. et soixante-treize pour celui de Munich. Le rythme
Seul le colophon du manuscrit Pococke 400, semble à peu près le même : environ une image
daté de 755H/1354, a été conservé. Les deux autres toutes les trois pages14.
codex et le feuillet isolé ont sans doute été copiés C’est incontestablement sur la copie d’Oxford,
et peints dans le Proche-Orient mamelouk à la fin Pococke 400, que le Kalīla wa-Dimna du MMA a pris
du xiiie siècle et dans le courant du xive siècle. Si modèle. Pour commencer, dans ces deux manus-
on s’en tient aux critères stylistiques, il est envisa- crits, la table des chapitres initiale est organisée à
geable que le manuscrit de Paris soit le plus ancien l’identique bien que, à partir de la moitié du texte
et le manuscrit de Munich le plus tardif. La page de environ (au folio 117 pour le manuscrit d’Oxford et
Cambridge est plus difficile à dater précisément ; au folio 89 pour le manuscrit de New York), l’agen-
toutefois, le style de l’écriture et celui des pein- cement des fables ne corresponde plus à l’ordre
tures m’amènent à proposer la fin du xiiie siècle13. indiqué dans le sommaire. Les deux manuscrits
présentent alors, l’un et l’autre, des dévelop-
pements un peu différents (voir Tableau 8.1).
Jahrtausenden (Wiesbaden : Reichert, 1982), no 90,
157, 159 ; Grube, « Prolegomena », no 4, 375 ; O’Kane, L’examen des cycles illustratifs a permis de mettre
Early Persian Painting, no 4, 39 et appendix 4, 221 ; en avant la proximité des programmes iconogra-
Paul G. Dannhauer et Helga Rebhan, Wertvolle orienta- phiques du Kalīla wa-Dimna d’Oxford et de New
lische Handschriften und seltene Drucke der Bayerischen York. Cinquante-huit peintures illustrent le même
Staatsbibliothek (Münich : Bayerischen Staatsbiblio-
thek, 2004), no 13, 9. épisode des fables et le cycle des images varie rela-
11 Laurence Binyon, James V.S. Wilkinson et Basil tivement peu (voir Tableau 8.2). Enfin, et c’est là
Gray, Persian Miniature Painting (Londres : Oxford un point important, certaines lacunes au sein des
University Press, 1933), no 4, 24-25 ; Walzer, « An peintures du manuscrit du Metropolitan Museum
Illustrated Leaf from a Lost Mamluk Kalilah wa
Dimnah Manuscript » ; Walzer, « The Mamlûk Illu-
correspondent exactement à des parties détério-
minated Manuscript of Kalîla wa-Dimna » ; Hal- rées dans le modèle d’Oxford (Figs. 8.8 et 8.11) : il y
dane, Mamluk Painting, 49-51 ; Esin Atıl, Kalila wa a donc eu copie directe d’un manuscrit sur l’autre,
Dimna: Fables from a Fourteenth-Century Arabic à une période où le Pococke 400 avait déjà subi
Manuscript (Washington, D.C. : Smithsonian Insti-
quelques dégradations.
tution Press, 1981) ; de Blois, Burzōy’s Voyage to India,
66 ; Grube, « Prolegomena », no 5, 375-376 ; O’Kane, Dans le cadre du projet consacré aux manus-
Early Persian Painting, no 6, 39 et appendix 6, 222. crits arabes illustrés de Kalīla wa-Dimna, l’examen
12 Au sujet de ce feuillet retrouvé dans la Genizah du du Pococke 400 par Annie Vernay-Nouri a permis
Caire, voir Walzer, « An Illustrated Leaf from a Lost d’identifier des trous de poncifs sur vingt-sept feuil-
Mamluk Kalilah wa Dimnah Manuscript » ; Wal-
zer, « The Mamlûk Illuminated Manuscript of Kalîla
lets du manuscrit (Fig. 8.18). Vingt-quatre pein-
wa-Dimna » ; Ettinghausen, La peinture arabe, 153-155 ; tures du Kalīla wa-Dimna du MMA correspondent
Haldane, Mamluk Painting, 95-99 ; de Blois, Burzōy’s exactement à ces poncifs. Parmi elles, vingt-trois
Voyage to India and the Origin of the Book of Kalīlah sont inversées par rapport aux originaux, et on
Wa Dimnah, 70 ; Grube, « Prolegomena », no 3, 374 ;
peut se demander pour quelle raison (Fig. 8.8, par
O’Kane, Early Persian Painting, no 5, 39 et appendix
5, 222 ; Béatrice Gruendler, The Rise of the Arabic Book exemple). Afin d’éviter de détériorer l’illustration
(Cambridge, Mass. : Harvard University Press, 2020),
153-154. 1250 a été proposée. Voir Gruendler, The Rise of the Ara-
13 Sofie Walzer propose une datation au xive siècle et une bic Book, 153-154.
autre chronologie pour ce groupe de manuscrits : Wal- 14 Si les illustrations de tous ces ouvrages sont indéniable-
zer, « The Mamlûk Illuminated Manuscript of Kalîla ment très proches les unes des autres, leur place dans
wa-Dimna ». Plus récemment, sur la base de l’analyse la page, les dimensions des manuscrits et les réglures
du papier, une fourchette chronologique entre 950 et diffèrent considérablement.
Les tribulations d’un manuscrit à peintures 243
figure 8.9 Le moine, le voleur et l’épouse du cordonnier, f. 15v (BnF, Arabe 3467)
Les tribulations d’un manuscrit à peintures 245
figure 8.10 Le moine, le voleur et l’épouse du cordonnier, f. 48v (BSB, Cod. Arab. 616)
246 de la Perrière
figure 8.11 Le moine, le voleur et l’épouse du cordonnier, f. 46 (Bodl. Libr., Pococke 400)
Les tribulations d’un manuscrit à peintures 247
tableau 8.1 Liste des chapitres dans les manuscrits de Kalīla wa-Dimna d’Oxford (OBL, Pococke 400) et de New York
(MMA, 1981.373)
f.5v (fragment)
f.5v
f.10
f.6
f.12v
f.8v
f.13v
f.16
f.18v
f.20v f.11v
f.21v f.12
f.22 f.12v
f.24 f.13v
f.25 f.15
f.16v
f.30 f.18
f.31 f.18v
f.32 f.19v
f.33 f.20
f.20v
f.36v f.22v
f.23v
f.24
f.25
f.25v
f.26
f.26v (le menuisier et le singe)
f.40v (les deux chacals) f.27v (les deux chacals, dans le style « safavide »
f.41v f.28v
f.43v f.29v
f.45 f.30v
f.46 f.31v
f.46v f.32
f.48 f.33
f.50 (le crabe et la cigogne)
f.51v f.35v
f.53 f.36v
f.58v f.39v
Les tribulations d’un manuscrit à peintures 249
f.60 f.40v
f.61v f.41v
f.42
f.63 f.42v
f.64v f.43v
f.65v f.44v
f.67 f.45v
f.46
f.70 f.47v
f.73v f.50
f.75v f.51
f.77v f.52v
f.80v f.54v
f.82v f.56
f.86 f.58
f.90v f.61
f.92v f.62v
f.94 (hiboux et corbeaux)
f.94v f.64v
f.97v (lièvres)
f.98 f.66
f.99 f.67
f.102v f.69
f.104v f.70v
f.107 f.72
f.108v f.73
f.111v f.75
f.114 f.76
f.114v f.77
f.78
f.116v f.79v
f.117 f.80v
f.121
f.123v
f.125 (combat corbeaux) f.99v
f.125v
f.129
f.130
f.131v f.81v
f.133
f.133v f.83
f.134 f.83v
250 de la Perrière
originale, il est possible qu’on ait passé la ponce copie plus ancien, datant de la fabrication même
emplie de poudre charbonneuse ou colorée au dos du manuscrit Pococke 400. Nous connaissons
de la peinture, les formes transposées apparais- quelques exemples attestant de l’utilisation d’ai-
sant dès lors en miroir sur la copie. Cependant, guilles (ou stylets) pour reporter un dessin d’un
l’absence de traces poudreuses sur les peintures du manuscrit à l’autre, sans ponce. Autrement dit,
Pococke 400, comme aux versos, laisse penser que l’original est disposé sur une page blanche et les
ces dernières n’ont probablement jamais été utili- contours percés à l’aide d’une aiguille afin d’at-
sées directement comme poncifs, mais qu’un feuil- teindre la page vierge en laissant les traces de ces
let a servi d’étape intermédiaire avec la copie15. Si perforations.
c’est effectivement le cas, les marques de perfora- Seize autres illustrations du Kalīla wa-Dimna de
tions figurant sur les illustrations de Pococke 400 New York sont très proches du manuscrit d’Oxford
peuvent avoir deux origines : elles ont pu être sans qu’il soit possible d’établir assurément quel
produites, involontairement, par l’aiguille qui a mode de reproduction a été utilisé car les pein-
servi à percer les trous sur la feuille intermédiaire, tures correspondantes dans le manuscrit d’Oxford
elle-même directement appliquée sur l’original. n’ont pas été perforées. Quinze autres illustrations
À moins, et c’est une hypothèse intéressante sur encore s’approchent de celles d’Oxford mais la
laquelle je reviendrai par la suite, que ces marques copie n’est pas servile. Enfin, dans trois peintures,
correspondent elles-mêmes à un phénomène de un nombre plus restreint d’éléments, parfois un
seul, est issu du manuscrit Pococke 400.
15 Il serait néanmoins indispensable de vérifier cette
Parallèlement, certaines formes ont servi deux
observation réalisée à l’œil nu en effectuant des ana- fois dans le manuscrit de New York (Fig. 8.12, 8.13) :
lyses à la loupe binoculaire. par exemple, le chacal du f.64 de Pococke 400
Les tribulations d’un manuscrit à peintures 251
est utilisé à la fois au f.43v et au f.50 du Kalīla est particulièrement intéressant. Les petites per-
wa-Dimna de New York. forations qui suivent les contours de certaines
De même, le groupe composé de trois chacals et figures dans le manuscrit d’Oxford semblent cor-
d’un lièvre figurant au f.75v du manuscrit d’Oxford respondre en tout point aux formes de la copie de
réapparaît à la fois au f.51 et au f.87v de la copie du New York, attestant par conséquent du rapport
MMA (Figs. 8.14 et 8.15). entre les deux œuvres, mais certains détails n’en
Une même figure peut aussi revenir au sein demeurent pas moins énigmatiques. En effet, si les
d’une même illustration : par exemple, dans la trous dans Pococke 400 suivent assez fidèlement
fable du lièvre et du lion, les reflets ont sans doute les formes des personnages, des architectures ou
été élaborés à partir du même poncif que les ani- des végétaux, d’autres éléments, comme le haut
maux qu’ils représentent (Fig. 8.16). des turbans et les auréoles, ne sont pas concernés.
Dans une autre peinture, au f.101 du manuscrit A contrario, certaines perforations dessinent les
de New York où est illustrée la fable du roi de l’Inde contours d’éléments fantômes : de manière fort
et des brahmanes, certains groupes de person- surprenante, la forme d’une ceinture, percée mais
nages sont reportés à la fois dans le haut et dans le non peinte dans le Pococke 400 (f.25), réapparaît
bas de la peinture pour figurer une foule (Fig. 8.17). plus tard dans la copie du MMA (f.15) (Figs. 8.18 et
8.19).
S’il n’est en rien surprenant que le peintre de cet
5 Les techniques de reproduction dans ouvrage ait pu prendre quelques libertés par rap-
le manuscrit du Metropolitan Museum port à sa source d’inspiration, ce qui l’est davan-
of Art tage c’est que ces changements se manifestent
déjà dans les perforations visibles sur le manuscrit
Nos connaissances en matière d’emploi des pon- source. Deux cas de figure sont dès lors envisa-
cifs dans les manuscrits islamiques, et plus encore geables : premièrement, si les trous de Pococke 400
dans les manuscrits arabes, sont aujourd’hui limi- sont issus de la préparation d’un poncif destiné au
tées. Bien que l’importance des processus de repro- manuscrit du MMA, on peut penser que le prépa-
duction dans la création picturale fasse consensus rateur du poncif envisageait certaines transforma-
parmi les historiens de l’art, très peu de travaux tions dans la conception de la copie. Toutefois, le
ont été consacrés au sujet et, jusqu’à présent, peu deuxième cas de figure paraît plus convaincant :
de manifestations concrètes de l’utilisation d’ou- les perforations sur Pococke 400 sont antérieures
tils de reproduction dans les peintures de manus- à l’élaboration du manuscrit du MMA, elles sont
crits (comme les trous de poncifs ou les marques la marque d’une étape plus ancienne, celle où
de calques) n’ont été repérées puis publiées, par- fut lui-même fabriqué Pococke 400, à partir d’un
ticulièrement pour les peintures des manuscrits ouvrage plus ancien. En effet, Pococke 400 consti-
arabes16. tue avec deux autres Kalīla wa-Dimna d’époque
Dans un tel contexte, le lien direct existant mamelouke (BnF, Arabe 3467 et BSB, Cod.
entre les peintures d’Oxford et celles de New York Arab. 616) un groupe dont les peintures sont très
proches les unes des autres ; il en a été question
16 L’article le plus récent sur ce sujet est celui d’Yves Porter, plus tôt dans cet article. Il est envisageable qu’un
« Models, Sketches, and Pounced Drawings in the Diez quatrième manuscrit, aujourd’hui disparu, ait
Albums: First Steps in the Making of Illustrated Manus- servi de modèle à l’ensemble de ces ouvrages. Ceci
cripts », in The Diez Albums. Contexts and Contents, éd. pourrait expliquer à la fois la très forte cohérence
Julia Gonnella, Friederike Weis et Christoph Rauch (Lei-
den : Brill, 2017), 353-379 ; voir aussi David J. Roxburgh,
du groupe et les dissemblances existant entre les
« Persian Drawing, ca. 1400-1450: Materials and Crea- images. Il est possible que le feuillet isolé décou-
tive Procedure », Muqarnas 19 (2002) : 44-77. vert dans la Genizah du Caire et aujourd’hui
254 de la Perrière
figure 8.18 Détail du f. 25 observé à la feuille lumineuse (Bodl. Libr., Pococke 400)
conservé à l’université de Cambridge soit un le but de reproduire les images dans le manuscrit
vestige de ce manuscrit source, à moins qu’il ne du MMA, comment se fait-il que les illustrations de
s’agisse encore d’une autre copie de même inspi- ce dernier en suivent si précisément les contours ?
ration. Les petits trous dessinant certains contours Le cas de l’illustration du Kalīla wa-Dimna du
dans Pococke 400 sont peut-être le résultat du pro- MMA, où apparaît le dessin d’une ceinture qui
cessus de copie que j’ai évoqué précédemment : le ne figure dans l’illustration du Pococke 400 que
modèle est directement disposé sur la feuille desti- sous la forme de perforations, est symptoma-
née à recevoir la copie, ses contours sont percés au tique. Cela pourrait-il signifier que, pour procéder
stylet de manière à marquer également de petits à la reproduction des images qui l’intéressaient,
trous la feuille vierge destinée à la reproduction. le peintre du manuscrit de New York a utilisé les
Ces perforations permettront au peintre de la perforations déjà présentes dans la copie d’Oxford
copie de retracer aisément les formes de l’original. depuis son origine ? On peut aussi envisager que
Une fois recouverts de peintures, les trous sur la l’artiste du manuscrit de New York ait eu accès
copie sont quasi invisibles à l’œil nu. aux poncifs qui ont servi à élaborer les images de
Toutefois, un problème se pose ici : si les perfora- Pococke 400. Il faut néanmoins garder à l’esprit
tions de Pococke 400 n’ont pas été effectuées dans que certaines peintures du Kalīla wa-Dimna du
Les tribulations d’un manuscrit à peintures 259
MMA reproduisent certaines marques de détério- mamelouk, est relativement uniforme dans l’en-
rations tardives de Pococke 400. C’est donc que semble du manuscrit. Dans une majorité d’illus-
le peintre du manuscrit de New York a eu directe- trations, la peinture est particulièrement diluée,
ment accès à celui d’Oxford et qu’il l’a utilisé17. La comme en sous-couche. À différents endroits,
question reste donc en suspens. elle a également été appliquée par tamponne-
On sait peu de choses également sur l’utilisa- ments, peut-être avec un autre instrument que
tion d’autres modes de reproduction comme les le pinceau, à moins qu’il ne s’agisse d’un pinceau
pochoirs, les formes et les tampons auxquels les à large brosse. Ce mode d’application de la pein-
artistes du Kalīla wa-Dimna du MMA paraissent ture sur de larges surfaces n’est pas surprenant : il
avoir eu recours dans un deuxième stade de l’éla- concorde avec l’économie du manuscrit qui vise
boration du manuscrit (Fig. 8.4). Ces techniques tout en même temps facilité, rapidité d’exécution
de reproduction ne sont pas évidentes à repérer et et emploi parcimonieux des matériaux. Plusieurs
à identifier, et la superposition des procédés dans méthodes ont probablement été utilisées, selon
le manuscrit n’en simplifie pas l’analyse. Il semble les périodes de fabrication du manuscrit, ou de
toutefois qu’on ait utilisé à plusieurs endroits des manière concomitante. Dans un premier scenario,
formes pleines ou évidées, de type pochoirs. Des le dessin disposé sur un fond vierge est simple-
petites zones laissées vierges au sein des peintures ment colorié, de la façon ordinaire. Dans un deu-
témoignent de l’emploi de cette technique parti- xième cas de figure, la matière a été appliquée en
culière (Fig. 8.8). grands aplats, soit librement, soit en délimitant les
Comme il a été évoqué plus haut, les peintures espaces à l’aide de formes pleines ou évidées, des
du Kalīla wa-Dimna de New York affichent une pochoirs. Les contours et les détails ont ensuite
grande disparité stylistique, mais aussi des varia- été surlignés par un trait plus sombre, voire noir.
tions notables en termes de qualité. À certains Les dessins copiés à partir de poncifs pouvaient
endroits, il semble que le modèle originel ait été apparaître en transparence sous la couche de
mal interprété, signalant le fait que le peintre peinture très diluée, puis être repris à la gouache
intervenant à ce stade de l’élaboration du manus- ou à l’encre noire et, enfin, colorés d’une couche de
crit n’y a sans doute pas eu accès. Les traits des peinture supplémentaire. Les poncifs ont pu éga-
dessins reportés lui étant étrangers, il n’en a pas lement être disposés directement sur les surfaces
totalement saisi le sens. La maladresse des tracés peintes, puis le contour rehaussé d’une deuxième
prouve qu’il s’agissait d’un novice et, de plus, un couche de peinture. En ce qui concerne les motifs
novice peu adroit ; le codex avait alors certaine- ornementaux qui couvrent certains fonds, ils ont
ment quitté les murs de l’atelier ou de l’échoppe été peints tardivement ; beaucoup d’ailleurs sont
du professionnel chargé de l’illustrer. inachevés (Fig. 8.7).
L’analyse de la mise en couleur dans la copie Pour finir, notons que quelques images du Kalīla
du MMA ne va pas non plus sans poser problème. wa-Dimna du MMA semblent inspirées de celui de
À quel moment la couleur a-t-elle été appliquée ? la BSB (Figs. 8.20, 8.21).
Il est difficile de répondre à cette question. La Il s’agit de certaines des illustrations de style
palette, tout à fait différente de celle du modèle safavide 2, où les personnages sont de plus petite
taille, le dessin plus fin et plus dépouillé. Certaines
de ces illustrations figurent en marge du manus-
17 À six reprises, les images du Kalīla wa-Dimna de New
York ne correspondent pas à celles d’Oxford mais elles
crit, ce qui laisse penser qu’elles ont été ajoutées
s’en sont de toute évidence inspirées, ce qui étayerait lors d’une ultime étape dans l’histoire de la fabri-
encore cette hypothèse. cation du manuscrit (f.20v, 23v, 24).
Les tribulations d’un manuscrit à peintures 261
Les épisodes représentés ne sont d’ailleurs pas l’Université d’Oxford, voire, plus précisément,
illustrés dans le Kalīla wa-Dimna de l’OBL alors entre 1637 et 1640, alors que Edward Pococke est
qu’ils figurent bien dans la copie de la BSB. On peut installé à Istanbul. Car c’est là le dernier séjour
en déduire que, lors d’une dernière phase d’élabo- d’Edward Pococke en Orient et probablement
ration du manuscrit, l’artiste peignant à la manière aussi le dernier séjour du manuscrit dans la région.
safavide 2 a complété les illustrations du manuscrit En d’autres termes, il convient de se demander si
du MMA à partir d’une autre copie, a priori celle de la copie du MMA a pu être exécutée à la demande
la BSB. Il n’est néanmoins pas possible de s’assurer de Pococke lui-même ou tout au moins avec son
que le codex munichois est l’unique source de ces assentiment. Cette hypothèse est peu plausible :
peintures car la réplique n’est pas servile : il s’agit les notes de secondes mains figurant sur la page
d’une copie à main levée, certes fidèle à l’esprit du de titre du Kalīla wa-Dimna du MMA, datées 1026
modèle, mais de style différent. H/161718, amènent à penser que la première étape
de sa fabrication est antérieure. En effet, si ces
dates sont exactes, ce que l’examen du codex tend
6 Les différentes étapes de la création à laisser croire, la copie est forcément antérieure,
picturale dans le Kalīla wa-Dimna du tout comme la première étape d’élaboration des
Metropolitan Museum of Art illustrations.
Les peintures qui diffèrent de celles du manus-
Les deux manuscrits les plus proches de celui du crit Pococke 400 dans le Kalīla wa-Dimna de New
MMA sont donc, d’abord, celui de l’OBL, puis, dans York permettent d’envisager la suite de son par-
une moindre mesure, celui de la BSB. La copie cours, les phases postérieures de son élaboration.
conservée à la BnF, voisine elle aussi, ne présente Les illustrations de style safavide 1 et 2, nous auto-
pas autant d’analogies avec le manuscrit du MMA risent à dater la deuxième période d’élaboration
que les deux autres. des peintures assez tardivement, dans la deuxième
Le manuscrit Pococke 400, qui est lui-même moitié du xviie siècle, peut-être même un peu
sans doute une copie d’un codex plus ancien, plus tard. D’un point de vue stylistique, il est dif-
constitue la première étape dans l’élaboration ficile d’affiner cette datation, aucun autre manus-
du Kalīla wa-Dimna du MMA. Acquis par Edward crit, à ma connaissance du moins, ne présentant
Pococke (1604-1691) entre 1630 et 1635, à Alep, en une iconographie similaire19. De plus, la gamme
Syrie, le manuscrit est acheté en 1692 par l’Uni- chromatique se rapproche de celles de peintures
versité d’Oxford. L’ouvrage est donc incontesta- des manuscrits arabes des xviie et xviiie siècles.
blement en Europe à la fin du xviie siècle. Par On distingue au sein du programme pictural du
conséquent, les traces de perforations visibles sur Kalīla wa-Dimna de New York divers ensembles
ses peintures ne peuvent dater que d’une période iconographiques qui semblent administrés par
antérieure à 1692, ce qui fournit de facto un ter- une seule et même main, comme si un unique
minus ante quem pour les peintures de la copie peintre avait utilisé des modèles issus de traditions
du MMA. diverses, dans l’ensemble du manuscrit et, parfois,
La question qui se pose est de savoir si le
manuscrit Pococke 400 a servi de modèle à celui 18 Le sceau n’est en revanche pas daté et il n’a pu être
du MMA avant d’être acheté par Edward Pococke identifié.
19 Les représentations sur d’autres supports, particulière-
aux environs de 1630, ou bien si le manuscrit de ment la céramique, pourraient peut-être enrichir nos
New York a été copié entre cette date et celle où comparaisons. Jusqu’à présent, nous n’avons cepen-
Pococke 400 est entré dans les collections de dant encore rien trouvé d’analogue.
262 de la Perrière
au sein même de certaines illustrations. En effet, fabrication, ainsi qu’à la fin, de deux manuscrits qui
par-delà la multiplicité des sources iconogra- se sont trouvés au Proche-Orient à un moment de
phiques, l’empreinte stylistique commune renvoie leur histoire : en Syrie pour le premier et en Égypte
sans doute à la deuxième phase d’exécution de pour le second. Par conséquent, une provenance
l’ouvrage. La diversité des modèles, mameloukes, proche-orientale paraît plausible pour la copie
safavides ou ottomans, est transcendée par une du MMA20.
manière, un trait, partagés. Ainsi, certaines figures
du Pococke sont revisitées dans le style « safavide
1 » (les visages, par exemple). La pose de la couleur 7 Conclusion. Un loyal réapprentissage
et des motifs ornementaux qui agrémentent les des styles passés ou une forgerie à
fonds semblent aussi dater de la même période. destination du marché européen ?
C’est sans doute à ce moment-là qu’ont été réu-
tilisées certaines formes issues du Pococke 400 L’étude menée sur ce manuscrit permet de pro-
(peut-être à l’aide des poncifs originaux, ou bien poser un possible déroulement des étapes de sa
calquées sur les images déjà reproduites) pour fabrication. À ce stade de l’analyse, les grandes
compléter, enrichir quelques compositions. De lignes de la chronologie du codex se dessinent et
toute évidence, à ce stade, les peintres semblent son origine géographique paraît moins problé-
avoir élargi les possibilités que leur offraient les matique. Ces conclusions remettent en question
systèmes de reproduction : réutilisation d’une certaines attributions passées, l’Inde, surtout,
forme déjà apparue quelques pages plus tôt, dont les illustrations du manuscrit auraient indé-
emploi répété de certains éléments dans la même niablement porté la marque21. Les recherches
peinture.
La mise en couleurs n’a pas été achevée, le 20 Il est plus délicat de savoir quand les images les plus
manuscrit a donc été, une deuxième fois, aban- maladroites du manuscrit du MMA ont été exécutées :
donné en cours d’exécution. Il est repris, plus tard, sont-elles l’œuvre d’un tout jeune apprenti ou bien d’un
dans un troisième temps, par l’auteur des pein- amateur à qui l’on aurait confié l’exécution de quelques
pages ? Cela paraît bien peu probable, la plupart des
tures « de style safavide 2 », faciles à distinguer des
peintures du manuscrit étant par ailleurs tout à fait
précédentes. Bien que stylistiquement différentes, soignées. On peut supputer qu’il s’agit de l’intervention
certaines de ces peintures peuvent être rappro- d’un amateur, survenue très tardivement sur quelques
chées des illustrations du Kalīla wa-Dimna de la peintures restées inachevées.
BSB : mêmes compositions, même disposition des 21 « Most of the seventy-eight painted folios in the
manuscript these two miniatures come from, which
personnages, mêmes gestuelles. Le manuscrit de was painted in western India in the mid-sixteenth cen-
la BSB a été largement mutilé, de nombreux feuil- tury, were derived from designs traceable to Egypt, to
lets sont manquants, il est acéphale et ne com- Mamluk prototypes that may in turn have been based
porte pas, ou plus, de colophon. Comme l’indique on long-lost Indian originals. But the Heeramaneck
manuscript contains many delightful, sometimes
l’ex-libris, il a été rapporté du Caire par l’orienta-
comical, innovations. Of particular interest is the
liste Jean-Jacques Marcel (1776-1854). Le manuscrit strong Ottoman flavor, especially apparent here in the
de la BSB était donc encore en Orient au xixe siècle windblown tulips in Burzuya’s Indian Mission (…).
où il a donc peut-être, lui aussi, servi de modèle au Other miniatures include characteristically Ottoman
Kalīla wa-Dimna du MMA. S’il demeure impossible textile patterns and architectural elements, some-
times in conjunction with figures wearing unmistaka-
de déterminer plus exactement quand cela s’est bly Safavid turbans. Dominating this stylistic melting
produit, il est intéressant de noter que le manus- pot, however, is an overwhelming Indian spirit. The
crit du MMA s’est inspiré au premier stade de sa smoldering palette is rich in burnt orange and colors
Les tribulations d’un manuscrit à peintures 265
iconographiques que nous avons menées au sein – pose de décors de style ottoman
du programme Kalīla wa-Dimna pour tenter d’ali- Le manuscrit n’est toujours pas achevé.
menter cette piste indienne se sont avérées infruc-
tueuses : rien de tangible n’autorise à attribuer Phase d’élaboration III (« safavide 2 ») : Égypte ?
le manuscrit au Gujarat, ni à nulle autre région xviiie siècle
du sous-continent. L’attribution à l’Empire otto- Peinture des illustrations manquantes à main
man par Ernst Grube quelques années plus tard levée dans le style « safavide 2 » (à partir du
demeure la plus plausible22. Cependant, comme il manuscrit de la BSB ?)
a été démontré plus haut, il est plus vraisemblable
que la copie du Metropolitan Museum ait vu le Pseudo-phase IV
jour dans le Proche-Orient à l’époque ottomane, Interventions ultérieures, non professionnelles,
qu’en Turquie même. Les quatre périodes qui se sur le manuscrit
distinguent dans le processus d’élaboration de la
copie du MMA sont donc les suivantes : On connaît encore très mal l’histoire de la peinture
de manuscrits tardive, c’est pourquoi il demeure
Phase d’élaboration I (sans doute un premier ate- difficile de proposer des attributions en se fondant
lier, ville 1) : Syrie, Alep ? avant 1617 sur des critères stylistiques. Après le xviie siècle,
– copie du texte, le marché des manuscrits orientaux, et plus parti-
– premiers jadval-s (encadrements seulement culièrement celui des manuscrits persans, semble
dans les 12 premiers feuillets) ? s’être étiolé. À ce propos, s’appuyant sur les ren-
– esquisses des illustrations sur le modèle du seignements fournis par l’orientaliste Heinrich
manuscrit Pococke 400 inachevées et sans mise Friedrich von Diez (1751-1817), Christoph Rauch
en couleur écrit : « According to Diez, illustrated manuscripts
were rare and for that reason very expensive.
Phase d’élaboration II (deuxième atelier ou deu- He reported that he had met Persians who had
xième groupe d’artistes, ville 2) : Empire ottoman ? been sent to Constantinople in order to obtain
fin du xviie siècle ou xviiie siècle illuminated Persian manuscripts, because they
– mise en couleur : remplissage des fonds, utilisa- wanted to reanimate the art of illumination in
tion des pochoirs et des formes Iran itself23. » Diez se trouvant à Istanbul de 1784
– ajouts de peintures de style « safavide 1 » et de à 1790, ces informations concernent donc la fin du
style ottoman xviiie siècle. Néanmoins, depuis près de deux
– calque et réutilisation de certains dessins de la siècles, la demande des bibliophiles européens n’a
phase I (Pococke 400) fait que croître, les manuscrits enluminés, et plus
– repeint de certains visages des figures issues particulièrement ceux exécutés à la manière per-
du manuscrit Pococke 400 dans le style « safa- sane, deviennent une denrée rare et recherchée.
vide 1 » On ne tient sans doute pas assez compte de l’im-
– tracé des contours à main levée pact d’un tel phénomène sur le marché du livre
otherwise known to us only in paintings from Sirohi, 23 Christoph Rauch, « The Oriental Manuscripts and
in the southwestern corner of the present-day state Albums of Heinrich Friedrich von Diez and the
of Rajasthan, not far from Gujarat border. » Welch, Perception of Persian Painting in his Time », in The
India, 139. Diez Albums. Contexts and Contents, éd. Julia Gonnella,
22 Grube, « Some Observations Concerning the Ottoman Friederike Weis et Christoph Rauch (Leiden : Brill,
Illustrated Manuscripts of the Kalilah wa Dimnah ». 2017), 80.
266 de la Perrière
à peintures dans le monde islamique : dans cette George, Alain. « Orality, Writing and the Image in the
perspective, il est fort possible que des manuscrits Maqamat: Arabic Illustrated Books in Context ». Art
inachevés aient été complétés, « revisités », pour History 35, no 1 (2012) : 10-37.
répondre à la demande. Le Kalīla wa-Dimna du George, Alain. « The Illustrations of the Maqāmāt and
Metropolitan Museum of Art est peut-être une the Shadow Play ». Muqarnas 28 (2011) : 1-42.
forgerie élaborée à partir de plusieurs manuscrits, Grube, Ernst J. « Prolegomena for a Corpus Publication
revue et corrigée à plusieurs reprises et finalement of Illustrated Kalilah wa Dimna Manuscripts ».
destinée à attirer l’œil de potentiels acquéreurs Islamic Art 4 (1990-91), 301-481.
européens. Grube, Ernst J. « Some Observations Concerning the
Quelle que soit en définitive la destination Ottoman Illustrated Manuscripts of the Kalilah
du manuscrit, sa stratigraphie constitue un cas wa Dimnah: Ali Çelebi’s Humayun-Name ». In
d’école, la preuve d’une longue filiation iconogra- Milletlerarası Türk Sanatları Kongresi, vol. 2 : 195-205.
phique entre le xiiie siècle et le xviiie siècle, une Ankara : Kültür Bakanlığı, 1995.
hérédité délibérée. Gruendler, Beatrice. The Rise of the Arabic Book.
Cambridge, Mass. : Harvard University Press, 2020.
Haldane, Duncan. Mamluk Painting. Warminster : Aris
Bibliographie & Phillips, 1978.
O’Kane, Bernard. Early Persian Painting: Kalila and
« Art of Asia Acquired by North American Museums Dimna Manuscripts of the Late Fourteenth Century.
1982 ». Archives of Asian Art 36 (1983) : 105. Londres : I.B. Tauris, 2003.
Atıl, Esin. Kalila Wa Dimna: Fables from a Fourteenth- Porter, Yves. « Models, Sketches, and Pounced Drawings
Century Arabic Manuscript. Washington, D.C. : in the Diez Albums: First Steps in the Making of
Smithsonian Institution Press, 1981. Illustrated Manuscripts ». In The Diez Albums.
Binyon, Laurence, James V.S. Wilkinson et Basil Contexts and Contents. Édité par Julia Gonnella,
Gray. Persian Miniature Painting. Londres : Oxford Friederike Weis et Christoph Rauch, 353-79. Leiden :
University Press, 1933. Brill, 2017.
Blois, François de. Burzōy’s Voyage to India and the Rauch, Christoph. « The Oriental Manuscripts and
Origin of the Book of Kalīlah Wa Dimnah. Londres : Albums of Heinrich Friedrich von Diez and the
Royal Asiatic Society, 1990. Perception of Persian Painting in his Time ». In The
Bothmer, Hans-Caspar. Kalila und Dimna: Ibn Diez Albums. Contexts and Contents. Édité par Julia
al-Muqaffaʿ’s Fabelbuch in einer mittelalterlichen Bil- Gonnella, Friederike Weis et Christoph Rauch, 74-117.
derhandschrift: Cod. arab. 616 der Bayerischen Staats- Leiden : Brill, 2017.
bibliothek München. Wiesbaden : Reichert, 1981. Roxburgh, David J. « In Pursuit of Shadows: al-Hariri’s
Dachs, Karl. Das Buch im Orient: Handschriften und Maqāmāt ». Muqarnas 30 (2013) : 171-212.
kostbare Drucke aus zwei Jahrtausenden. Wiesbaden, Roxburgh, David J. « Persian Drawing, ca. 1400-1450:
Allemagne : Reichert, 1982. Materials and Creative Procedure ». Muqarnas 19
Dannhauer, Paul G. et Helga Rebhan. Wertvolle orien- (2002) : 44-77.
talische Handschriften und seltene Drucke der Walzer, Sofie. « An Illustrated Leaf from a Lost Mamluk
Bayerischen Staatsbibliothek. Munich : Bayerischen Kalilah wa Dimnah Manuscript ». Ars Orientalis 2
Staatsbibliothek, 2004. (1957) : 503-505.
Ettinghausen, Richard. La peinture arabe. Genève : Walzer, Sofie. « The Mamlûk Illuminated Manuscript of
Skira, 1962. Kalîla wa-Dimna ». In Aus der Welt der islamischen
Les tribulations d’un manuscrit à peintures 267
Kunst: Festschrift für Ernst Kühnel zum 75 Geburstag Welch, Stuart C. et Marie L. Swietochowski. « And This
am 26-10-1957. Édité par Richard Ettinghausen, 195- is the Picture of the Leopard Relating to the Army
206. Berlin : Gebr. Mann, 1959. What the Lion Has Ordered ». The Metropolitan
Welch, Stuart Cary. India: Art and Culture, 1300-1900. Museum of Art, Notable Acquisitions 1981-1982, 15-16.
New York : Metropolitan Museum of Art, 1985. New York : Metropolitan Museum of Art, 1982.
9
Étude de la matière picturale de six Kalīla wa-Dimna et d’un
Maqāmāt (xiiie-xviiie siècle)
Traduit du persan au viiie siècle par Ibn Au total, sept manuscrits ont fait l’objet de cette
al-Muqaffaʿ, le recueil de fables animalières Kalīla étude, par le biais de l’imagerie scientifique
wa-Dimna a été l’un des rares textes illustrés dans associée aux analyses élémentaires et colorimé-
le monde arabe. La présence dans les collections triques. Ils ont tous été copiés entre le xiiie et le
de la Bibliothèque nationale de France de six xviie siècle, sur papier en écriture naskhī.
manuscrits arabes illustrés s’échelonnant sur plu- Trois manuscrits se rattachent à la période
sieurs siècles a constitué une opportunité excep- médiévale. Le premier, Arabe 34652, est le plus
tionnelle d’étudier les matières picturales utilisées ancien Kalīla wa-Dimna illustré connu ; il a été
et de suivre l’évolution de la palette. En effet, si les attribué au premier quart du XIIIe siècle et loca-
nombreuses analyses physico-chimiques publiées lisé en Égypte ou en Syrie sur la base de sa ressem-
nous permettent désormais d’identifier les cou- blance stylistique avec un autre manuscrit littéraire
leurs employées dans l’Occident médiéval latin et conservé à la BnF, les Maqāmāt d’al-Ḥarīrī (Arabe
d’en connaître les usages, leur connaissance dans 6094), daté de 619H/1222-3. Ces deux manuscrits
le monde islamique reste majoritairement réser- sont emblématiques de la « période classique » ;
vée au domaine persan et peu de travaux, à ce jour, celle-ci, qui se termine en 1258 avec la prise de
sont consacrés à la peinture d’illustration arabe1. Bagdad par les Mongols, est considérée comme
l’apogée de la peinture d’illustration.
Arabe 3465 est orné de quatre-vingt-dix-huit
miniatures ; quatre-vingt-dix sont contempo-
1 Tracey D. Chaplin et al., « Raman Spectroscopic Analysis raines de la copie médiévale, huit plus tardives se
of Selected Astronomical and Cartographic Folios from trouvent sur des feuillets refaits en début et fin de
the Early 13th Century Islamic ‘Book of Curiosities of the
Sciences and Marvels for the Eyes’ », Journal of Raman
volume. Les miniatures anciennes (P1)3 se carac-
Spectroscopy 37, no 8 (2006) : 865‑877 ; Anna Contadini, térisent par la richesse et la variété de la palette
A World of Beasts: a Thirteenth-Century Illustrated Arabic chromatique, le tracé ferme et élaboré du dessin,
Book on Animals (the « Kitāb Naʿt al-ḥayawān ») in the les nombreuses nuances dans la pose des couleurs.
Ibn Bakhtīshūʿ Tradition (Leiden : Brill, 2012) ; François
Elles portent à de nombreux endroits la trace de
Déroche, éd., Manuel de codicologie des manuscrits en
écriture arabe (Paris : Bibliothèque nationale de France, repeints maladroits, peut-être effectués par l’un
2000) ; Patricia Roger, Malika Serghini et François Déroche,
« Les matériaux de la couleur dans les manuscrits arabes
de l’Occident musulman. Recherches sur la collection 2 http://gallica.bnf.fr/ark:/12148/btv1b84229611/f1.item.
de la Bibliothèque générale et archives de Rabat et de la 3 Afin de simplifier la lecture, nous avons adopté les abré-
Bibliothèque nationale de France », Comptes rendus des viations suivantes : P1 pour le peintre médiéval, P2 (pein-
séances de l’Académie des Inscriptions et Belles-Lettres 148, tures du début) et P3 (peintures de la fin) pour les peintres
no 2 (2004) : 799‑830. tardifs.
des deux peintres tardifs. Les titres de chapitre Le troisième manuscrit médiéval, les Maqāmāt
et leur table sont réalisés à l’encre dorée dans la d’al-Ḥarīrī (Arabe 6094)10, n’appartient pas au
partie ancienne. Les miniatures plus récentes sont corpus des Kalīla. Il contient trente-neuf minia-
de facture assez grossière et semblent exécutées tures très proches de celles de la partie ancienne
par deux mains différentes (P2 pour les premières d’Arabe 3465. Des repeints grossiers altèrent les
pages4, P3 pour les dernières5). Le premier groupe peintures d’origine. La palette, hormis le marron,
tardif (P2) se distingue par sa palette restreinte est semblable dans les deux copies ainsi que le
limitée à cinq couleurs : vert, jaune, rouge brique, dessin, quoique plus élaboré dans les Maqāmāt.
noir et gris. Le dessin est malhabile, les couleurs Ces similitudes ont souvent fait attribuer ces deux
débordent, les visages sont juste esquissés et volumes à un atelier commun11.
les traits sommairement exécutés. Les couleurs Quatre manuscrits, dont trois datés, ont été pro-
ont été utilisées comme des substances liquides duits durant la période ottomane. Moins soignés,
posées directement sur le support, sans prépara- ces volumes sont stylistiquement très différents
tion. Le second groupe tardif (P3) présente une et plusieurs d’entre eux sont liés à des modèles
gamme de couleurs plus étendue, jaune ocré, plus anciens. Le premier, Arabe 347212, a été copié
rouge brique, vert, bleu, blanc et noir, posées avec pour Sīdī Aḥmad le 3 juin 1669 (3 muḥarram
peu de soin, soit comme une encre liquide, soit 1080). Rapporté d’Alep par Pierre Diyab en 1673,
en couche épaisse, comme c’est également le cas le volume est probablement d’origine syrienne. Le
pour les quatre autres manuscrits tardifs. Parce frontispice, peu commun, représente une façade
qu’Arabe 3465 est représentatif de la peinture de maison (f.1v). Le texte est orné de trente minia-
du xiiie siècle et qu’il comporte des parties plus tures, avec ou sans cadre ; trois peintures sont pla-
récentes, un nombre plus important de feuillets, cées à l’envers (ff.43, 44v et 80v). Très sommaire, le
soit douze, a été étudié. dessin tracé en gris est peint dans une gamme de
Caractéristique de la période mamelouke, le couleurs limitée au vert, orange, jaune et marron.
second manuscrit (Arabe 3467)6 a été copié en Le second manuscrit, Arabe 588113, est daté
Égypte ou en Syrie. Il a été daté autour de 13547 de fin mai 1681 (rabīʿ al-ʿāẖir 1092). Le nom du
pour sa très grande proximité avec un manuscrit commanditaire (f.2) est effacé. Le texte est dans
d’Oxford8. Incomplet et comprenant de nombreux un encadrement, les titres des chapitres sont
feuillets mutilés ou en désordre, il renferme cin- rubriqués, en vert ou en bleu. Il est illustré de
quante miniatures dont trois ajoutées postérieu- quatre-vingt-seize miniatures de facture naïve,
rement9. La palette se compose de couleurs vives néanmoins plus élaborées que celles du manuscrit
et contrastées comme le bleu, le vert, le mauve, le précédent. Les personnages se caractérisent par
rose, l’orange ou le jaune. leurs longues moustaches et leurs turbans blancs ;
les pourtours du corps et les visages sont dessinés
à l’encre noire ; les couleurs, principalement des
4 ff.3, 3v, 22, 25. ocres jaune et rouge, du vert, du bleu et du gris,
5 ff.138v, 139v, 141, 143.
sont délavées. Les décors sont réalisés avec plus de
6 http://gallica.bnf.fr/ark:/12148/btv1b84152188/f1.image.
7 Sofie Walzer, « The Mamlûk Illuminated Manuscript of soin que les figures humaines, particulièrement
Kalîla wa-Dimna », in Aus der Welt der islamischen Kunst:
Festschrift für Ernst Kühnel zum 75 Geburstag am 26-10- 10 http://gallica.bnf.fr/ark:/12148/btv1b8422967h.image.
1957, éd. Richard Ettinghausen (Berlin : Gebr. Mann, 1959), 11 Hugo Buchthal, « “Hellenistic” Miniatures in Early
195‑206. Islamic Manuscripts », Ars Islamica 7, no 2 (1940) :
8 Oxford, Bodleian Library, Pococke 400. 125‑150.
9 Le feuillet 1v, postérieur à la copie, a été analysé mais les 12 http://gallica.bnf.fr/ark:/12148/btv1b8453971j.image.
résultats ne seront pas présentés ici. 13 http://gallica.bnf.fr/ark:/12148/btv1b84539756.image.
270 Buisson et Vernay-Nouri
Ibn al-Muqaffaʿ, Kalīla wa-Dimna P1 : 20v, 30v, 34, 43v, 51, 57, A3465
Arabe 3465, Peintures anciennes P1 : Syrie ou Égypte ? c. 1222 ? 62, 71, 86
Peintures tardives P2 et P3 : xviie siècle ? Égypte ? P2 : 25v
Papier oriental (parties anciennes) et filigrané (parties refaites), P3 : 138 et 143
29 cm × 25,5 cm, 146 f.
Ibn al-Muqaffaʿ, Kalīla wa-Dimna 30, 40v, 61, 70, 98v A3467
Arabe 3467, Syrie ou Égypte ? c. 1350-60
Papier oriental, 30 cm × 23 cm, 119 f.
al-Ḥarīrī, al-Maqāmāt 13, 49v, 68, 133v A6094
Arabe 6094, Syrie ou Égypte ? 1222-23
Papier oriental (parties anciennes) et filigrané (parties refaites),
30 cm × 23 cm, 187 f.
Ibn al-Muqaffaʿ, Kalīla wa-Dimna, 33v, 143v A3472
Arabe 3472, Syrie ? 1669
Papier filigrané, 21 cm × 15 cm, 165 f.
Ibn al-Muqaffaʿ, Kalīla wa-Dimna, 11v, 14v A5881
Arabe 5881, Égypte ou Syrie ? 1681
Papier filigrané, 20 cm × 13 cm, 91 f.
al-Muqaffaʿ, Kalīla wa-Dimna, 20v, 196v A3475
Arabe 3475, Égypte ? 1761
Papier filigrané, 22 cm × 16,5 cm, 199 f.
Ibn al-Muqaffaʿ, Kalīla wa-Dimna, 50, 65, 102v A3470
Arabe 3470, Égypte ? xviie siècle ?
Papier filigrané, 31,5 cm × 23,5 cm, 122 f.
Vert Orpiment + Vert au Vert au Orpiment + Orpiment + Orpiment + Orpiment Orpiment + Mélange
indigo cuivre cuivre indigo indigo indigo + matériau indigo de vert
(mala- (malachite ?) (f. 61, 70) bleu + vert (f20v) au cuivre
chite ?) + au cuivre (malachite)
un peu de Vert au (malachite Vert au + orpiment
vermillon cuivre probable) + cuivre
(mala- carbonate (malachite)
chite ?) + de calcium (f. 196v)
blanc de
plomb
(f. 61v)
Jaune Orpiment Ocre jaune Orpiment Ocre jaune Orpiment Orpiment Orpiment Orpiment Orpiment
utilisé + un peu Orpiment + + orpiment (ff. 50, 102v)
seulement d’orpiment ocre jaune Orpiment +
en mélange + vermillon (f. 143) ocre jaune
+ blanc de (f. 65)
plomb
Bleu Indigo (ff. – Indigo Indigo Indigo – Indigo + Indigo Indigo
34, 57, 61, 84) (ff. 68, 133) (ff. 61, 98v) vermillon +
Lapis-lazuli Lapis-lazuli Lapis-lazuli carbonate
(ff. 5, 52, 71, (ff.13,49v, (ff. 70, 98) de calcium
84) 68)
Dorure Or en Particules Or en Or en Or en – – – Or en
paillettes d’or et paillettes paillettes paillettes paillettes
dispersées de cuivre dispersées dispersées dispersées dispersées
dans un dispersées dans un dans un dans un dans un
liant dans un liant liant liant liant
liant
Rouge Vermillon Vermillon Vermillon Vermillon Vermillon + Brun Vermillon Vermillon + Vermillon
(ff. 20v, 30v, + minium +minium + blanc minium rouge : minium +minium
34v, 43v) +noir de Vermillon + de plomb Ocre (ff. 50, 102v)
Vermillon + carbone ocre rouge (ff. 13, rouge+ Vermillon
minium 49v, 68) vermillon + + minium + Vermillon +
(ff. 57,62, Vermillon + minium orpiment ocre (f. 65)
71, 86) ocre rouge+ (f. 196v)
blanc de
plomb
(f. 133v)
274 Buisson et Vernay-Nouri
Il est l’un des premiers pigments synthétiques aux gens du commun, encre réservée aux livres
fabriqué depuis l’Antiquité21. religieux22.
Hormis les peintures d’A5881, il n’a pas été
retrouvé de pigments blanc pur dans les manus- 6.5 Le gris
crits des xviie-xviiie siècles. Le rendu du blanc La nuance gris argenté, mélange de noir de car-
résulte d’un effet visuel, par la mise en réserve bone et de blanc de plomb, n’a été observée qu’à
du papier. Le blanc de plomb a toutefois servi à deux reprises dans les manuscrits des xviie-xviiie
créer des rehauts blancs dans une peinture de P3 siècles : dans les peintures de P2 d’A3465 ainsi que
d’A3465, ainsi que dans celles d’A3475. Additionné dans celles d’A3470.
de noir de carbone, ce pigment a servi à réaliser les
gris argentés de P2 d’A3465. Enfin, il est employé 6.6 Les rouges
pour éclaircir les couleurs d’A3470, d’A3475 et des Sur les feuillets étudiés, deux rouges différents,
deux peintres tardifs d’A3465. Il est également un rouge vif et un rouge orangé, ont été obser-
intéressant de noter dans le cas des peintures vés et analysés. Les analyses XRF révèlent que la
d’A3475 que la mise en réserve pour obtenir un première teinte rouge se compose de mercure,
aspect blanc n’a pas du tout été utilisée. À l’excep- élément caractéristique du vermillon, le pigment
tion des rehauts mentionnés plus haut, ce manus- rouge le plus fréquemment employé dans notre
crit est totalement exempt de décor blanc. corpus. Ce matériau a été utilisé seul dans les
D’emploi plus rare, le carbonate de calcium n’a peintures les plus anciennes d’A3465 ainsi que
été identifié que dans les peintures d’A5881. Ce dans celles, plus récentes, d’A5881. Il a parfois été
matériau a servi à la fois pour exprimer la couleur additionné de blanc de plomb pour être légère-
blanche et pour diluer les autres couleurs. Compte ment éclairci (A6094). Le vermillon ou cinabre est
tenu de la facture modeste de ces peintures, le un sulfure de mercure (HgS) qui a été largement
faible coût de ce matériau a probablement justifié utilisé par le passé du fait de son accessibilité et
son emploi. de sa simplicité de préparation23. Forme naturelle
du pigment, il est employé depuis l’Antiquité24 ; le
6.4 Le noir vermillon en est la forme synthétique. Les annexes
Le noir a été observé dans toutes les miniatures des traités de Qadi Ahmad25 et de Sadiqi26 men-
à l’étude, à l’exception de celles d’A3472, qui n’en tionnent les deux façons de l’obtenir : soit à partir
renferme pas. Les noirs analysés sont tous issus de du cinabre naturel, soit à partir de la réaction du
noir de carbone, parfois additionné d’ocre rouge
(f.86 d’A3465). La façon d’obtenir des noirs est 22 Martin Levey, « Mediaeval Arabic Bookmaking and
largement documentée : il pouvait être produit Its Relation to Early Chemistry and Pharmacology »,
Transactions of the American Philosophical Society 52,
à partir de la calcination de matière végétale ou no 4 (1962) : 1-79.
recueilli à partir de vapeurs de combustion. Ibn 23 Abdelmajid El Bakkali et al., « Non-Invasive Micro
Bādīs au xie siècle répertorie deux grands types Raman, SERS and Visible Reflectance Analyses of
d’encre : les métallo-galliques et celles à base de Coloring Materials in Ancient Moroccan Islamic
Manuscripts », Journal of Raman Spectroscopy 44, no 1
suie. De nombreuses recettes sont mentionnées
(2013) : 114-120.
dans son traité : encre sèche ou utilisable immé- 24 Chaplin et al., « Raman Spectroscopic Analysis ».
diatement, encre ne requérant pas de feu pour 25 Qadi Ahmad, Calligraphers and Painters, trad. Vladimir
sa préparation, encre destinée aux voyageurs ou Minorsky (Washington : Smithsonian Institution,
1959).
26 Martin B. Dickson et Stuart C. Welch, « Appendix 1: The
Canons of Painting by Sadiqi Bek », in The Houghton
Shahnameh, éd. Martin B. Dickson et Stuart C. Welch
21 Chaplin et al., « Raman Spectroscopic Analysis ». (Cambridge : Harvard University Press, 1981) : 259-269.
276 Buisson et Vernay-Nouri
mercure et du soufre. Les techniques structurelles Il peut être additionné de blanc de plomb (A6094)
et spectroscopiques ne permettent pas de diffé- ou encore de minium pour donner un brun rouge
rencier ces deux produits. (A3472). On peut noter l’absence de couleur rouge
Concernant le second rouge, les analyses XRF dans les peintures postérieures d’A3467 hormis
ont décelé la présence de mercure et de plomb. pour tracer les contours des figures et des décors.
Cette teinte est donc issue du mélange vermillon
et minium. Le mélange semble assez commun 6.7 Les roses
car il a été identifié dans les peintures de P1 et P3 Les roses utilisés pour les carnations des peintures
d’A3465 ainsi que dans celles d’A3467, A3470 et médiévales (P1 d’A3465 et A3467) sont constitués
A3475. Le minium (Pb3O4) est l’un des premiers d’un mélange de vermillon et de blanc de plomb.
pigments fabriqués artificiellement et son uti- Le même rose renfermant plus de vermillon est
lisation remonte au ve siècle av. J.-C. en Chine27. observé sur le pelage des lapins des f.56 et 70
Ce mélange minium et vermillon a été trouvé d’A3467 (Fig. 9.1).
par P. Roger dans d’autres manuscrits de la BnF28 Les carnations d’A6094 renferment en plus
et dans plusieurs manuscrits islamiques des du minium et de l’ocre. Dans les peintures des
xvie-xviiie siècles29. Il n’était pas exclusivement xviie et xviiie siècles, (P2 et P3 d’A3465, A3470 et
limité au monde méditerranéen puisqu’on le A3472), elles sont obtenues par la mise en réserve
trouve dans des manuscrits français dès le xie du papier. Deux manuscrits dérogent cependant
siècle30. Il existe d’autres combinaisons possibles à cette règle : celles d’A3475 se composent d’un
de ce minéral, par exemple avec de l’orpiment31. mélange de vermillon éclairci avec du blanc de
À ce mélange de vermillon et de minium, il plomb, tandis que celles d’A5881 sont faites d’un
est parfois ajouté soit du noir de carbone (P2 mélange de vermillon et de minium éclairci par
d’A3465), soit de l’orpiment (A3475), pour obtenir du carbonate de calcium. D’autres exemples de
des nuances différentes. Un autre mélange com- rose réalisés également avec du vermillon ont été
posé de vermillon et d’ocre rouge est identifié dans trouvés dans un manuscrit persan datant du xvie
les peintures de P3 d’A3465 et dans celles d’A3470. siècle du poète Anvari32 ainsi que dans un manus-
crit turc du xvie siècle33.
27 Elisabeth W. Fitz Hugh, « Orpiment and Realgar », Une seconde nuance de rose a été observée sur
in Artists’ pigments: A Handbook of their History and
les feuillets de P1 d’A3465 et d’A6094. L’absence
Characteristics, éd. Robert L. Feller (Washington :
National Gallery of Art, 1986), 47-81. d’éléments caractéristiques de pigments minéraux
28 Roger, Serghini et Déroche, « Les matériaux de la décelables en XRF, à l’exception du plomb, sug-
couleur dans les manuscrits arabes de l’Occident gère que cette teinte est obtenue en mélangeant
musulman ». un colorant organique avec du blanc de plomb.
29 Lucia Burgio et al., « Pigment Analysis by Raman
Microscopy of the Non-Figurative Illumination in
L’observation des images composites en infrarouge
16th- to 18th-Century Islamic Manuscripts », Journal fausses couleurs montre une coloration orangée
of Raman Spectroscopy 39, no 10 (2008) : 1482-1493 ; claire pour l’ensemble des éléments de décors sui-
Dan A. Ciomartan et Robin J.H. Clark, « Raman vants : le brun-rose du manteau du sage du f.20v
Microscopy Applied to the Analysis of the Pigments
Used in Two Persian Manuscripts », Journal of the
Brazilian Chemical Society 7, no 6 (1996) : 395‑402. 32 Vânia S.F. Muralha, Lucia Burgio et Robin J.H. Clark,
30 Roger, Serghini et Déroche, « Les matériaux de la « Raman Spectroscopy Analysis of Pigments on 16-17th c.
couleur dans les manuscrits arabes de l’Occident Persian Manuscripts », Spectrochimica Acta Part A: Mole-
musulman ». cular and Biomolecular Spectroscopy 92 (2012) : 21‑28.
31 Silvia Bruni et al., « Micro-Raman Identification of the 33 Alicia Jurado-López et al., « Analysis of the Palette of a
Palette of a Precious XVI Century Illuminated Persian Precious 16th Century Illuminated Turkish Manuscript
Codex », Journal of Cultural Heritage 2, no 4 (2001) : by Raman Microscopy », Journal of Raman Spectroscopy
291-296. 35, no 2 (2004) : 119-124.
Étude de la matière picturale de six Kalīla wa-Dimna 277
figure 9.1
Pelage rose du deuxième lapin,
Paris, BnF, Arabe 3467n, f.70
figure 9.2a-b
Seconde nuance de rose montrant une
réponse orange clair en IRFC, Paris, BnF,
Arabe 3465, P1, f.20v
278 Buisson et Vernay-Nouri
figure 9.3a Spectres de réflectance des bleus mesurés sur les feuillets 34, 57 et 84 de Paris, BnF, Arabe 3465, comparés au
spectre de réflectance de l’indigo de référence
figure 9.3b Spectres de réflectance des bleus mesurés sur les feuillets 52, 52, 71 et 84 de Paris, BnF, Arabe 3465, comparés
au spectre de réflectance du lapis-lazuli de référence
280 Buisson et Vernay-Nouri
de détails fins. Il semblerait que l’usage des maté- se situent à Badakhshan en Afghanistan septen-
riaux bleus ait évolué au cours du temps dans les trional et sont exploitées depuis plus de six mille
manuscrits étudiés. Ceux des xviie-xviiie siècles ans pour l’utilisation en bijouterie, comme pierre
utilisent exclusivement l’indigo à l’exception de d’ornementation et comme pigment41. Ces mines
deux qui ne contiennent pas de bleu (P2 de A3465 ont fourni l’Extrême et le Proche-Orient, le monde
et A3472). Pour l’ensemble du corpus, les bleus ont méditerranéen et l’Occident42.
été éclaircis avec du blanc de plomb, sauf dans
A5881 où le peintre s’est servi de carbonate de 6.10 Les verts
calcium. Un vert composite, le vergaut, a été identifié dans
Le colorant indigo (C16H10N2O2) peut être fabri- les trois manuscrits médiévaux ainsi que dans trois
qué à partir de nombreuses plantes. L’espèce la des manuscrits des xviie-xviiie siècles (A3472,
plus répandue, indigofera tinctoria (Cardon et al., A5881, A3475). Les courbes obtenues en réflexion
1990), est originaire des zones tropicales. L’indigo spectrale de ces verts correspondent au spectre de
pourrait être d’origine indienne (le terme dérive référence de l’indigo (Figs. 9.4A-B). Les analyses
du grec indikon)38 ; il subsiste de nombreux témoi- en XRF avaient détecté de l’arsenic et du soufre.
gnages de sa culture en Palestine, en Égypte, au De l’orpiment a donc été mélangé à l’indigo pour
Yémen, en Afghanistan et au Maghreb. Il est utilisé obtenir cette teinte verte, appelée vergaut. A5881
dans le monde islamique comme plante médici- contient en plus du vert au cuivre. Cependant, il
nale, ornement de la peau, teinture pour les che- est difficile de se prononcer sur la nature précise
veux, mais son principal domaine d’utilisation de ce matériau vert. S’agit-il d’un mélange de plu-
reste la teinture des textiles39. sieurs matériaux (vergaut associé à du vert au
Le précieux lapis-lazuli est une roche dure cuivre) ou d’un repeint ?
d’un bleu profond, composée d’un mélange de Le vergaut a été utilisé comme une alternative
minéraux, comprenant généralement de la cal- au vert-de-gris par les peintres médiévaux. Ces
cite (veines blanches), de la pyrite de fer (cristaux mélanges de matériaux jaunes et bleus pour don-
jaunes cuivrés) et de la lazurite, ce dernier étant ner du vert ne sont pas rares et ont été identifiés
le constituant majoritaire. Le terme lapis-lazuli à la fois dans des manuscrits occidentaux et isla-
est souvent utilisé de façon incorrecte lorsqu’il miques43. Les combinaisons de ces matériaux sont
désigne la lazurite, qui ne constitue qu’un seul des multiples et constituent un reflet des approvision-
composants mais est le seul à conférer la couleur nements possibles44. Le mélange orpiment-indigo
bleue à la pierre. La lazurite est un sulfate d’alu- a été constaté à plusieurs reprises, par exemple
minosilicate sodo-calcique cubique de formule dans le manuscrit de Kells (vers 800)45, dans
générale (Na, Ca)8[(Al,Si)12O24]Sn. La couleur
bleue est due à la présence de radicaux anions 41 Eastaugh, Walsh et Chaplin, The Pigment Compendium:
soufrés40. Les principales mines de lapis-lazuli A Dictionary of Historical Pigments.
42 François Delamare, Bleus en poudres : de l’art à l’indus-
trie : 5 000 ans d’innovations (Paris : Presses de l’École
38 Nicholas Eastaugh, Valentine Walsh et Tracey Chaplin, des Mines, 2008).
The Pigment Compendium: a Dictionary of Historical 43 Katherine L. Brown, « Raman Microscopic and Com-
Pigments (Amsterdam : Elsevier, 2004). putational Studies of Artists’ Pigments and Molecular
39 Alain F. George, « Calligraphy, Colour and Light in Inorganic Compounds », (thèse de doctorat, Université
the Blue Qurʾan », Journal of Qurʾanic Studies 11, no 1 de Londres, 2002).
(2009) : 75-125. 44 Claude Coupry, « À la recherche des pigments », Revista
40 Eastaugh, Walsh et Chaplin, The Pigment Compendium: de História da Arte, série W, no 1 (2011) : 127-136.
A Dictionary of Historical Pigments; Muralha, Burgio, et 45 Susan Bioletti et al., « The Examination of the Book of
Clark, « Raman spectroscopy analysis of pigments on Kells Using Micro-Raman Spectroscopy », Journal of
16-17th c. Persian manuscripts ». Raman Spectroscopy 40, no 8 (2009) : 1043-1049.
Étude de la matière picturale de six Kalīla wa-Dimna 281
figure 9.4a Photographie en lumière directe figure 9.4b Image composite infrarouge fausses couleurs
de la robe de Paris, BnF, Arabe 3467, f.61
figure 9.5b Jaune d’orpiment pur, Paris, BnF, Arabe 3467, f.30
figure 9.6
Orange du f.133v de Paris,
BnF, Arabe 3472
284 Buisson et Vernay-Nouri
figure 9.7 Brun de la toque du gouverneur, Paris, BnF, Arabe 6094, f.133v
Étude de la matière picturale de six Kalīla wa-Dimna 285
figure 9.8a Or pur en paillettes dispersées dans un liant, figure 9.8b Or pur en paillettes dispersées dans un liant,
Paris, BnF, Arabe 6094, f.49 Paris, BnF, Arabe 3467, f.98v
figure 9.8c Or pur en paillettes dispersées dans un liant, figure 9.8d Particules d’or et de cuivre dispersées dans
Paris, BnF, Arabe 3470, f.50 un liant, Paris, BnF, Arabe 3465, P2, f.25v
coûtait moins cher que l’or en feuille et pouvait d’une colle de peau et d’un bol argileux. Les dorures
être appliqué avec un stylet. De plus, cette tech- de ce corpus ne présentent pas de relief notable, ce
nique permettait d’exécuter les détails avec plus de qui est le cas pour un grand nombre de manuscrits
finesse60. Il n’a pas été possible de déterminer le latins. La présence d’une telle assiette se recon-
type de colle employé. Cette technique de dorure naît en effet à l’aspect plus ou moins bombé des
n’utilise pas d’assiette colorée, consistant en dorures, qui leur donne un plus grand éclat61. Les
une couche préparatoire composée d’un mélange manuscrits A5881, A3475 et A3472 ne renferment
pas de dorure. Enfin, des substituts de l’or, comme
of the American Institute for Conservation 30, no 2 l’or mussif, un sulfure stannique (SnS2) ou le mica,
(1991) : 125-144 ; Burgio et al., « Pigment Analysis by largement utilisés par les peintres, n’ont pas été
Raman Microscopy of the Non-Figurative Illumination relevés dans le corpus des Kalīla wa-Dimna.
in 16th- to 18th-Century Islamic Manuscripts ».
60 Rushdya R.A. Hassan, « Analytical Study of a
Manuscript, “Tafsir Al Khazin” – the Seventeenth
Century AD », Current Science International 4, no 2 61 Déroche, Manuel de codicologie des manuscrits en écri-
(2015) : 196-207. ture arabe.
286 Buisson et Vernay-Nouri
7 Manières de peindre et pratiques générale, les artistes des xiiie et xive siècles uti-
d’artistes lisent un spectre de couleurs plus large et leur
peinture est plus élaborée qu’aux xviie et xviiie
Cette étude a fourni une occasion unique d’ap- siècles. Des différences apparaissent néanmoins,
profondir les connaissances sur les matériaux d’abord entre les copies des xiiie et xive siècles,
et les techniques utilisées dans la peinture figu- ensuite entre les deux plus anciennes.
rative arabe, champ peu étudié jusqu’à présent. La richesse chromatique de la période médié-
L’imagerie scientifique, la spectrocolorimétrie vale semble due non pas à un plus grand nombre
ainsi que la XRF portable ont permis de caracté- de pigments, mais plutôt aux nombreux mélanges
riser la palette utilisée dans les sept manuscrits et auxquels ils participent. Seul le lapis-lazuli, un
d’évaluer d’éventuels changements dans l’utilisa- matériau rare et précieux, n’apparaît que dans ce
tion des pigments au cours du temps. L’observation groupe. Les trois manuscrits médiévaux portent
à la loupe binoculaire a été un précieux auxiliaire clairement les traces d’un dessin préparatoire
pour mieux percevoir les processus de création ébauché à l’encre rouge. Ce tracé, particulièrement
des images. visible dans les zones où les pigments sont détério-
Les résultats ont confirmé en les affinant les rés, n’a pas toujours été respecté par le peintre au
observations faites à l’œil nu ; d’une manière
figure 9.9 Tracé préparatoire en rouge, mis en évidence par la détérioration de la matière picturale, Paris, BnF,
Arabe 6094, f.13
Étude de la matière picturale de six Kalīla wa-Dimna 287
figure 9.10 Dernière arcature du dais tracé à la pointe sèche, Paris, BnF, Arabe 3465, f.20v
moment de la mise en couleurs, le laissant appa- de robe, turbans, ṭirāz …) ou de décors. Le jaune
rent à de nombreux endroits (Fig. 9.9). (orpiment), pratiquement absent comme coloris,
Il peut s’accompagner d’un trait exécuté à la est employé mélangé à du bleu pour faire du vert.
pointe sèche pour exécuter certains décors archi- Les peintres ont privilégié en complément des
tecturaux géométriques, par exemple l’arcature teintes assez claires comme le rose pâle (beau-
faite au compas au f.20v d’A3465 (Fig. 9.10). coup de blanc de plomb, vermillon, colorant orga-
nique non identifié), le mauve (le précédent avec
7.1 Les manuscrits du xiiie siècle du bleu), le brun rose (même mélange que le rose
Les deux manuscrits du xiiie siècle (A3465 et pâle) et le blanc utilisé pur. Ces couleurs ont été
A6094) se distinguent par une gamme étendue réalisées en éclaircissant les pigments avec du
de couleurs due à leurs nombreuses déclinaisons blanc de plomb. Ainsi, à partir d’un même bleu, on
dans des tonalités claires ou foncées. Trois couleurs a obtenu plusieurs nuances en ajoutant des quan-
constituent la base de la palette : le rouge (ver- tités croissantes de grains blancs (Fig. 9.11A).
millon), le bleu (indigo et lapis-lazuli) et le vert, Pour le vert, composé d’un mélange d’orpiment
qui sont utilisés pour les personnages et le décor et d’indigo, on a procédé similairement, les grains
architectural ou végétal. L’or y est souvent associé jaunes d’orpiment étant plus ou moins finement
pour peindre des éléments du costume (revers concassés selon les endroits (Fig. 9.11B).
288 Buisson et Vernay-Nouri
figure 9.11a Nuances de bleu, Paris, BnF, Arabe 3465, figure 9.11b Nuances de vert, Paris, BnF, Arabe 3465,
f.20v f.43v
Les peintres ont rempli les surfaces à colorer de vif de vermillon à une grande quantité de blanc de
plusieurs manières. Pour les vêtements, ils ont plomb, a été appliquée à l’intérieur des contours.
posé une première couche plus ou moins homo- L’arrondi des pommettes, le front et le menton ont
gène et y ont ajouté des motifs ornementaux d’une été accentués par un orangé plus soutenu. Enfin,
couleur plus foncée ou contrastée. À d’autres un rehaut blanc plus ou moins épais a été par-
endroits, pour obtenir des tonalités différentes, ils fois passé sur les bras, les pieds et les décors des
ont juxtaposé des bandes s’échelonnant du plus vêtements pour figurer un semblant de modelé.
sombre au plus clair qui donnent un effet fondu Dans A6094, le traitement des figures est sen-
de dégradé. Ceci est particulièrement visible dans siblement le même, hormis le trait blanc qui est
la peinture des feuilles ou des rochers. Enfin, pour appliqué plus finement, pour accentuer les reliefs
des surfaces unies, comme le pelage des animaux du visage (Figs. 9.12A-B).
(l’éléphant f.6) ou les briques (f.11v), le peintre a Au regard de leur proximité stylistique, ces deux
étalé la couleur par petites touches, de façon non manuscrits ont souvent été attribués à un atelier
uniforme. L’ensemble concourt à une palette d’une commun, voir à un même peintre. L’observation
grande variété. à l’œil nu et à la loupe binoculaire ainsi que les
Dans les rares exemples où la peinture n’a pas résultats des analyses montrent néanmoins des
été retouchée dans A3465, il a été possible d’obser- différences qui semblent exclure l’hypothèse d’un
ver le mode de réalisation des figures humaines. seul artiste. On trouve dans A6094 plusieurs cou-
Le dessin préparatoire a été ébauché en rouge à leurs absentes d’A3465, un ton de marron particu-
grands gestes vifs. Un trait plus foncé a été passé lier composé d’ocre rouge mélangé à du vermillon
par-dessus pour définir les contours du visage, éclairci au blanc de plomb, quelques rares zones
du nez, de la bouche et des yeux, des mains et de couleur jaune à base d’ocre et d’orpiment ainsi
des pieds, parfois augmenté d’un trait noir pour que des associations spécifiques pour les rouges,
accentuer le rendu des paupières, des sourcils les roses et les mauves. Peut-être dues à l’appro-
et des pupilles. Du gris a été ensuite ajouté pour visionnement des pigments, ces modifications
modeler les sourcils et donner du relief au nez, et de la palette ne sont néanmoins pas suffisantes
l’intérieur de l’œil a été peint en blanc. La couleur pour différencier les artistes, mais elles s’ajoutent
de la carnation, obtenue par ajout de grains rouge à d’autres dissemblances comme le plus grand
Étude de la matière picturale de six Kalīla wa-Dimna 289
figure 9.12a Visage, Paris, BnF, Arabe 3465, f.71 figure 9.12b Visage, Paris, BnF, Arabe 6094, f.13
62 Par exemple les deux autres manuscrits des Maqāmāt 63 Duncan Haldane, Mamluk Painting (Warminster : Aris
conservés à la BnF (Arabe 6094 et Arabe 5847) et & Phillips, 1978) ; Ettinghausen, La peinture arabe ;
datant également de la première moitié du xiiie siècle. James, Arab painting.
290 Buisson et Vernay-Nouri
précédents. On trouve également un orange fait à des gris bleutés ou des bruns rosés pour figurer le
partir de minium, vermillon et orpiment. Ces cou- pelage des animaux, il se sert de rose, de mauve ou
leurs, qui couvrent des surfaces plus importantes d’orange pour les chacals, de jaune pur pour le lion,
qu‘auparavant, sont posées uniformément les unes de rose et d’orange pour les cervidés. Les rochers
à côté des autres sans effet de dégradé ni de fondu sont peints en bleu ciel, jaune, mauve ou rose vio-
et en s’opposant : les bleus et violets sont suivis de lacé, les troncs des arbres en rose ou violacé. Les
jaune, de vert ou de rouge, ce qui renforce la vivacité éléments du décor architectural sont traités avec
chromatique. Le peintre du xive siècle s’éloigne une juxtaposition de couleurs vives (Fig. 9.13), les
davantage de toute préoccupation « réaliste » dans vêtements sont souvent d’une seule couleur avec
sa représentation de la nature et des animaux. Là des motifs ornementaux plus tranchés comme les
où ses prédécesseurs utilisaient des ocres rouges, motifs floraux noir/jaune sur la robe du roi (f.110).
Étude de la matière picturale de six Kalīla wa-Dimna 291
Les figures humaines (Fig. 9.14) montrent égale- ont été peintes montre que la volonté de figurer le
ment des différences de réalisation. modelé du visage grâce aux rehauts blancs, survi-
Le dessin préparatoire, particulièrement visible vance de la peinture byzantine, a totalement dis-
au niveau des mains du personnage trônant, de la paru dans le manuscrit peint sous les Mamelouks.
bordure extérieure droite du trône ainsi que de la
bordure extérieure de la robe du personnage cen- 7.3 Les manuscrits des xviie et xviiie siècles
tral du f.98v, a été ébauché au trait rouge orangé. Datés ou datables des xviie et xviiie siècles, les
Un rouge d’un ton différent a été passé par-dessus quatre manuscrits post-mamelouk (A3472, A5881,
pour définir les contours du visage, du nez, de la A3475, A3470) ainsi que les pages refaites d’A3465
bouche, des paupières supérieures, des mains et (P2 et P3) présentent un certain nombre de carac-
des pieds. Le peintre a ensuite dessiné d’un trait téristiques communes malgré un aspect assez dis-
noir effilé les yeux en amande, les sourcils fins semblable. D’un format plus petit et d’une écriture
et les bordures extérieures des vêtements et des moins soignée, leur dessin est plus grossier et leur
décors. Enfin, le rendu des chairs, un rose très clair mise en couleurs plus maladroite. Les volumes
résultant du mélange de blanc à des grains rouge semblent destinés à un public moins fortuné que
vif a été appliqué en aplat à l’intérieur des contours les exemplaires des siècles précédents dont cer-
de manière uniforme. La façon dont les carnations tains ont vraisemblablement servi de modèles.
292 Buisson et Vernay-Nouri
figure 9.15 Diversité des couleurs de Paris, BnF, Arabe 3475, f.196v
Ces caractéristiques impliquent-elles des modifi- papier pour donner l’illusion de la blancheur. Il
cations dans le choix et l’utilisation des pigments ? n’est que très peu employé pour diluer les autres
Hormis la disparition du lapis-lazuli au profit du pigments et ainsi obtenir la grande variété de tons
seul indigo, la gamme des pigments employés qui caractérisaient les palettes anciennes. Des cou-
reste pour la plupart la même64. La palette chro- leurs comme le brun rose, le violet et le rose vio-
matique renferme principalement du jaune, du lacé ont disparu, et le rose n’est plus observé que
vert, du rouge, du noir, du gris et du bleu, à l’excep- dans deux ouvrages : A3475 et A5881. En revanche,
tion de P2 d’A3465 et A3472, qui ne contiennent les nuances orangées semblent plus présentes. Les
pas du tout de bleu. Le blanc en tant que couleur dorures n’y sont pas du tout utilisées, ou alors elles
est désormais remplacé par la mise en réserve du sont mélangées à du cuivre. Enfin, dans la majorité
des cas, les couleurs sont plus acides et le pigment
n’est plus étalé de manière uniforme, mais souvent
64 On peut se demander si l’absence du lapis-lazuli est préparé comme une encre liquide et posé sans
due à une rupture d’approvisionnement ou au fait que préparation (Figs. 9.15 et 9.16).
sa cherté en faisait un pigment réservé à des volumes Le mode de réalisation des peintures est égale-
plus luxueux. C’est le cas du manuscrit de la collec- ment moins élaboré. Le tracé préparatoire rouge
tion S, produit vraisemblablement au début de la
période ottomane en Syrie et dans lequel les analyses
très présent dans les peintures médiévales a dis-
physico-chimiques ont révélé la présence de ce pig- paru et a été remplacé par un trait noir ou gris,
ment. Voir le chapitre 11 dans ce volume. plus ou moins épais, excepté dans A3472 où le
Étude de la matière picturale de six Kalīla wa-Dimna 293
figure 9.16 Diversité des couleurs de Paris, BnF, Arabe 5881, f.14v
dessin a été directement exécuté. Dans A3470, montre clairement que l’on a utilisé des calques
on distingue en outre sur certaines peintures ou des pochoirs à partir d’un autre manuscrit,
(f.50) des traits gravés à la pointe sèche et posés probablement le Kalīla wa-Dimna de Rabat. Les
pour indiquer l’emplacement de la composition couleurs ont ensuite été posées et, dans certaines
avant la réalisation du dessin. Cela pourrait être peintures, un trait noir plus épais et plus sombre a
la preuve que ce manuscrit a été effectué à partir été appliqué par endroits sur la composition pour
d’un modèle, peut être A3465. Un autre volume, accentuer certains détails.
A3475, dont une partie des illustrations n’a pas Le traitement des visages et des autres parties
été achevée, nous permet de suivre les étapes de du corps a été effectué de manière plus grossière
la fabrication des images. Le dessin a été esquissé dans toutes les peintures tardives. Le rendu des
d’un trait noir très fin, peut-être au calame. La chairs est généralement obtenu par la mise en
répétition des mêmes formes à l’identique ou en réserve de papier, sauf dans A5881 où la carnation
inversé, particulièrement pour les animaux65, des personnages a été réalisée par un mélange de
carbonate de calcium, de vermillon et de minium.
65 Par exemple les éléphants (ff.103, 126, 126v ,160v et La question se pose enfin de savoir si les ana-
196v) et les chacals (ff.39, 41v, 47, 49v, 63v, 76, 82, 84v). lyses physico-chimiques corroborent l’examen à
294 Buisson et Vernay-Nouri
l’œil nu, à savoir l’hypothèse de deux artistes dif- différence de composition a été également déce-
férents dans les peintures tardives d’A3465 (P2 et lée dans la composition chimique des dorures, un
P3) et la possibilité d’un artiste commun à P3 et alliage d’or et de cuivre pour P2 et des paillettes
à A3470. À l’encontre de l’analyse faite par Leïla d’or pur pour P3.
Benouniche et Bernard O’Kane66, les investiga- En outre, les peintures de P2 sont très proches
tions effectuées semblent étayer l’hypothèse selon de deux manuscrits des Mille et Une Nuits conte-
laquelle deux peintres différents ont œuvré dans nant l’histoire du roi Nuʿman, l’un conservé à
les premières et dernières pages d’A3465. Malgré Manchester67, l’autre à Tübingen68. Les dégrada-
quelques ressemblances stylistiques, notamment tions dans le papier dues à l’acidité de l’encre sont
au niveau du dessin des visages, les caractéristiques de plus exactement les mêmes. Les miniatures pré-
codicologiques des premiers et derniers feuillets sentent la même palette de couleurs très diluées,
confortent ce postulat : les filigranes, le nombre utilisées de la même façon en bandes contrastées.
de lignes par page, la réglure sont nettement dif- Les motifs ornementaux sont semblables : coif-
férents. L’écriture, quant à elle, est attribuable à fures composées de petits demi-cercles superpo-
deux copistes bien distincts. Enfin, la présence de sés de couleur opposée, traits rehaussés de petites
chiffres coptes seulement sur les premiers feuillets boucles à l’encre noire. Tous ces éléments laissent
prouve que les pages en début et en fin de volume raisonnablement penser que P2 et P3 ont été exé-
n’ont pas été introduites au même moment. Ces cutés par deux mains différentes.
constatations sont confirmées par les analyses Pour ce qui est de l’hypothèse d’un artiste
physico-chimiques. Les photographies en lumière commun à A3470 et au deuxième peintre tardif
directe et en infrarouge fausses couleurs montrent d’A3465, elle semble confirmée par un traitement
des différences tant dans la technique d’exécution stylistique et une palette très proche. Si l’on com-
des personnages que dans les palettes utilisées pare les lions peints sur les feuillets d’A3470 et de
(Figs. 9.17A-B-C). P2 d’A3465 (f.138v), on constate que ceux d’A3470
P2 présente une palette assez terne, presque semblent tous issus d’un même modèle, avec des
éteinte, aux couleurs fortement diluées et utili- variantes dans le positionnement des pattes et de
sées à la manière des encres, tandis que celle de la queue ou encore dans leur orientation. Les traits
P3 est beaucoup plus vive. La première ne ren- de la tête sont très semblables : yeux très ronds
ferme pas de bleu et est riche en gris. Le rouge et naseaux épatés. Des différences apparaissent
est différent : pour le premier il s’agit d’un rouge néanmoins au niveau du traitement des crinières
brique composé de vermillon additionné d’un peu et de la couleur du pelage et des naseaux. Mis à
de minium et assombri par des grains de noir de part quelques petites variantes dans le traitement
carbone, tandis que le second utilise du vermillon du corps, le lion d’A3465 semble être une copie
mêlé à du minium ou à de l’ocre rouge. L’étude en inversée du lion du f.50 de l’A3470. En effet, le trai-
fluorescence des rayons X indique des mélanges tement des deux têtes ainsi que les couleurs utili-
différents dans le vert de cuivre : l’un est mélangé à sées sont assez similaires (Fig. 9.18).
une petite quantité de vermillon tandis que l’autre Il en est de même pour la manière dont sont
est dilué au blanc de plomb ou à l’orpiment. Une peints les chacals69 d’A3470 à partir d’un même
modèle et avec des positionnements différents.
66 Leïla Benouniche, « Observations sur un manuscrit de Celui d’A3465 (f.138v) en est très proche, hormis de
Kalila et Dimna », Revue suisse d’histoire de l’art et d’ar-
chéologie 2, no 40 (1983) ; Leïla Benouniche, Le « Kalila
et Dimna » de Genève : histoire d’un recueil de fables 67 Manchester, John Rylands Library, MSS Arabic 646
illustré (Genève : Slatkine, 1995) ; Bernard O’Kane, Early [706].
Persian Painting: Kalila and Dimna Manuscripts of the 68 Tübingen, Universitätsbibliothek, M.a VI.32.
Late Fourteenth Century (Londres : I.B. Tauris, 2003). 69 F.66v et f.105 d’A3470 et f.138v d’A3465.
Étude de la matière picturale de six Kalīla wa-Dimna 295
figure 9.17a-b Photographie en lumière directe et image composite infrarouge fausses couleurs du f.25 de Paris, BnF,
Arabe 3465
figure 9.17c-d Photographie en lumière directe et image composite infrarouge fausses couleurs du f. 143 de Paris, BnF,
Arabe 3465
figure 9.18 Comparaison des lions réalisés par le peintre de Paris, BnF, Arabe 3470 avec celui du peintre de Paris, BnF,
Arabe 3465
296 Buisson et Vernay-Nouri
légères variantes dans l’application de la couleur Antonio Sgamellotti. « MOLAB® Meets Persia: Non-
bleue plus claire et posée en épaisseur. Invasive Study of a Sixteenth-Century Illuminated
Le traitement des visages dans les deux manus- Manuscript ». Studies in Conservation 60, no S1
crits présente aussi de nombreuses similitudes. (2015) : 185-192.
Dans la plupart des cas, les visages sont très arron- Barrucand, Marianne. « Un manuscrit arabe illustré
dis, les yeux, ronds, le nez très fin, la bouche, très de Kalila wa Dimna du XIIIe siècle et sa copie otto-
petite et les sourcils, très arqués. mane ». Archéologie islamique, no 2 (1991) : 81-95.
Les photographies réalisées en lumière directe Benouniche, Leïla. Le « Kalila et Dimna » de Genève : his-
montrent l’utilisation de palettes de couleurs simi- toire d’un recueil de fables illustré. Genève : Slatkine,
laires, aux nuances très vives. L’observation des 1995.
images composites en infrarouge fausses couleurs Benouniche, Leila. « Observations sur un manuscrit de
et les analyses physico-chimiques confirment que Kalila et Dimna », Revue suisse d’histoire de l’art et
des matériaux de même nature ont été utilisés d’archéologie 2, no 40 (1983).
dans les deux manuscrits, en dépit de certaines Bioletti, Susan, Rory Leahy, John Fields, Bernard Meehan
différences comme l’absence de gris dans A3465. et Werner Blau. « The Examination of the Book of
Le même vert au cuivre éclairci avec du blanc de Kells Using Micro-Raman Spectroscopy ». Journal
plomb ou additionné d’orpiment a été employé. of Raman Spectroscopy 40, no 8 (2009) : 1043-1049.
L’orpiment a subi un traitement similaire : addi- Brown, Katherine L. « Raman Microscopic and
tionné de blanc de plomb pour être éclairci ou Computational Studies of Artists’ Pigments and
mélangé à l’ocre jaune pour obtenir la tonalité Molecular Inorganic Compounds ». Thèse de docto-
souhaitée. Le mélange vermillon et ocre rouge ne rat, Université de Londres, 2002.
se rencontre seulement que dans ces deux manus- Bruni, Silvia, Franco Cariati, Francesca Casadio, et
crits étudiés. Hormis quelques rehauts au blanc Vittoria Guglielmi. « Micro-Raman Identification of
de plomb, les blancs résultent d’un effet optique. the Palette of a Precious XVI Century Illuminated
Bien que nous ne puissions pas affirmer avec cer- Persian Codex ». Journal of Cultural Heritage 2, no 4
titude qu’il s’agit du même peintre, il est permis (2001) : 291-296.
de penser que les feuillets de ces deux manuscrits Buchthal, Hugo. « “Hellenistic” Miniatures in Early
peuvent être attribués à un même atelier ou à des Islamic Manuscripts », Ars Islamica 7, no 2 (1940) :
pratiques d’ateliers semblables. 125-150.
Bukhari, Y.K. « Pigments ». Marg 16, no 2 (1963) : ii-iii.
En conclusion, cette étude a démontré une fois Burgio, Lucia, Robin J.H. Clark, Vânia S.F. Muralha,
de plus l’utilité de bâtir un lien entre les conser- et Tim Stanley. « Pigment Analysis by Raman
vateurs, les historiens de l’art et les scientifiques, Microscopy of the Non-Figurative Illumination in
de façon à acquérir une connaissance approfon- 16th- to 18th-Century Islamic Manuscripts ». Journal
die des matériaux utilisés par les peintres entre le of Raman Spectroscopy 39, no 10 (2008) : 1482-1493.
xiiie et le xviiie siècle. Néanmoins, il convient de Cennini, Cennino. Le livre de l’art ou traité de la peinture.
rester prudent quant à l’extrapolation des résultats Paris : F. De Nobele, 1978.
vu le nombre réduit de manuscrits analysés. Chaplin, Tracey D., Robin J.H. Clark, Alison McKay et
Sabina Pugh. « Raman Spectroscopic Analysis of
Selected Astronomical and Cartographic Folios from
Bibliographie the Early 13th Century Islamic ‘Book of Curiosities of
the Sciences and Marvels for the Eyes’ ». Journal of
Anselmi, Chiara, Paola Ricciardi, David Buti, Aldo Raman Spectroscopy 37, no 8 (2006) : 865-877.
Romani, Patrizia Moretti, Kristine Rose Beers, Ciomartan, Dan A., et Robin J.H. Clark. « Raman
Brunetto Giovanni Brunetti, Costanza Miliani et Microscopy Applied to the Analysis of the Pigments
Étude de la matière picturale de six Kalīla wa-Dimna 297
Used in Two Persian Manuscripts ». Journal of the Characteristics. Édité par Elisabeth W. FitzHugh, vol. 3,
Brazilian Chemical Society 7, no 6 (1996) : 395-402. 109-140. Washington : National Gallery of Art, 1997.
Clarke, Mark. « Anglo-Saxon Manuscript Pigments ». Flood, Finbarr B. et Gülru Necipoğlu. « Frameworks
Studies in Conservation 49, no 4 (2004) : 231-244. of Islamic Art and Architectural History: Concepts,
Contadini, Anna. A World of Beasts: A Thirteenth-Century Approaches, and Historiographies ». In A Com-
Illustrated Arabic Book on Animals (the « Kitāb panion to Islamic Art and Architecture. Édité par
Naʿt al-ḥayawān ») in the Ibn Bakhtīshūʿ Tradition. Finbarr B. Flood et Gülru Necipoğlu, vol. 1, 2-56.
Leiden : Brill, 2012. Hoboken : Wiley-Blackwell, 2017.
Coupry, Claude. « À la recherche des pigments ». Revista George, Alain F. « Calligraphy, Colour and Light in the
de História da Arte, série W, no 1 (2011) : 127-136. Blue Qurʾan ». Journal of Qurʾanic Studies 11, no 1
Delamare, François. Bleus en poudres : De l’art à l’indus- (2009) : 75-125.
trie : 5 000 ans d’innovations. Paris : Presses de l’École Haldane, Duncan. Mamluk Painting. Warminster : Aris
des Mines, 2008. & Phillips, 1978.
Deneckere, A., M. Leeflang, M. Bloem, C.A. Chavannes- Hassan, Rushdya R.A. « Analytical Study of a
Mazel, B. Vekemans, L. Vincze, P. Vandenabeele et Manuscript, “Tafsir Al Khazin” – the Seventeenth
L. Moens. « The Use of Mobile Raman Spectroscopy Century AD ». Current Science International 4, no 2
to Compare Three Full-Page Miniatures from the (2015) : 196-207.
Breviary of Arnold of Egmond ». Spectrochimica Acta James, David L. Arab painting. New Delhi : Marg
Part A: Molecular and Biomolecular Spectroscopy 83, Publications, 1977.
no 1 (2011) : 194-199. Jurado-López, Alicia, Ornela Demko, Robin J.H. Clark et
Déroche, François, éd. Manuel de codicologie des manus- David Jacobs. « Analysis of the Palette of a Precious
crits en écriture arabe. Paris : Bibliothèque nationale 16th-Century Illuminated Turkish Manuscript by
de France, 2000. Raman Microscopy ». Journal of Raman Spectroscopy
Dickson, Martin B. et Stuart C. Welch. « Appendix 1: 35, no 2 (2004) : 119-124.
The Canons of Painting by Sadiqi Bek ». In The Levey, Martin. « Mediaeval Arabic Bookmaking and Its
Houghton Shahnameh. Édité par Martin B. Dickson Relation to Early Chemistry and Pharmacology ».
et Stuart C. Welch, vol. 1, 259-269. Cambridge : Transactions of the American Philosophical Society
Harvard University Press, 1981. 52, no 4 (1962) : 1-79.
Eastaugh, Nicholas, Valentine Walsh et Tracey Chaplin. Merrifield, Mary P. Medieval and Renaissance Treatises
The Pigment Compendium: A Dictionary of Historical on the Arts of Painting: Original Texts with English
Pigments. Amsterdam : Elsevier, 2004. Translations. Two Volumes Bound as One, Mineola,
El Bakkali, Abdelmajid, T. Lamhasni, M. Haddad, New York : Dover Publications, 1999.
S. Ait Lyazidi, S. Sanchez-Cortes, et E. del Puerto Muralha, Vânia S.F., Lucia Burgio et Robin J.H. Clark.
Nevado. « Non-Invasive Micro Raman, SERS and « Raman Spectroscopy Analysis of Pigments on
Visible Reflectance Analyses of Coloring Materials 16-17th c. Persian Manuscripts », Spectrochimica Acta
in Ancient Moroccan Islamic Manuscripts ». Journal Part A: Molecular and Biomolecular Spectroscopy 92
of Raman Spectroscopy 44, no 1 (2013) : 114-120. (2012) : 21-28.
Ettinghausen, Richard. La peinture arabe. Genève : O’Kane, Bernard. Early Persian Painting: Kalila and
Skira, 1962. Dimna Manuscripts of the Late Fourteenth Century.
FitzHugh, Elisabeth W. « Orpiment and Realgar ». In Londres : I.B. Tauris, 2003.
Artists’ Pigments: A Handbook of their History and Purinton, Nancy et Mark Watters. « A Study of the
Characteristics. Édité par Robert L. Feller, vol. 1, 47-81. Materials Used by Medieval Persian Painters ».
Washington : National Gallery of Art, 1986. Journal of the American Institute for Conservation 30,
FitzHugh, Elisabeth W. « Red Lead and Minium ». In no 2 (1991) : 125-144.
Artists’ Pigments: a Handbook of their History and
298 Buisson et Vernay-Nouri
Qadi Ahmad. Calligraphers and Painters. Traduit par of a Late-Medieval Manuscript with Total
Vladimir Minorsky. Washington : Smithsonian Reflection X-ray Fluorescence and Micro-Raman
Institution, 1959. spectroscopy ». Analyst 124, no 2 (1999) : 169-172.
Roger, Patricia, Malika Serghini et François Déroche. Villela-Petit, Inès et Bernard Guineau. « Le Maître de
« Les matériaux de la couleur dans les manuscrits Boucicaut revisité : palette et technique d’un enlu-
arabes de l’Occident musulman. Recherches sur la mineur parisien au début du xve siècle ». Art de l’en-
collection de la Bibliothèque générale et archives de luminure 6 (2003) : 2-33.
Rabat et de la Bibliothèque nationale de France ». Wallert, Arie. « Orpiment and Realgar. Some Pigment
Comptes rendus des séances de l’Académie des Characteristics ». Maltechnik-Restauro 4 (1984) :
Inscriptions et Belles-Lettres 148, no 2 (2004) : 799-830. 45-55.
Vandenabeele, Peter, Bernhard Wehling, Luc Moens, Walzer, Sofie. « The Mamlûk Illuminated Manuscript of
Brigitte Dekeyzer, Bert Cardon, Alex von Bohlen et Kalîla wa-Dimna ». In Aus der Welt der islamischen
Reinhold Klockenkämper. « Pigment Investigation Kunst: Festschrift für Ernst Kühnel zum 75 Geburstag
am 26-10-1957. Édité par Richard Ettinghausen, 195-
206. Berlin : Gebr. Mann, 1959.
partie 4
Adaptations orientales / Oriental Adaptations
∵
10
Simple exotisme ou contestation subliminale : les costumes mongols
dans le Kalīla wa-Dimna de Rabat (BRR 3655)
… en des instants rapides comme des fendu, petit nez) et pour les costumes (caftan et
traits, minces comme eux, minces ceinture, bottes, bonnets à fourrure)4.
comme ce paradis qu’on voit entre les L’intrusion des Mongols au Moyen-Orient eut
paupières d’un Mongol … des conséquences semblables, sans toutefois cou-
Jean Genet, Notre-Dame des Fleurs vrir une aussi longue période. Leur figure s’affiche
ainsi sur plusieurs types de supports (illustrations
de manuscrits, mais aussi métaux, textiles ou céra-
∵ miques). Cependant, après la chute des Il-khanides
(1353), les images de Mongols ne s’égrènent plus
qu’au fil de certains ouvrages historiques.
1 Introduction À titre liminaire, il faut rappeler que le manus-
crit du Kalīla wa-Dimna de la Bibliothèque royale
L’arrivée d’un envahisseur se traduit souvent, dans de Rabat est incomplet ; acéphale, il y manque
un pays conquis, par l’irruption, dans le champ des également un colophon, qui aurait pu fournir des
représentations, de ses attributs ethnoculturels, renseignements sur le lieu et la date d’exécution de
au premier rang desquels se trouve le costume1. ce codex5. L’étude détaillée des peintures comme
Celui-ci devient ainsi un « signe extérieur » de de la graphie ont pourtant fourni des éléments
la nouvelle « hégémonie culturelle », pour utili- pour cerner le contexte dans lequel cette œuvre
ser la locution d’Antonio Gramsci2. Dans les arts a vu le jour. Ainsi, selon Marianne Barrucand,
de l’Orient musulman, un précédent au déferle- « il a été copié et illustré dans un milieu à la fois
ment des Mongols, comparable dans une certaine empreint de culture citadine arabe et marqué
mesure par ses répercussions dans les arts visuels, par des apports il-khanides originaux et donc (…)
fut la prise de pouvoir des Turcs (Ghaznavides
puis Seljoukides)3. Leur domination imposa en
effet de nouveaux traits esthétiques, à la fois pour
les canons de la beauté physique (visage rond, œil 4 Voir à ce sujet la remarquable étude de Priscilla Soucek,
« Ethnicity in the Islamic Figural Tradition: the Case of the
‘Turk’ », Tārīḫ 2 (1992) : 73-103.
1 À nouveau, toute ma gratitude à Richard Castinel, qui 5 Voir Marianne Barrucand, « Le Kalila wa Dimna de la
insuffle vibration et profondeur à ses « relectures ». Bibliothèque royale de Rabat : un manuscrit illustré
2 Antonio Gramsci (1891-1937), philosophe et théoricien il-khanide », Revue des Études islamiques, numéro spécial
politique marxiste, a notamment développé la théorie de en l’honneur de Dominique Sourdel, n°54 (1986) : 17-51 ;
« l’hégémonie culturelle ». Marianne Barrucand, « Un manuscrit arabe illustré de
3 On pourrait bien sûr ajouter un précédent plus ancien : Kalila wa Dimna du XIIIe siècle et sa copie ottomane »,
l’invasion arabo-musulmane, qui a durablement intro- Archéologie islamique, no 2 (1991) : 81-95 ; Ernst J. Grube,
duit, entre autres, des codes vestimentaires (turban, djel- « Prolegomena for a Corpus Publication of Illustrated
laba) fort différents de ceux en usage à la fin de la période Kalilah wa Dimna Manuscripts », Islamic Art 4, n°2 (1990-
sassanide. 91) : 374.
figure 10.1 Le roi et Bilad, Fable de Shedram, Iblad et Irakht, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat,
BRR, ms. 3655, f.104
figure 10.2 Khusraw ouvre son trésor pour Burzuya, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.2
Le plus fréquent est un bonnet illustré dans trente al-tavārīkh on distingue nettement deux types de
et une occurrences ; systématiquement de couleur plumes : de longues pennes d’aigle et des plumes
dorée, il se particularise par une forme asymé- duveteuses de hibou14. Sheila Blair précise que
trique et un plumet. Tous les exemples montrent sans doute, à l’origine, le nombre de plumes d’aigle
un individu de trois-quarts face ; du côté de la révélait un rang social15 ; clairement, cette notion
nuque, on distingue une sorte de queue terminée est évacuée dans notre manuscrit. Un portrait de
en pointe (Fig. 10.1). Kubilaï Khan réalisé en Chine présente également
En comparant cette coiffe avec les illustrations ce type de coiffe, sans plumet toutefois16.
des chroniques, on comprend que cet appendice Un deuxième type de coiffure, décliné sous
est en réalité un protège-nuque. Les représenta- plusieurs formes, est une sorte de chapeau plat,
tions historiques présentent volontiers ce type de rappelant le pétase grec mais en plus rigide, qui
couvre-chef de couleur noire et doublé d’un tissu apparaît dans huit occurrences de l’ouvrage (f.2,
rouge ; c’est notamment le cas dans les exemples
tirés des albums Diez13. Dans toutes les occur- 14 Sheila Blair décrit ces bonnets à plumes sans préciser
rences du manuscrit de Rabat, ce bonnet est garni l’origine de cette différence ornithologique dans Blair,
d’un plumet. Cependant, les plumes sont ici indif- « A Mongol Envoy », 46.
15 Blair, « A Mongol Envoy ».
férenciées alors que dans les peintures des Jāmiʿ 16 Seconde moitié du xiiie siècle, National Palace, Taipei ;
voir Komaroff et Carboni, The Legacy of Genghis Khan,
13 Voir Kadoi, « The Mongols Enthroned », 270. fig. 40, 19.
Simple exotisme ou contestation subliminale 305
figure 10.3 Le roi et Bilad, Fable de Shedram, Iblad et Irakht, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.105
3v, 12, 63v, 105). Les illustrations qui nous occupent couvre-chef avec son protège-nuque est conservé
l’enjolivent quasi systématiquement d’un plumet. à l’Académie des Sciences d’Ulan Bator, ce qui per-
Les couleurs utilisées sont le noir ou le rouge ; la met d’en connaître le matériau, en l’occurrence la
doublure est soit invisible, soit à peine suggérée soie20. Cette forme apparaît encore au xve siècle,
(Figs. 10.2 et 10.3). schématisée jusqu’à évoquer un bicorne, notam-
Les albums Diez déclinent une forme simi- ment dans la copie datable de 1430-1434 des Jāmiʿ
laire, parfois plus souple17. Un modèle dérivé, et al-Tavārīkh de Rashīd al-Dīn faite à Hérat21.
plus complexe, apparaît dans la copie des Jāmiʿ Un dernier type de couvre-chef masculin n’ap-
al-Tavārīkh conservée à Tachkent18. Une dernière paraît que dans une seule occurrence (f.12) ; il
variation, agrémentée d’un protège-nuque et ressemble vaguement à une barrette ou « bonnet
d’un bouton sommital, s’illustre dans un portrait carré »22. Comme on le verra plus bas, il serait inté-
peint en Chine19. Un exemple concret d’un tel ressant de vérifier si ces différents types de coiffe
entendent décrire une hiérarchie sociale (Fig.
10.4).
17 Voir Scène de trône, dans Komaroff et Carboni, fig.
84, 80. Voir aussi les illustrations plus tardives d’une
Anthologie poétique copiée en 713-14/1314-15, dans 20 Kadoi, « The Mongols Enthroned », fig. 9.8, 259.
Komaroff et Carboni, fig. 165, 139. 21 Paris, BnF, Supplément persan 1113, voir par exemple
18 Reproduit dans El’mira Marufovna Ismailova, Oriental ff.139v. ; 227v-228 ; 273v d’après Gallica.
Miniatures of Abu Raihon Beruni Institute of Orientology 22 Peut-être peut-on mettre cette forme de bonnet en
of the UzSSR Academy of Sciences (Tashkent : Gafur parallèle avec le kolâh-i chahâr-par mentionné par
Gulom Nomidagi Adabiët va San”at nashriëti, 1980). Bayhaqi, et que P. Soucek compare avec le « four-gored
19 https://commons.wikimedia.org/wiki/File:YuanEmpe qalpāq » ? Voir Soucek, « Ethnicity in the Islamic
rorAlbumTemurOljeituFull.jpg. Figural Tradition », 80.
306 Porter et Castinel
figure 10.4 Fable du marchand et du joueur de cymbales, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.12
2.2 Le caftan, la ceinture, les bottes des sacoches présente en outre une pointe dépas-
De manière systématique, le caftan figuré dans sant de sa base (Fig. 10.3).
nos illustrations se ferme en croisant de gauche à Un tel élément se retrouve, mais de façon
droite, par des nœuds ou boutons sur le côté. Une exceptionnelle, dans les images « historiques ».
large bande médiane évoque une ceinture, encore Dans un folio d’un album d’Istanbul provenant
que celle-ci puisse être partie intégrante du vête- d’un tel ouvrage dépecé, un homme est agenouillé
ment, sous la forme d’un empiècement de couleur devant le couple impérial. De son étroite ceinture,
unie, le plus souvent rouge. De même, d’autres probablement en cuir noir orné d’éléments métal-
empiècements unis figurent parfois au niveau des liques dorés, pendent une sacoche arrondie noir
épaules23. Quelques autres exemples décrivent et or ainsi qu’un objet en forme d’étroit fourreau
pour leur part de riches étoffes évoquant des bro- blanc et or, peut-être un plumier24.
carts ou lampas de soie à motifs (voir étoffe du per- Par ailleurs, le folio 32v de notre manuscrit
sonnage de droite de la Fig. 10.2). dépeint un homme avec une ceinture noire étroite
De nombreuses images montrent de petites et un autre avec une large bande de tissu rouge
sacoches ou bourses pendant de la ceinture ; l’une (Fig. 10.5).
figure 10.5 Les hommes tuent le cobra, Fable du corbeau, du cobra et du chacal, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat,
BRR, ms. 3655, f.32v
Cependant, dans des scènes de trône comme la en mongol). Celles-ci, pointues et avec un talon
page d’Istanbul Hazine 2153, f.148v déjà citée, une peu marqué, sont généralement rouges ou noires
majorité d’hommes ne porte pas de ceinture25. (Figs. 10.4 et 10.5).
Ainsi, ces éléments (ceinture et sacoches, voire Dans certains cas (f.99v), c’est une sorte de
plumier) paraissent avoir été empruntés sans tunique à manches mi-longues, ornée d’un galon
réelle connaissance ou prise en considération doré aux manches et au bas, qui est portée sur
d’une fonction précise ou d’une signature sociale une robe plus longue (Fig. 10.6). Plus rarement,
induite. comme f.12a, les pans de la robe ou manteau de
Les caftans « mongols » sont généralement plus dessus sont relevés et coincés par la ceinture
courts que les robes portées par les « Arabes »26, et (Fig. 10.4).
laissent notamment apparaître des bottes (gutul,
2.3 Costume féminin
25 Le fait d’enlever sa ceinture est, chez les Mongols, un Sur onze pages illustrant des épisodes dans les-
geste de soumission ; Roux, « Le bonnet et la ceinture », quels interviennent des personnages féminins,
17-18. seules deux (f.97, 108v) montrent des femmes
26 Le terme « arabe » est simplement utilisé ici pour arborant un costume mongol. Une telle rareté est
décrire les personnages « non Mongols » ; il n’a donc pas
de véritable connotation ethnolinguistique. On pour-
en soi susceptible de comporter une dimension
rait aussi bien parler de personnages « autochtones » sociale. Il s’agit de fait du même personnage, en
face à des « étrangers », les Mongols en l’occurrence. l’espèce une reine (Fig. 10.7).
308 Porter et Castinel
figure 10.6 Fable de Shedram, Iblad et Irakht, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.99v
figure 10.7 Le roi repousse Irakht pour avoir choisi la couronne, Fable de Shedram, Iblad et Irakht, Ibn al-Muqaffaʿ,
Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.108v
Simple exotisme ou contestation subliminale 309
Les deux femmes – dans deux scènes très simi- Les neuf autres pages présentent des person-
laires – se distinguent notamment par le port de nages en costume mongol ; un cavalier est figuré
la coiffe dite bughtaq, qui se caractérise par sa f.37 – mais l’image s’appuie sans doute sur la seule
forme tubulaire accompagnée d’un plumet27. mention d’un cheval rapide29. Un autre cavalier
La comparaison avec les illustrations des chro- (f.109) illustre pour sa part un chasseur tuant deux
niques fait apparaître les coiffes de l’exemplaire lionceaux. Rien, dans le texte, ne vient fournir à ces
de Rabat comme très exagérées, tant par la hau- divers cavaliers un quelconque « contexte » dont
teur du cylindre que par celle du plumet. Une telle on puisse déduire un statut social indubitable.
exagération incite à dénier aux personnages de Par ailleurs, des hommes tout à fait « ordinaires »
type mongol de notre manuscrit toute prétention sont également représentés vêtus de costumes
à décrire une réalité ethnoculturelle précise. En mongols :
revanche, il n’est pas impossible que cette forme − f.17v : Les périls de la vie ;
exagérée ait un rapport avec la narration, comme − f.69v : Le loup et l’arc30 : chasseur mort ;
on le verra plus bas. − f.71v : l’hôte/ou le dévot, réveillé par le rat31 ;
− f.85v : homme qui dort piqué par un serpent32 ;
− f.93v-94 : le dévot et ses rêves33 ;
3 Les Mongols mis en scène dans le − f.95 : l’homme et la mangouste34.
manuscrit de Rabat À ces figures isolées, on peut ajouter celle des deux
hommes du folio 32v : « Histoire du cormoran et
En préambule, il est entendu que les histoires l’écrevisse ».
racontées dans le Kalīla wa-Dimna sont bien large- Six images montrent des conversations entre
ment antérieures aux invasions mongoles, et que deux hommes assis au même niveau (Figs. 10.2
par conséquent, aucun « véritable » Mongol ne et 10.15 ; f.2, 3, 3v, 50, 107v, 111v) ; ces six scènes
figure dans les fables d’origine. opposent systématiquement un homme « mon-
Sur les cent vingt-deux illustrations que compte gol » et un autre « autochtone ». Les personnages
le manuscrit, quarante font apparaître des per- des folios 3-3v (mais probablement aussi le f.2),
sonnages « mongols » ; soixante-huit ne figurent sont légendés : al-malik Kasrā et Buzurjmihr,
que des animaux, alors que les quatorze restantes autrement dit, en transposant au général, un sou-
montrent des personnages uniquement habillés à verain et son vizir. Ici, l’accoutrement révèle une
la manière « autochtone ». dualité historiquement reconnaissable : le sou-
On pourrait penser, suivant l’action dépeinte verain Il-khanide et son vizir « persan »35. C’est
dans les illustrations, que le type de costume
peut nous éclairer sur le caractère et les qualités
des personnages, ainsi que sur la nature de leurs
interactions. le benêt 47a ; ce type de couvre-chef semble donc avoir
Treize illustrations ne s’animent que d’un seul une connotation sociale clairement dépréciative.
personnage ; sur ce nombre, seules quatre pages 29 ʿAbd Allāh Ibn al-Muqaffaʿ, Le Livre de Kalila et Dimna,
trad. André Miquel (Paris : Klincksieck, 2012), 84.
montrent des non Mongols, curieusement tous de
30 Ibn al-Muqaffaʿ, 135.
basse extraction (paysans : f.5 et 76v ; charpentier : 31 Ibn al-Muqaffaʿ, 137.
20 ; un chasseur de mine patibulaire : 65v)28. 32 Ibn al-Muqaffaʿ, 139.
33 Ibn al-Muqaffaʿ, 184.
34 Ibn al-Muqaffaʿ, 187.
27 Kadoi, « The Mongols Enthroned », fig. 9.6 A-B, 9.7, 258. 35 En l’occurrence, il n’y a pas de différence vestimentaire
28 Les images suivies d’un astérisque montrent des per- fondamentale entre « Persans » et « Arabes » ; comme
sonnages coiffés d’un bonnet pointu. On le retrouve signalé plus haut, la principale opposition se fait entre
également sur le voleur f.9a, les deux pêcheurs 35a, et « autochtones » et « étrangers ».
310 Porter et Castinel
le cas le plus évident où le vêtement établit clai- deuxième peinture, il s’agirait de la conclusion de
rement une distinction que l’on peut lire comme l’histoire, habilement menée par Iblad (Fig. 10.7).
« hiérarchique ». Les folios 110v, 104, 105, 106 et 108, de compo-
Ce type de situation peut être éclairé par le paral- sition comparable, décrivent sûrement encore
lèle avec le frontispice du Tarīkh-i Jahān-Gushay-i les longs échanges entre Shedram et Iblad. Ces
Juvaynī (Histoire de la conquête du monde) de ʿAṭā images qui semblent très stéréotypées s’inspirent
Malik al-Juvaynī (m. 1283), dans la copie datée de probablement de celles qui apparaissent dans
1290 conservée à la BnF. Cette page montre l’au- les ouvrages historiques38. Le souverain est assis,
teur à gauche, nommé par une inscription, coiffé soit sur une banquette (dorée : f.97, 99v, 106, 108v ;
d’un bonnet mongol doublé de rouge ; devant lui rouge : f.100v, 104, 105), soit sur un tabouret pliant
se trouve un jeune homme imberbe debout, éga- rouge (f.108), ou des coussins (f.98v). Cette der-
lement en Mongol, qui doit être Hülegü ou Abaqā, nière scène compte – outre le roi – quatre hommes
au service desquels fut Juvaynī36. Rappelons au « mongols », probablement les Brahmanes évo-
passage que la famille Juvaynī eut la charge du qués dans l’histoire (Fig. 10.8).
gouvernement de l’Iraq jusqu’en 1282. Il reste que En revanche, d’autres illustrations dépeignant
seule l’inscription éclaire sur les deux person- également des face-à-face entre « Arabe » et
nages, qui pourraient, sans cela, passer tous deux « Mongol » sont probablement plus ambiguës.
pour « authentiquement » mongols. Ainsi, les images des folios 46v à 47v illustrent
Dans le manuscrit de Rabat, une nuance peut l’épisode du Fourbe et du benêt, sans qu’il soit pos-
être suggérée : le souverain incarne bien entendu sible de déterminer leurs qualités respectives39.
le pouvoir, mais le vizir représenterait quant à lui Peut-être la page suivante (47), qui montre un
une sorte d’autorité légitime ; nous reviendrons individu à chapeau pointu en train de creuser sous
plus bas sur cette idée. l’arbre, pendant que l’autre l’observe, signifierait
La peinture folio 111v représente le saint homme que le premier est le benêt, l’autre le fourbe ; on
et son hôte37 (Fig. 10.15). Les deux personnages sont remarque au passage que l’« Arabe » est habillé dif-
dépeints en conversation, leurs mains en direction féremment dans les deux pages. Mais le folio 47v,
l’un de l’autre, seulement séparés par une coupe illustrant la conclusion de l’anecdote, ne fait figu-
pleine de sucreries. On pourrait penser que le per- rer aucun personnage mongol, sans doute parce
sonnage « arabe » est le saint homme, le Mongol que l’image représente le moment où le juge arrive
étant alors son hôte. L’illustration folio 107v est sur les lieux alors que le fourbe n’a pas encore été
d’une composition assez semblable ; pourtant, elle démasqué40 (Figs. 10.9, 10.10 et 10.11).
décrit certainement l’un des nombreux échanges On pourrait provisoirement en conclure que
entre le roi Shedram et son confident Iblad. les rôles, comme les costumes, sont alors inter-
Les folios 97 et 108v montrent la reine Irakht changeables, ou plus vraisemblablement, qu’ils
devant Shedram, tous deux en Mongols. Il est ne dépeignent pas un personnage pour son
probable que la taille exagérée de la coiffe de
type bughtaq mentionnée précédemment tra- 38 Il s’agit notamment de ce que Y. Kadoi nomme
« Miniature-Sized Enthronement Scenes » ; voir
duise le rôle de « déclencheur » que la couronne
notamment Kadoi, « The Mongols Enthroned »,
choisie par la reine joue dans cet épisode. Dans la 270. Voir aussi la version datable c. 1300 conservée à
Tachkent, Bibliothèque A.R. Biruni, n°1620.
39 Un cas similaire est représenté au folio 50, avec l’his-
36 Paris, BnF, Supplément persan 205, ff.1v-2 ; Francis toire du Marchand et le Dépositaire infidèle ; Ibn
Richard, Splendeurs persanes : manuscrits du XIIe au al-Muqaffaʿ, Le Livre de Kalila et Dimna, 97.
XIV e siècle (Paris : Bibliothèque nationale de France, 40 Sur le thème du juge à dos d’âne (ou plutôt de mulet),
1997), no 7, 41. voir Barrucand, « Le Kalila wa Dimna de la Bibliothèque
37 Ibn al-Muqaffaʿ, Le Livre de Kalila et Dimna, 253. royale de Rabat », 25.
Simple exotisme ou contestation subliminale 311
figure 10.8 Fable de Shedram, Iblad et Irakht, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.108
312 Porter et Castinel
figure 10.9 Fable du fourbe et du benêt, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.46v
Simple exotisme ou contestation subliminale 313
figure 10.10 Fable du fourbe et du benêt, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.47
314 Porter et Castinel
figure 10.11 Fable du fourbe et du benêt, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.47v
Simple exotisme ou contestation subliminale 315
statut social mais plutôt pour sa position de force de costume entre ces deux personnages indiquent
circonstancielle. une claire différence sociale. La scène transcrit
Cette idée s’applique en particulier à d’autres toutefois une situation triviale de transaction com-
cas beaucoup plus conflictuels : ainsi, dans l’his- merciale où chaque protagoniste est tour à tour
toire de l’amant surpris par le mari, une première en situation d’exiger une prestation de l’autre. Là
image (10v) montre l’amant comme un jeune encore aucune légitimité statutaire n’est à l’œuvre,
homme imberbe habillé à la mongole, alors que mais un rapport de force permet à chacun d’impo-
l’image suivante, dans laquelle l’amant est battu ser une exigence.
par le mari (11a), l’amant est représenté comme Le folio 14 montre deux hommes devant le
un « autochtone » barbu et le mari en Mongol à cadi de Marv ; l’un est habillé en « autochtone »,
barbiche41. Cette inversion des rôles/costumes l’autre en Mongol. Le cadi rend d’abord un juge-
est évidemment très surprenante, voire décon- ment favorable à l’un, puis un deuxième favorable
certante : le costume apparaît là comme non à l’autre. Il est impossible ici de distinguer l’un de
réellement signifiant, en ce qu’il traduirait une l’autre ces deux plaignants. En revanche, la figure
caractérisation du personnage de nature statu- du juge se pare d’un costume rappelant ceux des
taire dans l’ordre du social ou du récit. Il souligne cadis dans les Maqāmāt (« Séances ») d’al-Ḥarīrī de
en revanche la position de force brute que le per- la dite « école de Bagdad » (voir note 6), des philo-
sonnage accoutré en Mongol exerce dans la scène sophes grecs de la copie syrienne d’al-Mubashshir
(Figs. 10.12 et 10.13). (m. 1087 ?) datant de la première moitié du
L’histoire suivante (f.12) raconte comment un xiiie siècle42, ou encore le frontispice des Rasāʾil
marchand loue les services d’un artisan afin qu’il Ikhwān al-Ṣafāʾ 43 (Fig. 10.14).
fore des perles ; mais, trouvant des cymbales (ou Un bilan provisoire de ces apparitions autorise
une harpe, suivant les versions), il demande à l’ar- difficilement des interprétations tranchées : en
tisan d’en jouer toute la journée. Le soir venu, l’ar- dehors du cas du souverain (et de son épouse),
tisan demande à être payé (Fig. 10.4). les représentations de Mongols ne semblent
L’image montre, dans ce cas, les deux person- pas rigoureusement viser à des caractérisations
nages habillés à la mongole : mêmes bottes, cein- sociales, ethnographiques ou autres44. À l’inverse,
ture rouge avec sacoches, robe gris-bleue. L’artisan elles montrent parfois le même personnage tantôt
qui joue des cymbales porte le chapeau plat, alors en « Arabe/autochtone », tantôt en « Mongol », de
que le marchand est coiffé du bonnet à quatre façon quasiment interchangeable. Il est peu pro-
pans (le seul de ce type dans notre manuscrit, bable pour autant que le hasard de l’inspiration
comme signalé plus haut). Les basques de son du peintre ait seul guidé le choix de telle ou telle
manteau de dessus sont relevées et coincées par mise. L’idée d’une pleine conscience du ou des
une ceinture noire à plaques de métal doré, sans auteurs de la portée de leurs choix se renforce d’un
doute pour faciliter ses pas de danse. Pourtant, il constat : les tenants effectifs du pouvoir (le roi et la
paraît difficile d’affirmer que les petites disparités
42 Mukhtār al-Ḥikam wa Maḥāsin al-Kalim d’al-
Mubashshir, Syrie, première moitié du xiiie s., Topkapı
41 Dans l’histoire des « Deux Perroquets et la Femme du Sarayı Library, Ahmet III 3206.
Satrape », la première image (f.63) montre à gauche 43 Pour des reproductions de ces œuvres, voir Etting
un « Arabe » devant la cage où se trouvent les perro- hausen, La peinture arabe, 75‑77, 99, 106‑107.
quets et, à droite, deux personnages dont au moins l’un 44 On notera que notre avis est en cela plus nuancé que
en « Mongol » ; voir Ibn al-Muqaffaʿ, Le Livre de Kalila celui de Marianne Barrucand, qui précise : « Nous
et Dimna, 120-22. L’homme de gauche pourrait bien constatons que les vêtements dans ce manuscrit
être le fauconnier. L’image suivante (63v), illustrant caractérisent assez clairement différentes couches
le moment où le fauconnier se fait crever les yeux, sociales » : Barrucand, « Le Kalila wa Dimna de la
montre celui-ci habillé en Mongol. Bibliothèque royale de Rabat », 28.
316 Porter et Castinel
figure 10.12 Fable de l’amant surpris par le mari, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.10v
Simple exotisme ou contestation subliminale 317
figure 10.13 Fable de l’amant surpris par le mari, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.11
318 Porter et Castinel
figure 10.14 Fable du qadi de Merv, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.14
Simple exotisme ou contestation subliminale 319
reine) sont systématiquement vêtus en Mongols, (m. 1304), n’avaient pas encore officiellement
tout comme le cadi est toujours en autochtone. adopté l’Islam. Cette conversion permit du reste
Ces invariants, autour desquels les attributs et les d’amorcer un processus de réconciliation entre les
fonctions se rejoignent, fournissent une matrice élites mongoles et leurs sujets, majoritairement
de compréhension, respectueuse à leur niveau des persans46. C’est probablement au génie politique
statuts qui sont les leurs. Les mutations et incohé- de certains vizirs et conseillers que l’on doit en
rences apparentes des autres personnages dans partie ce retournement.
leur habillement s’inscrivent alors dans un mou- Toute une série de scènes agrémentant notre
vement qui questionne la pertinence des statuts manuscrit opposent les deux types « ethnolin-
pour les dégrader au rang de position contingente. guistiques » Mongol/Arabe (ou autochtone). À ce
La porte s’ouvre à une critique politique ou morale. stade, il est d’ailleurs important de souligner que
Même s’il faut rester prudent en l’absence le monde strictement « persan » semble complète-
d’autres corpus d’images de la même période ment absent de ce volume47. Ce constat irait dans
validant un mouvement général, la piste d’une le sens d’une commande émanant d’un membre
stigmatisation circonstancielle d’une situation de des élites arabophones de Bagdad, et positionne
pouvoir pour le personnage « Mongol », éventuel- en tout état de cause le manuscrit dans une rela-
lement opposé à la légitimité d’une autorité du tion entre Mongols et Arabes48.
personnage « Arabe », semble toutefois faire sens. Cette opposition entre types « mongol » et,
L’apparente « schizophrénie » engendrée par les dans le cas qui suit, « arabo-persan » (voir note
modulations vestimentaires d’un même person- 29) est illustrée sur une exceptionnelle tapisserie
nage ne laisse pour autant pas d’interroger le lec- multicolore conservée à la David Collection de
teur ; un autre niveau de lecture nous permet de Copenhague49. Là, cependant, les personnages
questionner cette nouvelle dimension.
46 Clifford E. Bosworth, The Islamic Dynasties: A
Chronological and Genealogical Handbook (Edinburgh :
4 Pouvoir mongol versus autorité Edinburgh University Press, 1967), 150.
légitime ? 47 On connaît mal la peinture persane antérieure aux
années 1300, époque des « Petits Shāh-nāmā mon-
gols », par exemple, mais ce qu’on en distingue montre
Nous avons précédemment laissé entendre qu’en
un style bien différent de celui qui apparaît dans ce
dépit d’une différence « hiérarchique », notam- manuscrit de Rabat, comme du reste dans les copies
ment visible dans les scènes faisant intervenir un « irakiennes » des Maqāmāt datant d’avant la conquête
roi et un vizir, le souverain « Mongol » représen- mongole, ainsi que dans la copie datée et située à
terait logiquement la figure du pouvoir en place Bagdad des Ikhwān al-Ṣafāʾ citée plus haut. Du reste,
d’après certains chercheurs, ce serait bien à Bagdad
à Bagdad dans le dernier quart du xiiie siècle, qu’auraient été produits au moins certains des petits
tandis que le vizir habillé en « Arabe » illustrerait Shāh-nāmā mongols ; ceux-ci démontreraient ainsi
peut-être une sorte d’autorité « légitime »45. un changement radical dans le style jusqu’ici ratta-
En effet, la question de la légitimité des souve- ché à l’ancienne capitale des califes. Pour les petits
Shāh-nāmā, l’ouvrage de référence reste celui de
rains mongols, fossoyeurs du califat abbasside et
Marianna S. Simpson, The Illustration of an Epic: The
jusqu’alors ennemis de l’Islam est encore, à cette Earliest Shahnama Manuscripts (New York : Garland
époque, très durement ressentie par les popula- Pub., 1979).
tions locales. Il faut rappeler que les souverains 48 Rappelons qu’une copie de la traduction persane d’Abū
il-khanides, jusqu’à la conversion de Ghazan Khan al-Maʿāli Naṣrallāh, datant des premières décennies du
xiiie siècle et située à Bagdad, est conservée à Istanbul
(Topkapı Sarayı Library, Hazine 363).
45 Ce cas de figure ne se vérifie toutefois pas sur le frontis- 49 Kjeld von Folsach, Torben Lundbæk et Peder
pice du Juvaynī de la BnF, mentionné plus haut. Mortensen, éd., Sultan, Shah, and Great Mughal: the
320 Porter et Castinel
sont disposés de telle sorte dans la composition illustrée par la peinture du folio 111v, représentant
qu’ils apparaissent clairement hiérarchisés : un le saint homme (à gauche, en Arabe) et son hôte52
souverain mongol couronné trône sous un dais, en (Fig. 10.5). Cette grille de lecture peut-elle être
position centrale ; ses cheveux sont noirs, comme appliquée à d’autres images ? Nous avons évoqué
sa moustache et sa barbiche, indiquant qu’il est plus haut certains épisodes troublants, comme
dans la force de l’âge. À sa gauche, un personnage l’histoire de l’amant battu par le mari (f.10v-11) ;
barbu grisonnant et portant turban est assis sur dans un premier temps, l’amant, imberbe, est
un tabouret, tandis qu’à sa droite se tient debout habillé en Mongol, peut-être pour figurer un bour-
un jeune Mongol imberbe. L’image évoque certai- reau des cœurs ou un transgresseur du bon droit.
nement le cas de plusieurs sultans il-khanides qui Puis, la page d’après, il serait un homme victime
ont été secondés par des vizirs persans (Juvaynī, d’un châtiment, le mari évoquant dans l’exercice
Naṣīr al-Dīn al-Ṭūsī, Rashīd al-Dīn, par exemple)50. concret de sa puissance brute l’oppresseur mongol.
Nous avons évoqué le cas le plus flagrant, sous Dans le cas des personnages isolés ou secon-
la forme du souverain et de son vizir. Cependant, daires habillés en Mongols, on pourrait déceler
d’autres cas de figure iraient également dans ce une certaine idée de soumission au pouvoir mon-
sens : il s’agit notamment des représentations gol en place ; à l’inverse, le personnage « autoch-
du cadi. tone » incarnerait alors une sorte de résistance
Nous avons vu, au folio 14 le jugement du cadi passive, nourrie de légitimité.
de Marv ; la figure d’autorité qu’il incarne, rap- De toute évidence, le costume à lui seul pré-
pelant la représentation de cette charge dans les sente plusieurs niveaux de lectures, avec des déca-
Maqāmāt d’al-Ḥarīrī de ladite « école de Bagdad », lages qui ne sont sans doute pas toujours aisés
révèle l’association autorité/costume autochtone. à percevoir : le pouvoir, l’oppression, l’étranger,
Il semble évident, en contexte musulman, que la s’opposeraient alors à l’autorité légitime locale.
fonction de cadi soit exercée par un coreligion- Cette « division » est peut-être d’autant plus mar-
naire, au fait des subtilités de la charia, bien diffé- quée que tous les « Arabes » ne sont pas au même
rente en cela des codes juridiques mongols (yasa). niveau ; ils ne se valent pas, ceux au chapeau pointu
Le personnage du juge se retrouve également étant clairement socialement « en-dessous » de
dans l’histoire de la femme du cordonnier (f.29), ceux portant turban.
dans laquelle les trois personnages sont habillés à D’autres images sont indubitablement plus
l’« arabe ». De même, dans la conclusion de l’his- « neutres », du moins en apparence. Ainsi en est-il
toire du fourbe et du benêt (f.47v), on remarque de quelques étoiles décorées au lustre métallique.
que le cadi est monté sur un mulet, ce qui le place L’une d’elles, conservée au Musée Grobet-Labadié
au-dessus des autres personnages. de Marseille, illustre deux personnages imberbes,
Il est possible que cette même notion d’autorité51 assis au même niveau et donc d’égale « impor-
– plus « spirituelle » dans ce cas – soit aussi tance » qui se font face. L’un est habillé à la mon-
gole (à gauche), tandis que l’autre porte un costume
History and Culture of the Islamic World (Copenhague : « arabo-persan ». Sans « légende » pour en saisir
National Museum, 1996), no 133, 174-175.
le contenu iconographique, cette représenta-
50 Sur la question des rapports entre souverains mongols
et vizirs persans, voir aussi Assadulah S. Melikian- tion pourrait évoquer une sorte de conversation
Chirvani, « Le Livre des rois, miroir du destin », Studia
Iranica 17 (1988) : 7-46 ; Assadulah S. Melikian-Chirvani,
« Le Livre des rois, miroir du destin. II. Takht-e
Soleymân et la symbolique du Shâh-Nâme », Studia celui du Dioscorides daté 626H/1229 de la biblio-
Iranica 20 (1991) : 33-148. thèque d’Ahmet III, folio 2v ; Ettinghausen, La peinture
51 Difficile de ne pas évoquer d’autres « images d’auto- arabe, 71.
rité », en l’occurrence les « portraits d’auteur », dont 52 Ibn al-Muqaffaʿ, Le Livre de Kalila et Dimna, 253.
Simple exotisme ou contestation subliminale 321
figure 10.15 Fable du saint homme et son hôte, Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Rabat, BRR, ms. 3655, f.111v
322 Porter et Castinel
transculturelle53. Cependant, ici, un quatrain de Akbar en 1596 dévoile toute une série de peintures
Jalāl al-Dīn Rūmī (m. 1273) évoque l’amour dévo- illustrant les glorieux ancêtres ; une page conser-
rant de l’un pour l’autre ; si le Mongol est bien celui vée à la Freer Gallery de Washington, peinte par
qui a des yeux plus beaux que ceux des gazelles Kesu Kālān et Kamāli Chela, nous présente ainsi
et des boucles qui enchaînent les lions, il est alors Kubilai Khan et son épouse arborant tous deux
l’objet de l’amour de l’autre54. des coiffures dérivées des modèles mongols vus
plus haut57. Cependant, et comme on pourrait s’y
attendre, cette image s’affadit avec le temps. Les
5 Conclusion connotations qu’elle véhicule voient s’émousser la
rigueur descriptive au profit d’un cadre graphique
Comme on l’a évoqué plus haut, la figure du plus allusif et donc susceptible d’incarner des
Mongol, très identifiable, s’établit certes dans la messages déconnectés de toute expérience histo-
longue durée dans l’Orient musulman, mais se rique. Du reste, ce phénomène dépasse le cadre
fige, après leur disparition, dans une iconographie de l’Orient musulman puisqu’on le retrouve éga-
historique quasiment mnémotechnique – à la lement dans l’Occident médiéval. En effet, les arts
façon des tableaux dynastiques des empereurs de de l’Europe du Moyen-Âge s’émaillent de quelques
Chine apparaissant dans la copie de 1314 des Jāmiʿ apparitions de Mongols58.
al-tavārīkh de Rashīd al-Dīn55. Ainsi, leur présence Un exemple dans une fresque italienne de la
se décèle encore dans des images datant de la première moitié du xive siècle est notamment l’oc-
période timouride ou issues de l’Inde moghole de casion, pour Michele Bernardini, de faire un bilan
la toute fin du xvie siècle. Remarquons au passage de l’iconographie du Mongol59. Dans la fresque de
que dans les deux cas (Timourides et Moghols), il la Crucifixion du Sacro Speco de Subiaco, celui-ci
s’agit de descendants de Tamerlan, qui revendi- apparaît en effet parmi les « Romains » qui se
quait lui-même, du moins en partie, une ascen-
dance mongole. Cette iconographie figée serait 57 Reproduit dans Amina Taha-Hussein Okada, Le Grand
alors, à son tour, une manière de faire connaître Moghol et ses peintres : miniaturistes de l’Inde aux XVIe
cette revendication. et XVIIe siècles (Paris : Flammarion, 1992), no 101, 98.
Ainsi, les illustrations du manuscrit Supplément Pour un autre exemple, voir Déploration de la mort
d’Abaqa Khan, par Mukund et Banwâri Kālān, page du
persan 1113 des Jāmiʿ al-Tavārīkh, datant de c. 1430-
même manuscrit dispersé, Inde moghole, 1596, dans
34, « sont étroitement inspirées d’un manuscrit la collection S. Aga Khan (M.220) ; reproduit dans
réalisé un siècle plus tôt »56. Dans l’Inde moghole, Sheila R. Canby, Princes, Poètes & Paladins : miniatures
un exemplaire du Livre de Gengis Khan conçu pour islamiques et indiennes de la collection du prince et de la
princesse Sadruddin Aga Khan (Genève : Musée d’art et
d’histoire, 1999), no 90, 122.
53 Aurélie Bosc et Mireille Jacotin, éd., Le Goût de l’Orient : 58 C’est du reste à la diplomatie du pape Innocent IV puis
collections & collectionneurs de Provence (Milan : de Louis IX que l’on doit des tentatives de rapproche-
Silvana, 2013), 253. ment avec les Grands Khans, qui auront des consé-
54 Abdallah Qouchani, Ashʿār-i fārsī-i kāshīhā-yi Takht-i quences désastreuses. Une première mission est celle
Sulaymān (Téhéran : Markaz-i našr-i dānišgāhī, 1992), de Jean de Plan Carpin (1246-1247), suivie plus tard
no 48, 67. Ey gorosne-ye mehr-e to seyrân-e jahân/Tarsân par celle de Guillaume de Rubrouck (1253-1254) ; voir
ze farâq-e to dalirân-e jahân/bâ cheshm-e to ahu’ân che Jean Richard, La Papauté et les Missions d’Orient au
dârand be dast/Ey zolf-e to pây-band-e shirân-e jahân. Moyen Âge, XIIIe-XVe siècles (Rome : École française de
55 Voir L’Empereur Xuandi de la dynastie des Liang posté- Rome, 1977).
rieurs, Londres, collection Khalili, MSS 727, folio 14b ; 59 Michele Bernardini, « Un mongolo nella “Crocifissione”
reproduit dans Michael Rogers, éd., Arts de l’islam : trecentesca del Sacro Speco di Subiaco », in Studi in
chefs-d’œuvre de la collection Khalili (Paris : Hazan, Onore di Umberto Scerrato per il suo settantacinquesimo
2009), 149. compleanno, éd. Maria V. Fontana et Bruno Genito
56 Richard, Splendeurs persanes, no 40, 76. (Naples : Università degli Studi di Napoli, 2003), 77-103.
Simple exotisme ou contestation subliminale 323
partagent les dépouilles du Christ. Cette figure éventuellement dans une distanciation par rap-
négative n’est du reste pas isolée, puisqu’on la port au récit stricto sensu.
retrouve dans d’autres peintures60. Pourtant, Même si aucun commanditaire n’est identifié,
quelques décennies plus tard, l’iconographie occi- il reste probable qu’un tel ouvrage s’adressait à un
dentale du Mongol prend un tour plus positif, ou membre de l’élite baghdādī, réceptive à la critique
du moins plus neutre bien qu’exotique, dans l’illus- politique ou morale que porte le texte comme ses
tration du Livre des Merveilles de Marco Polo, par illustrations. C’est, au-delà de ce destinataire, au
exemple61. lecteur que s’adresse une interrogation identitaire :
Ces glissements de champs de sens remontent Qui suis-je, dans l’ordre moral autant que dans
comme par capillarité à l’époque contempo- l’ordre politique, moi qui contemple ces images ?
raine. Le film de Parviz Kimiavi, Moghôl-hâ, (Les Quelle est ma place et quelles sont mes valeurs en ce
Mongols, 1973), long-métrage de fiction, réfléchit bas monde agité d’incertitudes63 ?
sur la modernité et l’impérialisme culturel qui Les processus d’identification et de distancia-
règnent dans l’Iran des années 1970. Entre rêve et tion du lecteur à l’égard des différents personnages
réalité, il pose des questions sur l’identité et l’an- de ces histoires sont autant de cheminements
xiété que produisent des avancées technologiques possibles pour accéder à une ou plusieurs lectures
telles que la télévision, présente y compris dans du monde en transcendant la situation qui est la
de petits hameaux en bordure de désert. Tentant sienne. Il s’agit notamment de s’élever du particu-
une sorte de reconstitution historique, le metteur lier au général. On passe alors de l’archétype, com-
en scène (dans le film, joué par Kimiavi lui-même) pris comme la cristallisation d’un personnage dans
emploie des acteurs turkmènes pour jouer des ses caractéristiques descriptives, vers le stéréotype,
Mongols. Le décalage entre ces personnages à élaboration d’une matrice plus souple à valeur de
l’allure de braves paysans et ce que le metteur référence indicative et susceptible d’accueillir des
en scène leur demande de terrifiante sauvagerie valeurs circonstancielles. En synthèse, les figures
« mongole » fait vivre jusqu’au cocasse les dis- mongoles du manuscrit de Rabat illustrent l’idée
tances qui finissent par s’instaurer entre la forme de la mutabilité, parfois jusqu’à l’évanescence, des
signifiante et les réalités signifiées62. formes descriptives favorisant la présentification,
Le message, au final, que porte notre manus- dans un contexte politico-culturel donné, des
crit et que déclinent les exemples cités ci-dessus, valeurs auxquelles les fables de Kalīla wa-Dimna
touche à la problématique des apparences telles donnent à réfléchir.
qu’elles masquent ou révèlent le fond, la réalité.
Dans un contexte historique écrasé par les vicis-
situdes du Destin, le genre littéraire que consti- Bibliographie
tue la fable se prête à des tours de passe-passe
entre les personnages, et les illustrations jouent, Barrucand, Marianne. « Le Kalila wa Dimna de la
dans le registre de l’image, de ces mécanismes, Bibliothèque royale de Rabat : un manuscrit illustré
il-khanide ». Revue des Études islamiques, numéro
spécial en l’honneur de Dominique Sourdel, n°54 of Sciences. Tashkent : Gafur Gulom Nomidagi
(1986) : 17-51. Adabiët va San”at nashriëti, 1980.
Barrucand, Marianne. « Un manuscrit arabe illustré Kadoi, Yuka. « The Mongols Enthroned ». In The Diez
de Kalila wa Dimna du XIIIe siècle et sa copie otto- Albums. Contexts and Contents. Édité par Julia
mane ». Archéologie islamique, no 2 (1991) : 81-95. Gonnella, Friederike Weis et Christoph Rauch, 243-
Bernardini, Michele. « Un mongolo nella “Crocifissione” 275. Leiden : Brill, 2017.
trecentesca del Sacro Speco di Subiaco ». In Studi Komaroff, Linda et Stefano Carboni, éd. The Legacy of
in Onore di Umberto Scerrato: per il suo settantacin- Genghis Khan: Courtly Art and Culture in Western
quesimo compleanno. Édité par Maria V. Fontana et Asia, 1256-1353. New York : Metropolitan Museum of
Bruno Genito, vol. 1 : 77-103. Naples : Università degli Art, 2002.
Studi di Napoli, 2003. Melikian-Chirvani, Assadulah S. « Le Livre des rois,
Blair, Sheila. « A Mongol Envoy ». In The Iconography of miroir du destin ». Studia Iranica 17 (1988) : 7-46.
Islamic art: Studies in Honour of Robert Hillenbrand. Melikian-Chirvani, Assadulah S. « Le Livre des rois,
Édité par Bernard O’Kane, 45-60. Edinburgh : miroir du destin. II. Takht-e Soleymân et la sym-
Edinburgh University Press, 2005. bolique du Shâh-Nâme. » Studia Iranica 20 (1991) :
Bosc, Aurélie et Mireille Jacotin, éd. Le Goût de l’Orient : 33-148.
collections & collectionneurs de Provence. Milan : O’Kane, Bernard. « The Uses of Captions in Medieval
Silvana, 2013. Literary Arabic Manuscripts ». In Arab Painting: Text
Bosworth, C. Edmund. The Islamic Dynasties: A Chrono- and Image in Illustrated Arabic Manuscripts. Édité
logical and Genealogical Handbook. Edinburgh : par Anna Contadini, 135-44. Leiden : Brill, 2007.
Edinburgh University Press, 1967. Qouchani, Abdallah. Ashʿār-i fārsī-i kāshīhā-yi Takht-i
Canby, Sheila R. Princes, poètes & paladins : miniatures Sulaymān. Téhéran : Markaz-i našr-i dānišgāhī, 1992.
islamiques et indiennes de la collection du prince et Richard, Francis. Splendeurs persanes : manuscrits du
de la princesse Sadruddin Aga Khan. Genève : Musée XIIe au XVIIe siècle. Paris : Bibliothèque nationale de
d’art et d’histoire, 1999. France, 1997.
Ettinghausen, Richard. La peinture arabe. Genève : Richard, Jean. La papauté et les missions d’Orient au
Skira, 1962. Moyen Âge, XIIIe-XVe siècles. Rome : École française
Folsach, Kjeld von, Torben Lundbæk, et Peder de Rome, 1977.
Mortensen, éd. Sultan, Shah, and Great Mughal: Rogers, Michael, éd. Arts de l’islam : chefs-d’œuvre de la
The History and Culture of the Islamic World. collection Khalili. Paris : Hazan, 2009.
Copenhague : National Museum, 1996. Roux, Jean-Paul. « Le bonnet et la ceinture ». Turcica, no
Grabar, Oleg. « What Does “Arab Painting” Mean? ». In 7 (1975) : 50-64.
Arab Painting: Text and Image in Illustrated Arabic Simpson, Marianna S. The Illustration of an Epic: the
Manuscripts. Édité par Anna Contadini, 17-22. Earliest Shahnama Manuscripts. New York : Garland
Leiden : Brill, 2007. Pub., 1979.
Grube, Ernst J. « Prolegomena for a Corpus Publication Soucek, Priscilla. « Ethnicity in the Islamic Figural
of Illustrated Kalilah wa Dimna Manuscripts ». Tradition: The Case of the ‘Turk’ ». Tārīḫ 2 (1992) :
Islamic Art 4 (1990-91), 301-481. 73-103.
Ibn al-Muqaffaʿ, ʿAbd Allāh. Le Livre de Kalila et Dimna. Taha-Hussein Okada, Amina. Le Grand Moghol et ses
Traduit par André Miquel. Paris : Orients-Klincksieck, peintres : miniaturistes de l’Inde aux XVIe et XVIIe
2012. siècles. Paris : Flammarion, 1992.
Ismailova, El’mira M. Oriental Miniatures of Abu Raihon
Beruni Institute of Orientology of the UzSSR Academy
11
Entre deux mondes : une anthologie illustrée inédite inspirée de
Kalīla wa-Dimna
Aïda El Khiari et Nathalie Buisson, Frantz Chaigne, Françoise Cuisance, Rajana Fatima
Amalarajah, Hoa Perriguey, Valérie Saurel
quelques études récentes consacrées à la culture de 1516-1517 parachève de rapprocher ces uni-
matérielle contribuent, fort heureusement, à lever vers qui vont s’enchevêtrer tout particulièrement
le voile sur le paysage artistique de l’Égypte et de la durant la première moitié du xvie siècle9.
Syrie ottomanes et soulignent les transformations
qui se font jour dans les arts et l’architecture5. La
première moitié du xvie siècle en particulier est 2 Le manuscrit de la collection S.
une période fascinante qui marque l’apogée des
processus de collusion artistique et culturelle Un manuscrit, conservé dans une collection par-
entamés un siècle auparavant entre les domaines ticulière, offre l’occasion de revenir sur les consé-
mamlouk, ottoman et persan6. On trouve en effet quences de la provincialisation de l’Égypte et de
de nombreuses traces d’interactions artistiques la Syrie dans le domaine des arts du livre10. Copié
entre les mondes mamluk et ottoman dès le xive en 943H/1537 de notre ère, très vraisemblable-
siècle7 et ces interactions se poursuivent et s’ac- ment à Damas, par Badr al-Dīn Muḥammad b.
croissent le siècle suivant8. La conquête ottomane al-Ghazzī al-ʿĀmirī (m. 984H/1577), cette antho-
logie poétique richement illustrée renferme trois
materiały 5. Ogólnopolskiej Konferencji Arabistycznej,
textes arabes en vers inspirés de Kalīla wa-Dimna
Poznań, 9-10 czerwca 1997, éd. Henryk Jankowski (Poznań : d’Ibn al-Muqaffaʿ (m. 139-142H/756-759 ?) : Miftāḥ
Uniwersytet im. Adama Mickiewicza w Poznaniu. Katedra al-fiṭna fī naẓm Kalīla wa-Dimna de Jalāl
Orientalistyki i Bałtologii, 1998), 177‑83. al-Dīn al-Naqqāsh11 (m. xe/xve siècle), al-Ṣādiḥ
5 En ce qui concerne la mesure de cette transition dans
l’ordre urbain et les arts mobiliers, on se référera aux tra-
vaux suivants : Doris Behrens-Abouseif, Egypt’s Adjust- The Arts of the Mamluks in Egypt and Syria: evolution
ment to Ottoman Rule: Institutions, Waqf and Architecture and impact, éd. Doris Behrens-Abouseif (Göttingen :
in Cairo (16th and 17th Centuries) (Leiden : Brill, 1994) ; Bonn University Press, 2012), 267‑81.
Çiğdem Kafescioǧlu, « “In The Image of Rūm”: Ottoman 9 La première moitié du xvie siècle, période de transi-
Architectural Patronage in Sixteenth-Century Aleppo and tion jusqu’alors peu étudiée, fait l’objet d’une attention
Damascus », Muqarnas 16 (1999) : 70‑96 ; Doris Behrens- nouvelle ces dernières années. Voir à ce sujet les tra-
Abouseif, « The Ottoman Conquest of Egypt and the Arts », vaux de Lellouch, Les Ottomans en Égypte : historiens
in Conquête ottomane de l’Égypte (1517), 303‑26 ; Julien Loi- et conquérants au XVIe siècle (Paris : Peeters, 2006) ;
seau, « La ville démobilisée : Ordre urbain et fabrique de la Benjamin Lellouch et Nicolas Michel éd., Conquête
ville au Caire avant et après 1517 », in Conquête ottomane de ottomane de l’Égypte (1517) : arrière-plan, impact, échos
l’Égypte (1517), 269‑84. (Leiden : Brill, 2013).
6 Au sujet des relations entre l’empire ottoman et le sul- 10 Dans le cadre du programme de recherche Kalīla
tanat mamlouk, voir les travaux de Cihan Yüksel Muslu, wa-Dimna, le manuscrit a pu être étudié grâce à sa
« Ottoman-Mamluk Relations and the Complex Image mise en dépôt à la Bibliothèque nationale de France.
of Bāyezīd II », in Conquête Ottomane de l’Égypte (1517), Durant l’année 2015/2016, de nombreux spécialistes
49‑76 ; Yüksel Muslu, The Ottomans and the Mamluks: issus du monde de la recherche, des bibliothèques et
Imperial diplomacy and warfare in the Islamic world des musées, des restaurateurs indépendants ainsi que
(Londres : I.B. Tauris, 2014) ; Yüksel Muslu, « Patterns of des étudiants en Master et en doctorat ont collaboré
Mobility between Ottoman and Mamluk Lands », in The lors de séminaires mensuels sous la direction d’Éloïse
Mamluk Sultanate from the Perspective of Regional and Brac de la Perrière et Annie Vernay-Nouri et coor-
World History, éd. Reuven Amitai et Stephan Conermann donnés par Aïda El Khiari. Nous avons également eu
(Göttingen : V&R unipress, 2019), 391‑430. la chance de bénéficier de l’avis de plusieurs spécia-
7 Zeren Tanındı, « The Arts of the Book: Patrons and listes et chercheurs parmi lesquels Francis Richard,
Interactions in Erzincan between 1365 and 1410 », in Béatrice Gruendler, Nourane Ben Azzouna, Elaine
At the Crossroads of Empires: 14th-15th Century Eastern Wright, Jean-Louis Estève, Aurélia Stréri et Ségolène
Anatolia. Proceedings of the International Symposium Held Walle.
in Istanbul, 4th-6th May 2007 (Istanbul : Institut Français 11 Jalāl al-Dīn al-Ḥasan b. Aḥmad al-Naqqāsh (parfois
d’Études Anatoliennes, 2012), 221‑38. appelé Ibn al-Iṣfahānī) est un écrivain baghdādī quasi-
8 Zeren Tanındı, « Two Bibliophile Mamluk Emirs: Qansuh ment inconnu, dont deux œuvres seulement nous ont
the Master of the Stables and Yashbak the Secretary », in été transmises : la versification Miftāḥ al-fiṭna fī naẓm
Entre deux mondes 327
wa-l-bāghim d’Ibn al-Habbāriyya12 (m. 509H/ Le codex apparaît en 1981 dans une vente aux
1115-6 ?) et enfin al-Laṭāʾim wa-l-ashnāf d’Ibn enchères organisé à New-York par Hans Peter
Makānis13 (m. 794H/1392). Le manuscrit est un Kraus, célèbre libraire et collectionneur de livres
objet unique tant dans l’association de ces trois anciens14. La notice du catalogue de vente le
textes que par ses peintures qui sont le seul exemple décrit en des termes élogieux mais partiellement
conservé de ces textes illustrés. Il s’agit également inexacts : ainsi seuls deux des trois textes qui com-
du premier manuscrit daté d’après la conquête posent le manuscrit sont identifiés et l’un d’entre
de Selīm Ier pouvant être attribué à un peintre eux est attribué au mauvais auteur15. En outre, la
ottoman exerçant dans les anciens territoires notice détaillée, disponible sur demande, précise
mamlouks. la date erronée de 843H/1439. L’examen minutieux
du manuscrit mené dans le cadre de cette étude
a en effet révélé que sa structure matérielle a été
Kalīla wa-Dimna, rédigée en 828H/1425 et une seconde entièrement remaniée au cours du xxe siècle16
versification intitulée Tashmīs al-budūr wa-taḫmīs et que les dates des colophons ont été altérées
al-shuḏūr composée en 816H/1413. Voir au sujet de la afin d’antédater le manuscrit d’un siècle : on y lit
versification : Martin Sprengling, « Kalīla Studies »,
American Journal of Semitic Languages and Literatures,
désormais 843H/1439 en place de 943H/1537. Une
no 40 (1924) : 82 ; Edward D. Ross, « An Arabic and a décennie plus tard, Ernst Grube publie une courte
Persian metrical version of Burzoe’s autobiography description du codex dans son « Prolegomena » et
from “Kalila and Dimna” », Bulletin of the School of s’appuyant sur des éléments qu’il n’explicite pas, le
Oriental and African Studies 4, no 3 (1927) : 441‑72 ; Carl
Brockelmann, « Kalīla wa-Dimna », in Encyclopaedia of
Islam, Second Edition, éd. Peri Bearman et al., consulté
en ligne le 22 novembre 2018. 14 Hans P. Kraus, Illuminated Manuscripts from the
12 Poète arabe de l’époque seldjoukide, Al-Sharīf Abū Eleventh to the Eighteenth Centuries (New York : H.P.
Yaʿlā Muḥammad ibn Muḥammad al-Hāshimī al-ʿAb- Kraus, 1981), 83.
bāsī (connu sous le nom d’Ibn al-Habbāriyya) naquit 15 La notice donne le Miftāḥ al-fiṭna fī naẓm Kalīla
et étudia à Bagdad (selon toute vraisemblance). Il est wa-Dimna et Al-Ṣādiḥ wa-l-bāghim, tous deux étant –
l’auteur, entre autres, de deux versifications de Kalīla toujours d’après la notice – d’Ibn al-Habbāriyya. Nous
wa-Dimna : Al-Ṣādiḥ wa-l-bāghim et Natāʾij al-fiṭna fī tenons à remercier vivement Samer Ben Brahim pour
naẓm Kalīla wa-Dimna. Voir à son sujet : Aḥmad ibn son aide dans l’identification du texte de Jalāl al-Dīn
Muḥammad Šams al-Dīn Ibn Khallikān, Ibn Khallikan’s al-Ḥasan b. Aḥmad al-Naqqāsh.
Biographical Dictionary, trad. par Mac Guckin De 16 S’il est possible de situer la réfection du manuscrit
Slane (Paris : Printed for the Oriental translation fund entre les années 1940 et 1980, rien n’indique que la
of Great Britain and Ireland, 1843), vol. 3, 155 ; Nefeli modification des colophons ait eu lieu au même
Papoutsakis, « Ibn al-Habbāriyya », in Encyclopaedia of moment. De nombreux exemples de manuscrits dont
Islam, THREE, éd. Kate Fleet, Gudrun Krämer, Denis les colophons ont été modifiés peu de temps après leur
Matringe, John Nawas, Everett Rowson, consulté en copie sont attestés et la littérature arabe médiévale
ligne le 2 décembre 2018. regorge de récits de faussaires, l’un des plus fameux
13 Poète de l’époque mamlouke, Fakhr al-Dīn ibn ʿAbd étant peut-être celui du calligraphe Ibn al-Bawwāb se
al-Rahmān ibn ʿAbd al-Razzāq al-Qibṭī al-Ḫanafī félicitant d’avoir réussi, sans éveiller le moindre soup-
al-Qāhirī (connu sous le nom d’Ibn Makānis) naquit au çon, à reproduire un juz’ manquant d’un Coran d’Ibn
Caire en 1344. Il fut nommé au vizirat de Damas en 1376 al-Muqla. Voir à ce sujet, Abolala Soudavar, « The
puis au vizirat d’Égypte et mourut empoisonné sur la Concepts of ‘al-Aqdamo Aṣaḥḥ’ and ‘Yaqin-e Sābeq’,
route qui l’y menait en 1392. Il est l’auteur, entre autres, and the Problem of Semi-fakes », Studia Iranica 28,
de l’ouvrage Al-Laṭāʾim wa-l-ashnāf. Voir à son sujet : no 2 (1999) : 255‑73 ; Francis Richard, « Forgeries IV. Of
Ali Hayder, « Ibn Makanis (sa vie, son œuvre) sous les Islamic Manuscripts », in Encyclopædia Iranica, 2011,
Mamlouks » (Thèse de doctorat, Bordeaux, Université http://www.iranicaonline.org/articles/forgeries-iv,
Bordeaux 3, 1981) ; Geert J. van Gelder, éd., Classical consulté en ligne le 20 avril 2020 ; Jan J. Witkam,
Arabic Literature: A Library of Arabic Literature « Increasing the Value or Hiding the Defects? The Case
Anthology (New York : New York University Press, 2013), of the Stuttgart Hātifī Manuscript », Journal of Islamic
102. Manuscripts 8, no 2 (2017) : 240‑53.
328 El Khiari et al.
rattache à Basra à la fin du xve siècle17. À l’excep- texte n’est quasiment pas vocalisé ; lorsqu’il l’est,
tion de la brève notice de Grube, le manuscrit n’a seule la lettre de rime reçoit un signe de vocalisa-
fait jusqu’alors l’objet d’aucune étude. tion. La plupart des points diacritiques sont notés
et les signes orthoépiques incluent la notation –
non systématique – des hamza et des shadda.
3 Un objet codicologique complexe Texte et peintures ont été disposés sur un papier
oriental18 non filigrané de couleur crème, brillant,
Le manuscrit se présente sous la forme de deux très lisse et relativement épais19. Il compte sept
volumes : l’un comporte soixante-quatre feuillets, vergeures par centimètre environ, orientées per-
l’autre vingt-neuf. Ils sont tous deux richement illus- pendiculairement à la couture et qui peuvent être
trés, le premier renfermant quatre-vingt-quatre rectilignes ou bien légèrement incurvées. Les fils de
peintures, le second quarante-six. Ils présentent chaines, parallèles à la couture, sont difficilement
une parfaite homogénéité au niveau du papier, de discernables et il est compliqué d’en déduire leur
la mise en page, de la graphie et des illustrations disposition20. Le papier a été enduit d’une charge
qui indique qu’ils constituaient très vraisembla- minérale blanche destinée à recevoir le texte et les
blement originellement un seul et même volume.
Le premier volume mesure 26,5 cm × 17,5 cm et 18 Au xvie siècle, les manuscrits copiés sur papier non
son épaisseur, mesurée sur le dos du manuscrit, est filigrané et sur papier européen existent en quanti-
de 3 cm. Il est recouvert d’une reliure orientale en tés sensiblement égales dans l’Empire ottoman. Au
plein maroquin rouge dont les plats sont incrus- xviie siècle, la concurrence européenne est telle qu’en
Égypte et en Syrie l’immense majorité des manuscrits
tés de papier doré et gaufré et qui évoque le décor est copiée sur papier filigrané. Toutefois, l’existence
des reliures persanes à grand décor. Le second de papiers orientaux non filigranés datés des xviie et
volume mesure 25 × 17 cm et 1,5 cm d’épaisseur. xviiie siècles laisse supposer l’existence d’une pro-
Sa reliure, faite de basane de couleur rouge brun duction locale mineure. L’inventaire d’un marchand
de papier égyptien daté de 1650 faisant référence à
est décorée sur ses plats de motifs en creux dorés un papier local (waraq baladī) corrobore cela. Voir à
et peints représentant un médaillon central en ce sujet François Déroche, éd., Manuel de codicologie
forme de mandorle et quatre écoinçons (Fig. 11.1). des manuscrits en écriture arabe (Paris : Bibliothèque
Les dimensions des feuillets sont de 25 × 17 cm. La nationale de France, 2000), 64 ; Helen Loveday, Islamic
Paper: A Study of the Ancient Craft (Londres : The Don
réglure est de vingt-neuf lignes par page et l’écri-
Baker Memorial Fund, 2001), 27 ; Terence Walz, « The
ture couvre une surface d’environ 20 × 12 cm. Les Paper Trade of Egypt and the Sudan in the Eighteenth
deux volumes ne gardent aucune trace de foliota- and Nineteenth Centuries and its Re-Export to the
tion originelle, de marques de cahiers ou de col- Bilād as-Sūdān », The Trans-Saharan Book Trade.
lation pas plus que de marques de possessions ou Manuscript Culture, Arabic Literacy and Intellectual
History in Muslim Africa, 8 (2010) : 72‑72.
de lecture. Quelques corrections et commentaires 19 L’apprêture du papier, le texte et la présence de nom-
en arabe situés dans la marge sont de la main du breuses illustrations rendent malaisé l’examen du
copiste. Le corps du texte est noté au moyen d’un papier et c’est essentiellement au niveau des marges
naskh cursif de petit module à l’encre noire. Le que l’on a pu en discerner quelques caractéristiques.
Les observations sur le papier ont été menées à l’œil nu
avec l’aide d’une table éclairante et d’une feuille lumi-
17 Ernst J. Grube, « Prolegomena for a Corpus Publication neuse. Le polissage et l’apprêture du papier sont très
of Illustrated Kalilah wa Dimna Manuscripts », Islamic usés ou lacunaires par endroits – notamment en gout-
Art 4 (1990-91), n°40, 391. Il le mentionne à nouveau tière – laissant découvrir l’aspect du papier brut, plutôt
très rapidement dans Grube, « Some Observations pelucheux, de couleur blanc légèrement rosé et mou.
Concerning the Ottoman Illustrated Manuscripts of La pâte à papier, assez homogène, comporte quelques
the Kalilah wa Dimnah: Ali Çelebi’s Humayun-Name », petits amas de fibres et brindilles.
in Milletlerarası Türk Sanatları Kongresi, vol. 2 (Ankara : 20 Les plus repérables sont le plus souvent groupées par 2
Kültür Bakanlığı, 1995), 195‑205. ou par 3, à une distance de 1,1/1,3 cm ou 1,8 /1,9 cm.
Entre deux mondes 329
décors. Les analyses physico-chimiques menées d’amidon de riz ou de blé24, les analyses conduites
sur le manuscrit ont détecté la présence de cal- sur les manuscrits de Kalīla wa-Dimna montrent
cium laissant supposer l’utilisation de craie ou de au contraire que l’usage du gypse ou de la craie
gypse en surface. Cet encollage de très faible épais- s’est maintenu au moins jusqu’au xviie siècle dans
seur a été retrouvé dans l’intégralité des manus- la zone syro-égyptienne.
crits de Kalīla wa-Dimna analysés par Nathalie
Buisson21 ainsi que dans un ensemble de manus-
crits sur parchemin et sur papier datant du xiie 4 Une réfection majeure au cours du
au xve siècle et provenant du Maghreb, d’Égypte xxe siècle
et de Syrie22. Si certains auteurs ont suggéré que
les recettes à base de gypse ou craie23 auraient été L’analyse codicologique détaillée du manuscrit
progressivement remplacées par l’encollage à base révèle que sa structure matérielle a été profondé-
ment modifiée et qu’il a fait l’objet d’une réfection
majeure au cours du xxe siècle. Celle-ci a consisté
21 Voir l’article 9 dans ce même volume. dans un premier temps à démonter la reliure et à
22 Les analyses menées sur ce groupe de manuscrits par défaire tous les cahiers. Les feuillets ont ensuite été
Bernard Guineau ont été publiées dans Déroche, éd.,
Manuel de codicologie des manuscrits en écriture arabe,
144‑67. 24 Déroche, éd., Manuel de codicologie des manuscrits en
23 Loveday, Islamic Paper, 42. écriture arabe, 59.
330 El Khiari et al.
découverte, peu avant la publication de cet article, reste du texte. Dans les deux cas, la couleur du
d’un feuillet isolé illustré conservé au Museum für papier autour de l’écriture du mot thamānimiʾa
Islamische Kunst de Berlin (I. 6940)29 et apparte- est plus foncée que sur le reste de la page, ce qui
nant sans nul doute au même manuscrit, permet suggère que l’on a effacé le mot qui s’y trouvait
de préciser la datation relative de la réfection. Ce auparavant. Dans le troisième colophon, celui
feuillet, acquis en 1942 à Paris auprès du marchand d’al-Latāʾim wa-l-ashnāf (vol. 2, f. 28v), la mention
d’art George Tabbagh30, présente une réclame – de l’année est notée en chiffres et aucun signe de
contrairement au reste du manuscrit conservé manipulation ultérieure n’apparaît à première vue.
dans la collection privée dont les réclames ont été Un examen plus attentif montre cependant qu’en
intégralement grattées. Il est donc probable que le dessous du chiffre huit se devine la barre d’un neuf
manuscrit ait été remanié après cette date. qui a été gratté. Les colophons restitués donnent
les dates suivantes :
– Le texte du Miftāḥ al-fiṭna fī naẓm Kalīla
5 Colophons et copiste wa-Dimna de Jalāl al-Dīn al-Naqqāsh a
été achevé en un mois, en shaʿbān 943 /
Chacun des trois textes du recueil présente en sa janvier-février 1537 (vol. 2, f. 29v)
fin un colophon donnant la date d’achèvement – al-Ṣādiḥ wa-l-bāghim d’Ibn al-Habbāriyya, le 12
de la rédaction ainsi que le nom du copiste. Si on ramaḍān 943 / 22 février 1537 (vol. 1, f. 62v)
y lit désormais 843H/1439, un examen à la loupe – al-Laṭāʾim wa-l-ashnāf d’Ibn Makānis, le 13
binoculaire révèle que les dates ont été manipu- ramaḍān 943 / 23 février 1537 (vol. 2, f. 28v)
lées postérieurement à la copie afin d’antédater (Fig. 11.4)
le manuscrit. Si l’on examine les mentions du Les colophons fournissent également l’identité
siècle (thamānimiʾa) dans les colophons du Miftāḥ du copiste : Badr al-Dīn Muḥammad b. al-Ghazzī
al-fiṭna (vol. 2, f. 29v) et d’al-Ṣādiḥ wa-l-bāghim al-ʿĀmirī. Figure intellectuelle majeure de la pre-
(vol. 1, f. 62v), on constate que l’écriture y est irré- mière moitié du xvie siècle, il naît en 904H/1499
gulière et d’une graphie plus épaisse que celle du au sein d’une célèbre famille de savants damas-
cènes31. Après avoir été formé au Caire et reçu son
adhéré les uns aux autres sur la presque totalité de leur ijāza de deux des plus grands savants mamlouks,
marge, du côté du dos et qu’ils ont été désolidarisés à
Zakariyyā al-Ansārī (m. 1520) et Jalāl al-Dīn
l’occasion de la dernière restauration, réalisée de nos
jours. Plusieurs indices l’attestent : papier arraché et al-Suyūtī (m. 1505), il s’installe à Damas où il
dédoublé avec report de motifs ou de fragments. occupe les plus hautes fonctions religieuses : mufti
29 Voir au sujet du feuillet : Oktay Aslanapa, Turkish art shāfiʿī de la ville, imam de la grande mosquée des
and architecture (Londres : Faber and Faber, 1971), Umayyades et enseignant dans de nombreuses
313 ; Burchard Brentjes, Chane-Sultane-Emire: der
Islam vom Zusammenbruch des Timuridenreiches bis
madrasas. Il est l’auteur de plusieurs œuvres poé-
zur europäischen Okkupation (Leipzig : Koehler und tiques dont l’une particulièrement controversée à
Amelang, 1974), 16 ; Karin Rührdanz, éd., Orientalische
illustrierte Handschriften: aus Museen und Bibliotheken
der Deutschen Demokratischen Republik Ausstellung im 31 Voir à son sujet : Fatih Çollak et Cemil Akpınar,
Islamischen Museum der Staatlichen Museen zu Berlin, « Gazzi, Bedreddin », in Türkiye Diyanet Vakfı İslâm
Hauptstadt der DDR (Berlin : Staatliche Museen zu Ansiklopedisi (Istanbul : Türkiye Diyanet Vakfı, 1996) ;
Berlin, 1984), 122‑24. Ralf Elger, « Badr al-Dīn Muḥammad al-Ġazzī », in
30 Nous souhaiterions remercier le Dr. Deniz Erduman- Essays in Arabic Literary Biography, 1350-1850, éd.
Çalış, conservateur des manuscrits au Museum für Joseph E. Lowry et Devin J. Stewart, vol. 2 (Wiesbaden :
Islamische Kunst de Berlin, de nous avoir transmis Harrassowitz, 2009), 98‑106 ; Helen Pfeifer,
cette information ainsi que d’avoir fait l’effort, en pleine « Encounter after the Conquest: Scholarly Gatherings
pandémie, de se rendre au musée afin de prendre une in 16th-Century Ottoman Damascus », International
photographie du feuillet. Journal of Middle East Studies 47 (2015) : 219‑39.
332 El Khiari et al.
figure 11.5 Frontispice (vol. 1, f. 64) et finispice (vol. 2, f. 29) du Miftāḥ al-fiṭna fī naẓm Kalīla wa-Dimna, collection
particulière
poursuivait par al-Ṣādiḥ wa-l-bāghim long de centrale est composée de quatre couronnes prin-
2000 vers et s’achevait sur Al-Laṭāʾim wa-l-ashnāf cipales concentriques dont les remplissages font
composé de 570 vers. Plusieurs structures enlumi- appel aux répertoires végétalisant (arabesques) et
nées permettaient en outre de mettre en exergue géométrique (rubans, tresses). Au verso, se trouve
les divisions et hiérarchisations au sein du texte. un bandeau de ʿunwān qui marque le début du pre-
Le volume s’ouvrait sur la page de titre enlu- mier texte et présente un cartouche aux extrémités
minée du Kitāb miftāḥ al-fiṭna fī naẓm Kalīla en arcs kelīl prolongés de chaque côté par un motif
wa-Dimna de Jalāl al-Dīn al-Naqqāsh, actuelle- de demi-palmettes affrontées. Ce premier texte
ment située au f. 64 du volume 1. Ce folio, en très s’achève au f. 29v du volume 2 par une shamsa qui
mauvais état34, présente une shamsa centrale occupe la partie résiduelle de la page de colophon.
dorée en “cloud collar” et un bandeau supérieur Elle entre en résonance avec la shamsa initiale,
portant un cartouche en accolades35. La shamsa tant par la palette que par le choix d’un motif en
« cloud collar » placé au centre. (Fig. 11.5)
Le second texte, Al-Ṣādiḥ wa-l-bāghim, s’ouvre
34 La détérioration du feuillet d’ouverture du manuscrit
accrédite la thèse d’une humidification accidentelle du
sur une page de titre enluminée, située au f. 1 du
manuscrit. volume 2, dans laquelle l’artiste fait preuve de
35 Le bandeau inférieur n’est pas conservé. son talent dans la variété des solutions formelles.
334 El Khiari et al.
Le bandeau supérieur affecte ainsi la forme d’un construction plutôt malhabile et suggèrent la main
trapèze tandis que le bandeau inférieur reçoit un d’un apprenti ou tout du moins d’un artiste moins
cartouche à côtés en arc kelīl. La shamsa centrale qualifié. Les titres de chapitres des deux autres
est composée de trois couronnes concentriques. textes sont simplement copiés à l’encre dorée.
(Fig. 11.6) Ce second texte s’achève au f. 62v du Le programme des enluminures a été pensé avec
volume 1 par une autre shamsa qui prend place beaucoup de soin et les éléments se répondent,
dans la partie inférieure au texte du colophon. créant une grande harmonie graphique. Celle-ci
Moins élaborée que ses sœurs, elle présente un passe tout d’abord par la palette qui fait appel à
centre, vacant, entouré d’une épaisse couronne l’or, au bleu, à l’orange et au rose. La composition
garnie d’arabesques tracées aux encres orange et les motifs décoratifs sont également source de
et bleue dans des compartiments définis par des cohérence visuelle : le motif du « cloud collar », en
médaillons végétalisants. premier lieu, qui est utilisé dans l’enluminure du
Le troisième et dernier texte, al-Laṭāʾim wa-l- frontispice, du finispice et dans le shamsa conclu-
ashnāf, s’ouvre sur une page de titre enluminée, sive du premier texte. L’artiste a introduit de
située au f. 2 du volume 1 dont la shamsa est com- légères variations dans chacun des cas : le « cloud
posée de deux couronnes concentriques. Elle est collar » du frontispice présente ainsi huit fleurons,
accompagnée de deux bandeaux à la construc- ce qui en fait un rare exemple de « cloud collar »
tion élaborée contenant chacun deux cartouches complet. Les motifs décoratifs affichent également
oblongs polylobés. (Fig. 11.7) une grande homogénéité : fleurs de lotus, rinceaux
Enfin, un finispice très richement enluminé clôt fleuris spiralés, demi-palmettes dorées …
le volume36. Il se déploie sur une pleine page et
sa composition sophistiquée rappelle l’ensemble
des éléments de décor utilisés au fil du manuscrit. 7 Textes et peintures
La shamsa présente quatre cercles concentriques
dont le décor fait appel aux mêmes répertoires Le manuscrit de la collection S est le seul exemple
que celles des shamsa précédentes37. Le motif illustré connu de ces trois textes. S’il s’agit
central en « cloud collar » évoque celui du fron- aujourd’hui d’un unicum, un document révèle qu’à
tispice tandis que le bandeau inférieur en forme l’époque médiévale, ce n’était pas le cas. Le cata-
de trapèze évoque le bandeau supérieur trapé- logue de la bibliothèque de l’Ashrafiyya à Damas,
zoïdal de la page de titre enluminée du second rédigé dans les années 1270, indique la présence de
texte. Le bandeau supérieur quant à lui fusionne deux exemplaires illustrés d’al-Ṣādiḥ wa-l-bāghim,
deux éléments : le cartouche en accolades du le second texte du recueil38. Le document signale
bandeau supérieur du frontispice et le décor des également l’existence d’un manuscrit illustré d’une
demi-palmettes que l’on retrouve dans les ban- versification intitulée Muḍahat Kalīla wa-Dimna
deaux enluminés inférieurs des pages de titres des de Muḥammad Ibn Ḥusayn al-Yamanī, dont nous
deuxième et troisième textes. (Fig. 11.8) ne conservons aujourd’hui aucun exemplaire
Au sein du premier texte, des bandeaux enlu- illustré39. Étonnamment, aucune copie – textuelle
minés accueillent les titres copiés en graphie ou illustrée – de Kalīla wa-Dimna d’Ibn al-Muqaffaʿ
thuluth à l’encre dorée bordée d’un filet noir, sur
un fond orné de rinceaux. Ces bandeaux sont de 38 Konrad Hirschler, Plurality and Diversity in an Arabic
Library: The Ashrafiya Library Catalogue (Edinburgh :
Edinburgh University Press, 2016), 30, 240. Il note éga-
36 F. 63, vol. 1. lement l’existence de sept copies non illustrées de ce
37 F. 64, vol. 1 ; f. 29v, vol. 2 ; f. 1, vol. 2 ; f. 62v, vol. 1 ; f. 2, texte.
vol. 1. 39 Hirschler, Medieval Damascus, 286.
Entre deux mondes 335
figure 11.7 Frontispice d’al-Laṭāʾim wa-l-Ashnāf, vol. 1, figure 11.8 Finispice du manuscrit, vol. 1, f. 63
f. 2
n’est citée dans le catalogue, ce qui témoigne du 7.1 Palette chromatique et analyses
succès dont jouissaient au xiiie siècle à Damas ces physico-chimiques
versions concurrentes en vers. Le manuscrit est L’examen des peintures permet de mettre en évi-
très richement illustré de cent trente et une pein- dence une palette chromatique variée avec pour
tures. Le premier texte de soixante-deux feuillets une même couleur, un grand nombre de nuances :
reçoit quatre-vingt-six illustrations, le deuxième bleu clair et sombre, bleu gris, vert sombre et clair,
de vingt-deux feuillets, trente-six et le troisième orange vif, rouge vif, mauve sombre et pâle, rose
enfin de sept feuillets, neuf peintures. Le format clair et rose orangé, jaune vif, jaune ocré, blanc,
des peintures est assez varié allant de la pleine gris et noir.
page à la simple vignette. Quatre illustrations sont La caractérisation des pigments a été rendue
en pleine page (vol. 1, ff. 5v, 6, 52v, 54) – dont une possible par l’analyse scientifique menée par
double page – et 65% des peintures occupent la Nathalie Buisson, anciennement responsable du
moitié ou plus de la page ce qui témoigne de la laboratoire du département de la Conservation
place importante accordée à l’image au sein du de la BnF. Le manuscrit a tout d’abord fait l’ob-
texte. jet d’observations à la loupe binoculaire dans le
Entre deux mondes 337
but de déceler des particularités dans la couche Gris Noir de carbone éclairci au blanc de plomb
picturale. Ensuite, deux types d’analyses com- Noir Noir de carbone
plémentaires ont été réalisés afin de déterminer Doré Or en paillettes
la nature des pigments et des colorants présents Argenté Argent en paillettes
dans les peintures. La majorité des pigments a pu
être identifiée par fluorescence de rayons X (XRF), Ces analyses mettent en lumière la grande variété
technique très utile pour l’analyse non destructive de pigments minéraux et de colorants organiques
des matériaux du patrimoine40. La spectrocolori- utilisés. Dix pigments ont ainsi été identifiés : ver-
métrie, autre méthode d’analyse non destructive, millon, minium, indigo, orpiment, blanc de plomb,
est venue compléter l’identification des matières lapis-lazuli, noir de carbone, or, argent et un vert
colorantes41. Six feuillets représentatifs des au cuivre qui n’a pu être caractérisé complète-
couleurs employées dans les peintures ont été ment. Ceux-ci ne correspondent pas à l’intégralité
sélectionnés42 et leur analyse a permis de mettre des pigments qui ont été utilisés : deux pigments
en évidence les pigments suivants : n’ont pas pu être caractérisés par les analyses car
ils étaient de nature organique ; d’autres encore
Rouge Vermillon mélangé à un peu de minium n’ont certainement pas été décelés, seulement six
Orange Minium mélangé à un peu de vermillon feuillets ayant été analysés. Il est ainsi peu pro-
Vert – Indigo mélangé à de l’orpiment bable que l’indigo, utilisé en mélange avec l’orpi-
– Vert de cuivre (malachite ?) mélangé à ment pour faire du vert, n’ait pas été mis en œuvre
de l’orpiment pour la couleur bleue. Il faut donc se montrer
Rose – Vermillon éclairci au blanc de plomb prudent quant à l’interprétation des résultats de
– Colorant organique rouge inconnu ces analyses, qui n’ont pas vocation à se montrer
éclairci au blanc de plomb exhaustives.
Jaune – Orpiment Néanmoins, quelques points méritent d’être
– Orpiment mélangé à du minium soulignés. En premier lieu, la préciosité de la palette
Bleu Lapis-lazuli qui emploie avec abondance le lapis-lazuli, l’or et
Mauve Colorant organique inconnu l’argent. L’intégralité des ciels sont ainsi peints à
Blanc Blanc de plomb l’or tandis que l’argent est utilisé pour peindre l’eau
et quelques détails décoratifs. Si l’or était fréquem-
40 Une analyse élémentaire de la couleur a été réalisée ment utilisé dans la peinture de manuscrits arabes,
par spectrofluorescence de rayons X portable (Tracer en particulier sous les Mamlouks où les fonds
III-SD, Brucker). Cette technique d’analyse élémen- des peintures étaient parfois entièrement dorés,
taire sans contact ne nécessite pas de prélèvement et l’usage de l’argent est beaucoup plus rare. Il n’a pas
présente ainsi beaucoup d’avantages. En revanche, l’ex-
ploitation des résultats est assez complexe car l’analyse
du tout été employé dans les manuscrits arabes
se fait sur l’ensemble des couches de matière picturale de Kalīla wa-Dimna conservés à la Bibliothèque
et ne donne pas accès aux éléments légers du tableau nationale de France comme le montrent les ana-
périodique, en dessous du calcium. lyses physico-chimiques menées sur ce corpus43.
41 Le spectrocolorimètre portable NCS-RUBY, qui permet
En revanche, l’argent est employé dans la peinture
des mesures non destructives et sans contact, a été uti-
lisé pour la détermination de certains pigments bleus persane et ottomane où il est habituellement uti-
et verts. Les courbes de réflexions diffuses sont obte- lisé pour peindre l’eau44.
nues en éclairant le matériau avec une lumière blanche
fournissant un flux continu dans le visible ; la lumière
diffusée par l’échantillon est recueillie et analysée. Ce
spectre caractéristique du matériau est ensuite com- 43 Voir le chapitre 9 dans ce même volume.
paré à ceux d’une base de données de référence. 44 Voir Penley Knipe et al., « Materials and Techniques of
42 Il s’agit des feuillets 4v, 6v, 10v, 40, 54 et 61 du volume 1. Islamic Manuscripts », Heritage Science 6, no 55 (2018).
338 El Khiari et al.
La palette est également caractérisée par une composite, appelé vergaut, mélange d’indigo et
grande sophistication. Elle est composée de pig- d’orpiment. Ce mélange indigo-orpiment n’est pas
ments qui ont été utilisés soit individuellement, rare et a été identifié dans de nombreux manus-
soit mélangés ensemble pour produire de nou- crits arabes et persans48 ainsi que dans un manus-
velles couleurs. Le blanc de plomb a été utilisé crit turc49. Un deuxième vert a été utilisé pour
pour éclaircir telle ou telle couleur et produire une peindre les vêtements du feuillet 45v. Il n’a pas
grande variété de tons différents. Dans certains été possible de le caractériser avec précision mais
cas, comme celui des couleurs vert, rose et jaune, les analyses suggèrent la présence de malachite
deux mélanges de pigments ont été effectués afin à laquelle a été ajouté de l’orpiment50. De nom-
d’obtenir deux teintes différentes. breuses études signalent la présence de malachite
L’usage différencié des matériaux verts et roses dans les manuscrits persans51 du xvie au xviiie
est particulièrement intéressant ; en effet, deux siècle et dans un manuscrit turc du xvie siècle52.
verts et deux roses, différents en composition et
en utilisation ont été relevés dans les peintures
du manuscrit. Un premier rose a servi à réaliser le 48 On retrouve ce mélange dans plusieurs manuscrits de
Kalīla wa-Dimna (voir chapitre 9) : Arabe 3465, Arabe
fond du registre inférieur du feuillet 40. Il n’a pas 3467, Arabe 3472, Arabe 3475 ainsi que dans le manus-
été possible de le caractériser entièrement mais crit Arabe 6094 des Maqāmāt. On le retrouve égale-
il est probable qu’un colorant organique rouge, ment dans deux feuillets illustrés du Kitāb fī Maʿrifat
éclairci avec beaucoup de blanc de plomb soit à al-Ḥiyal al-Handasiya produits au Caire en 1354 : voir
l’origine de cette nuance45. Une deuxième nuance à ce sujet Knipe et al., « Materials and Techniques
of Islamic Manuscripts ». Pour d’autres exemples
rose a servi à peindre le pelage des lapins du voir Tracey D. Chaplin et al., « Raman Spectroscopic
registre supérieur du feuillet 40. Elle est compo- Analysis of Selected Astronomical and Cartographic
sée de vermillon éclairci avec du blanc de plomb, Folios from the Early 13th Century Islamic ‘Book of
mélange qui se retrouve dans plusieurs manuscrits Curiosities of the Sciences and Marvels for the Eyes’ »,
Journal of Raman Spectroscopy 37, no 8 (2006) : 865‑77 ;
arabes46, persans et turcs47. Deux teintes diffé- Robin J.H. Clark et Sigrid Mirabaud, « Identification
rentes de vert ont également été utilisées. Le vert of the Pigments on a Sixteenth Century Persian Book
ayant servi à réaliser le fond d’herbe sur lequel of Poetry by Raman Microscopy », Journal of Raman
ont été peints les lapins du feuillet 40 est un vert Spectroscopy 37, no 1‑3 (2006) : 238 ; Lucia Burgio
et al., « Pigment Analysis by Raman Microscopy
of the Non-Figurative Illumination in 16th- to
45 À cette époque, des colorants d’origine végétale, 18th-Century Islamic Manuscripts », Journal of Raman
comme la garance ou l’orseille ou encore d’origine ani- Spectroscopy 39, no 10 (2008) : 1482‑93 ; Chiara Anselmi
male, comme les rouges de cochenilles (rouge kermès, et al., « MOLAB® Meets Persia: Non-Invasive Study of
laque rouge, etc.) étaient largement utilisés. a Sixteenth-Century Illuminated Manuscript », Studies
46 On retrouve ce mélange dans plusieurs manuscrits de in Conservation 60, no 1 (2015) : 185‑92.
Kalīla wa-Dimna (voir chapitre 9) : Arabe 3465, Arabe 49 Jurado-López et al., « Analysis of the Palette of a
3467 et Arabe 3475. On le retrouve également dans Precious 16th Century illuminated Turkish Manuscript
deux feuillets illustrés du Kitāb fī Maʿrifat al-Ḥiyal by Raman Microscopy », 121.
al-Handasiya produits au Caire en 1354 : voir à ce sujet 50 Un broyage fin de la malachite peut réduire de beau-
Knipe et al., « Materials and Techniques of Islamic coup l’intensité de sa couleur verte et une étude a
Manuscripts ». relevé le fait que pour obtenir une tonalité verte satis-
47 Alicia Jurado-López et al., « Analysis of the Palette faisante, les peintres pouvaient appliquer un résinate
of a Precious 16th Century Illuminated Turkish de cuivre transparent ou encore, un pigment jaune.
Manuscript by Raman Microscopy », Journal of Raman 51 Knipe et al., « Materials and Techniques of Islamic
Spectroscopy 35, no 2 (2004) : 122 ; Vânia S.F. Muralha, Manuscripts », 25.
Lucia Burgio, et Robin J.H. Clark, « Raman spectros- 52 Jurado-López et al., « Analysis of the Palette of a
copy analysis of pigments on 16-17th c. Persian manus- Precious 16th Century Illuminated Turkish Manuscript
cripts », Spectrochimica Acta Part A: Molecular and by Raman Microscopy » ; on trouve dans ce manuscrit
Biomolecular Spectroscopy 92 (2012) : 21‑28. de la malachite mélangée à un peu de lazurite.
Entre deux mondes 339
Cet usage sélectif du vert est attesté dans la lit- tuel que le peintre décide de modifier son premier
térature scientifique portant sur des manuscrits tracé et de nombreux manuscrits conservent des
persans et turcs des xvie et xviie siècles53. Ces exemples de ce genre d’errata.
études révèlent que dans certains manuscrits, le
vergaut était réservé pour peindre la végétation 7.2 Sources iconographiques
tandis que les pigments à base de cuivre, comme Le cycle illustré est un unicum qui ne connaît pas
la malachite, étaient employés pour peindre les d’équivalent56. Il est néanmoins possible d’établir
autres éléments du décor, comme les vêtements. des analogies avec l’iconographie de manuscrits
On en trouverait également un exemple dans un de Kalīla wa-Dimna. Le premier texte du recueil,
manuscrit de Kalīla wa-Dimna produit en Égypte celui du Miftāḥ al-fiṭna fī naẓm Kalīla wa-Dimna,
ou en Syrie au début du xive siècle (BnF, Arabe s’y prête particulièrement bien puisqu’il suit de
3467). Au feuillet 61, la végétation a été peinte avec très près la version arabe d’Ibn al-Muqaffaʿ. Les
du vergaut tandis que la robe du personnage a été deux autres textes en revanche s’en éloignent
peinte avec du vert au cuivre54. Cet usage sélectif beaucoup et présentent, à l’exception des grands
reste toutefois à nuancer car il n’a pas fait ici l’objet poncifs que sont les scènes de trône et de combats,
d’une étude systématique. une iconographie tout à fait originale. De nom-
Le mode de réalisation des figures humaines breuses scènes, considérées comme des leitmotivs
et animales a également pu être étudié et permet de l’iconographie de Kalīla wa-Dimna – aussi bien
de dessiner les contours du processus artistique. dans les manuscrits arabes que persans – ne sont
Le dessin préparatoire a été ébauché au trait gris ici pas représentées. C’est le cas du combat du lion
tandis que le rendu des chairs a été appliqué à et du bœuf, des périls de la vie, du singe et du char-
l’intérieur des contours. Dans le cas des figures pentier, du lion et du lièvre ou encore de la tortue
humaines, l’arrondi des pommettes, le front et et des canards. D’autres épisodes en revanche, peu
le menton ont été accentués par une teinte rose ou pas représentés ailleurs, reçoivent une illus-
orangée plus soutenue que celle de la carnation. tration : l’histoire de l’éléphant et de l’alouette57,
Un trait brun rouge a été passé par-dessus le dessin de l’intendant des mers58 ainsi que plusieurs
préparatoire pour définir les contours du visage, du épisodes issus du chapitre du fils du roi et de ses
nez, de la bouche et des mains. Enfin, un trait noir compagnons59 viennent souligner la grande ori-
a été passé pour accentuer le rendu des yeux (pau- ginalité du cycle illustré. Celui-ci est une combi-
pières et pupilles), des sourcils et des moustaches. naison tout à fait particulière d’éléments issus de
L’alignement du dessin final sur l’esquisse n’est pas répertoires iconographiques distincts.
toujours exact ce qui explique que des traces du Certaines peintures se rapprochent des copies
dessin préparatoire soient clairement visibles à de illustrées arabes de Kalīla wa-Dimna et des codex
nombreux endroits55. Il n’était en effet pas inhabi- mamlouks en particulier. La fable du marchand
et du joueur de harpe, située au feuillet 19v du
53 Anselmi et al., « MOLAB® Meets Persia: Non-Invasive
Study of a Sixteenth-Century Illuminated Manus-
cript » ; Chaplin et al., « Raman Spectroscopic Analysis 56 Un article présentant le cycle iconographique complet
of Selected Astronomical and Cartographic Folios from et inédit de ce manuscrit ainsi que la reconstitution
the Early 13th Century Islamic ‘Book of Curiosities of de l’ordre des feuillets est en préparation par Aïda El
the Sciences and Marvels for the Eyes’ » ; Jurado-López Khiari.
et al., « Analysis of the Palette of a Precious 16th Cen- 57 La fable de l’éléphant et de l’alouette fait l’objet de trois
tury illuminated Turkish Manuscript by Raman Micros- illustrations : deux au f. 7 et une au f. 7v (vol. 1).
copy », 121 ; Knipe et al., « Materials and Techniques of 58 F. 26v (vol. 1).
Islamic Manuscripts », 20‑21. 59 La fable du fils du roi et de ses compagnons donne lieu
54 Voir le chapitre 9 à ce sujet. à six peintures : f. 59, f. 59v (deux illustrations), f. 60,
55 Voir par exemple au feuillet 40 du volume 1. f. 60v, f. 61.
340 El Khiari et al.
volume 1 par exemple, est très proche dans sa et certains détails particuliers le rapprochent du
composition et son iconographie, de ce que l’on plus ancien manuscrit arabe des fables61 (on peut
trouve dans les manuscrits mamlouks. La mise prendre pour exemple l’iconographie du combat
en scène épurée inscrit les personnages dans un entre le bœuf et le lion62). Dans le cas de l’épisode
simple cadre pourvu d’écoinçons. Cette architec- du marchand et du joueur de harpe, le peintre du
ture indique que l’action se déroule en intérieur manuscrit de la collection S., suit un schéma très
et se retrouve dans de nombreux manuscrits proche de celui du codex de Cambridge63 (Fig. 11.9)
mamlouks. L’iconographie et la composition de (Fig. 11.10). Le marchand est représenté à droite,
la scène apparentent le manuscrit de la collec- jambes croisées, battant les mains de joie tandis
tion S. à un codex mamlouk de Kalīla wa-Dimna que face à lui un personnage joue des cymbales.
conservé à Cambridge, au Corpus Christi College
sous la cote ms. 578. Il est daté de 791 H/1388 et 61 Au sujet de ce manuscrit, conservé à la Bibliothèque
présente l’un des cycles illustrés les plus dévelop- nationale de France sous la cote Arabe 3465, voir Hugo
pés du groupe des manuscrits arabes de Kalīla Buchthal, « “Hellenistic” Miniatures in Early Islamic
wa-Dimna. Son iconographie le distingue très Manuscripts », Ars Islamica 7, no 2 (1940) : 125‑50 ;
Hugo Buchthal, « Three Illustrated Hariri Manuscripts
nettement du reste des manuscrits mamlouks qui in the British Museum », The Burlington Magazine
forment un ensemble extrêmement homogène60 for Connoisseurs 77, no 452 (1940) : 144-152 ; Grube,
« Prolegomena for a Corpus Publication of Illustrated
60 Au sujet de ces manuscrits mamlouks à savoir les Kalilah wa Dimna Manuscripts », 374 ; Lucy-Anne Hunt,
manuscrits Arabe 3467 à la Bibliothèque nationale de « Christian-Muslim Relations in Painting in Egypt of
France , Cod Arab 616 à la Staastbibliothek de Munich, the Twelfth to Mid-Thirteenth Centuries: Sources of
Pococke 400 à la Bodleian Library à Oxford et un feuil- Wall Painting at Deir es-Suriami and the Illustration
let isolé T-S Ar.51.60 à la bibliothèque universitaire de of the New Testament ms Paris, Copte-Arabe 1/ Cairo,
Cambridge, au Royaume Uni, voir : Sofie Walzer, « An Bible 94 », Cahiers archéologiques 33 (1985) : 111‑55 ;
Illustrated Leaf from a Lost Mamluk Kalilah wa Dimnah Hunt, « Manuscript Production by Christians in
Manuscript », Ars Orientalis 2 (1957) : 503‑5 ; Walzer, 13th-14th Century Greater Syria and Mesopotamia
« The Mamlûk Illuminated Manuscript of Kalîla and Related Areas », Aram Periodical 9‑10 (1997) :
wa-Dimna », in Aus der Welt der islamischen Kunst: 289‑326 ; O’Kane, Early Persian Painting, 218 ; Anna
Festschrift für Ernst Kühnel zum 75 Geburstag am 26-10- Contadini, « Ayyubid Illustrated Manuscripts and their
1957, éd. Richard Ettinghausen (Berlin : Gebr. Mann, North Jaziran and ʿAbbasid Neighbours », in Ayyubid
1959), 195‑206 ; Duncan Haldane, Mamluk Painting Jerusalem. The Holy City in Context 1187-1250, éd. Robert
(Warminster : Aris & Phillips, 1978), 81-82, 85-88, 95-96 ; Hillenbrand et Sylvia Auld (Londres : Altajir Trust,
Knut Holter, « Die frühmamlukische miniaturmale- 2009), 179‑94.
rei », Die graphischen Künste, no 2 (1978) : 1‑14 ; Esin 62 Au sein du manuscrit de Kalīla wa-Dimna Arabe 3465,
Atıl, Kalila Wa Dimna: Fables from a Fourteenth-Century l’épisode du combat du lion et du taureau donne lieu
Arabic Manuscript (Washington, D.C. : Smithsonian à une peinture où l’on voit le lion attaquer le taureau
Institution Press, 1981) ; Esin Atıl, Renaissance of Islam: par l’arrière en le saisissant par l’épaule (Paris, BnF,
Art of the Mamluks (Washington, D.C. : Smithsonian Arabe 3465, f. 71v). Cette représentation qui rappelle les
Institution Press, 1981) ; François de Blois, Burzōy’s reliefs sassanides sera par la suite abandonnée dans les
Voyage to India and the Origin of the Book of Kalīlah manuscrits mamlouks de Kalīla wa-Dimna où le lion
Wa Dimnah (Londres : Royal Asiatic Society, 1990), 70 ; est vu attaquant le taureau de face (Paris, BnF, Arabe
Robert Hillenbrand, « Mamlūk and Īlkhānid Bestiarie : 3467, f. 40v ; Munich, BSB, Cod. Arab. 616, f. ; Oxford,
Convention and Experiment », Ars Orientalis 20 (1990) : Bodleian Library, Pococke 400, f. 63). Le manuscrit de
149‑87 ; Grube, « Prolegomena for a Corpus Publication Cambridge est le seul manuscrit mamlouk de Kalīla
of Illustrated Kalilah wa Dimna Manuscripts », 375- wa-Dimna qui reprenne l’iconographie du manuscrit
376 ; Julian Raby, « Between Sogdia and the Mamluks: A Arabe 3465 (Cambridge, Corpus Christi College, Parker
Note on the Earliest Illustrations to Kalila wa Dimna », Library, ms. 578, f. 52v). Outre cet épisode, ces deux
Oriental Art 33, no 4 (1991) : 381‑98 ; Bernard O’Kane, manuscrits présentent de nombreuses similarités ico-
Early Persian Painting: Kalila and Dimna Manuscripts nographiques qui demandent à être détaillées.
of the Late Fourteenth Century (Londres : I.B. Tauris, 63 Cambridge, Corpus Christi College, Parker Library, ms.
2003), 221-222. 578, f. 24v.
Entre deux mondes 341
figure 11.9 Fable du marchand et du joueur de figure 11.10 Fable du marchand et du joueur de harpe, f. 24,
harpe : vol. 1, f. 19 1388 ?, ms. 578, Cambridge, Parker Library
Les cymbales sont, dans les deux cas, tenues à la paroxystique du récit – se rapproche là encore du
verticale par le joueur, contrairement à ce qui est manuscrit de Cambridge66 dans lequel un cha-
figuré dans les autres manuscrits mamlouks où les meau, affalé sur ses pattes, est en train de se faire
cymbales sont à l’horizontale. Ce détail se retrouve dévorer par les animaux (Figs. 11.11 et 11.12). Tous
également dans le plus ancien manuscrit arabe les autres manuscrits arabes médiévaux adoptent
des fables64 qui représente, à l’instar du manuscrit une représentation différente de la scène, le cha-
de la collection S, l’ensemble des détails évoqués meau étant représenté à l’envers, le cou tordu de
par le texte, à savoir un coffret à bijoux (figuré douleur67. Le corbeau crevant les yeux du cha-
dans les deux cas près du marchand comme le meau se retrouve uniquement dans le plus ancien
spécifie le récit) et les outils du joaillier. On peut
également évoquer l’épisode du lion, du loup, du
corbeau, du chacal et du chameau. La représenta-
tion de la mise à mort du chameau65 – moment 66 Cambridge, Corpus Christi College, Parker Library, ms.
578, f. 49v.
67 Paris, BnF, Arabe 3465, f. 66 ; Rabat, Bibliothèque
64 Paris, BnF, Arabe 3465, f. 40. Royale de Rabat, ms. 3655, f. 41v ; Munich, BSB, Cod.
65 F. 26, volume 1. Arab. 616, f. 59v ; Oxford, Bodl. Libr., Pococke 400, f. 60.
342 El Khiari et al.
figure 11.11 Fable du lion, du loup, du corbeau, du figure 11.12 Fable du lion, du loup, du corbeau, du
chacal et du chameau, vol. 1, f. 26 chacal et du chameau, f. 49v, 1388 ?, ms. 578,
Cambridge, Parker Library
manuscrit arabe68 des fables ainsi que dans un la scène, est également très similaire à ce que l’on
manuscrit persan69. trouve dans les manuscrits persans, et plus parti-
D’autres peintures adoptent une composition culièrement dans un manuscrit jalayride daté des
dérivée de schémas iconographiques issus de la années 1370-74. Ce manuscrit conservé dans la
tradition persane. Prenons pour exemple l’épisode Bibliothèque de l’Université d’Istanbul sous la cote
du voleur et du rayon de lune, situé au f. 18v du F. 1422 a servi de modèle à de nombreux manus-
volume 1 (Fig. 11.13). L’iconographie de la scène se crits persans ; de fait, son iconographie est fré-
rapproche de l’épisode tel qu’il est illustré dans les quemment reproduite70. La copie de la collection
cycles persans. Le voleur est ainsi représenté non S. suit scrupuleusement le texte en représentant le
pas en train de tomber, comme on le voit dans l’en- rayon de lune que le voleur a voulu utiliser pour
semble des manuscrits arabes, mais allongé au sol
et battu par le mari. La structure hexagonale du
toit, au travers duquel trois complices observent
70 Istanbul, Istanbul University Library, F. 1422. Au sujet
de ce manuscrit datable des années 1370-74 et proba-
68 Paris, BnF, Arabe 3465, f. 66. blement produit à Tabriz voir O’Kane, Early Persian
69 Istanbul, Topkapı Sarayı Müzesi Kütüphanesi, H. 363, f. Painting, 42 ; appendix 15, 234-253 et illustration repro-
68v. duite en planche 15 du même ouvrage, p. 100.
Entre deux mondes 343
descendre dans la maison. En cela, il diffère des tandis que la représentation des personnages arbo-
représentations persanes qui ne figurent jamais rant de grands visages ronds dont les traits sont
la lune ou les rayons lumineux, à l’exception du systématiquement représentés de trois-quarts
manuscrit persan cité précédemment qui dépeint affiche une facture relativement similaire dans
partiellement le rayon de lune71. Au sein du cor- les deux manuscrits. L’influence de la production
pus arabe, seul le plus ancien manuscrit illustré turkmène de Chiraz des années 1480-90 y est,
des fables – Arabe 3465 – représente ce détail sous dans les deux cas, particulièrement sensible. On
la forme de rayons lumineux émanant de l’astre y trouve ainsi l’usage fréquent du pavage végétal
nocturne72. en écailles de poisson utilisé aussi bien pour cer-
taines représentations d’arbres que comme motif
de remplissage des fenêtres74. Ces éléments qui
8 Un représentant de la peinture pourraient être considérés par un peintre persan
ottomane de la première moitié du xvie du xvie siècle comme des “archaïsmes” font par-
siècle tie du répertoire habituel de la peinture ottomane
de la première moitié du xvie siècle. L’apport de
Malgré la diversité des sources iconographiques, la tradition illustrée persane dans la peinture otto-
le style des peintures au sein du manuscrit de la mane de la seconde moitié du xve et de la pre-
collection S. est très homogène ce qui pourrait mière moitié du xvie siècle est bien documenté75.
indiquer qu’un même artiste a supervisé l’ou- Les références à la peinture turkmène de Chiraz
vrage voire peint l’ensemble. Les peintures se sont prédominantes mais des références tabrizi
rapprochent de deux manuscrits dont le style voire herati ne sont pas à exclure. Celles-ci sont
très particulier caractérise une des productions venues s’agréger au répertoire ottoman au fur et à
du règne du sultan Süleyman (r. 1520-1566). L’un mesure des conquêtes et leur importance a long-
d’entre eux retient plus particulièrement l’atten- temps fait considérer, aux historiens de l’art, que
tion : un volume du Selīmnāma, copié en 152773, la peinture ottomane n’était qu’un mauvais ersatz
présente de nombreuses similarités stylistiques de la peinture persane76. Cette tendance a été
avec le présent manuscrit (Fig. 11.14). Le traite-
ment des motifs décoratifs, comme les carreaux 74 On retrouve cette utilisation du pavage végétal en
écailles de poisson comme motif de remplissage des
hexagonaux et les compositions spiralées à décor
fenêtres dans un manuscrit turc du İskendernāma
de fleurs, évoquent le manuscrit de la collection S conservé à Venise à la Biblioteca Nazionale Marciana
sous la cote Cod. Or.XC = 57. Au sujet de ce manuscrit,
71 Istanbul, Istanbul University Library, F. 1422. Voir au voir Ayşi̇n Yoltar-Yıldırım, « The Role of Illustrated
sujet de cet épisode et sa représentation O’Kane, Early Manuscripts in Ottoman Luxury Book Production:
Persian Painting, 52. 1413-1520 » (Thèse de doctorat, New York University,
72 Paris, BnF, Arabe 3465, f. 38. Les manuscrits mamlouks 2003), 101‑45.
représentent le rayon de lune sous la forme d’un 75 Voir par exemple Yoltar-Yıldırım, « The Role of
bloc monolithique de forme rectangulaire débutant Illustrated Manuscripts in Ottoman Luxury Book
soit au niveau de la lune (Munich, Cod. Arab. 616, Production: 1413-1520 » ; Yoltar-Yıldırım, « A 1498-99
f. 36), soit émanant du plafond de la maison (Oxford, Khusraw va Shīrīn: Turning the Pages of an Ottoman
Pococke 400, f.31). Le manuscrit de Cambridge ne Illustrated Manuscript », Muqarnas 22 (2005) : 95‑109 ;
représente pas le rayon de lune tandis que le manuscrit Lâle Uluç, Turkman Governors, Shiraz Artisans and
de Paris présente une lacune au niveau de cet épisode. Ottoman Collectors: Sixteenth Century Manuscripts
73 Il est conservé à Jérusalem, au sein de la Bibliothèque (Istanbul : Türkiye İş Bankası, 2006) ; Serpil Bağcı et al.,
nationale d’Israël sous la cote Yahuda Ms.Ar.1116 et il est Ottoman Painting, 2e éd. (Ankara : Ministry of Culture
daté du 1er shaʿbān 930/1524 ; Efraim Wust, Catalogue and Tourism Publications, 2010).
of the Arabic, Persian and Turkish Manuscripts of the 76 Thomas W. Arnold et Adolf Grohmann, The Islamic
Yahuda Collection of the National Library of Israel, vol. 2 Book: A Contribution to its Art and History from the
(Leiden : Brill, 2020), 684‑86. VII-XVIII Century (Paris : Pegasus Press, 1929), 94‑95.
Entre deux mondes 345
figure 11.16 Représentation d’une ville miniature, détail, vol. figure 11.17 Fable du chameau et du chamelier,
1, f. 9 (Cliché d’Aïda El Khiari) détail, vol. 2, f. 5v
récit illustré de la campagne d’Iraq entreprise par fera que très peu œuvre de mécène. Il n’en mène
le Sultan Süleyman en 1534. L’expansion de l’em- pas moins une importante politique d’appropria-
pire entraîne en effet le développement de la car- tion culturelle, humaine et matérielle dont les
tographie dès le début du xvie siècle et les images retombées se feront sentir sous son successeur.
miniatures de villes et de paysages vont acquérir Ces déportations humaines et ces prélèvements
une place importante dans la peinture ottomane matériels, à destination d’Istanbul, étaient des ins-
du xvie siècle. truments traditionnels de la politique ottomane.
Ainsi, en 1514, lors de la défaite de Chaldiran, Selīm
négocie la déportation d’artistes et d’artisans tabrizi
9 Un artiste ottoman à Damas parmi lesquels de nombreux artisans du livre88.
En Égypte et en Syrie, il fait confisquer les fonds
Si le texte du Selīmnāma relate les grands épisodes de plusieurs bibliothèques : on peut mentionner
du règne du sultan Selīm, les deux exemplaires
illustrés conservés ont été produits durant le règne
de son fils, le sultan Süleyman (r. 1520-1566). Selīm 88 Norah Titley, « Early Ottoman Miniature Painting: Two
passera en effet la majeure partie de son règne à Recently Acquired Manuscripts in the British Library »,
étendre les frontières de l’empire ottoman et ne The British Library Journal 9, no 2 (1983) : 135.
348 El Khiari et al.
entre autres celle de l’Ashrafiyya à Damas89, les travailler dans les anciens territoires mamlouks94.
bibliothèques des madrasas al-Maḥmūdiyya, Cette mesure, qui n’a pas suscité beaucoup d’inté-
al-Muʿayyadiyya et al-Ṣarjatmichiyya au Caire rêt dans la littérature, se révèle d’une importance
ainsi que la bibliothèque de la mosquée du Sultan fondamentale dans la compréhension du manus-
al-Ghawrī90. Un document explicite la mise en crit de la collection S. Jusqu’alors, l’un des premiers
œuvre de ces confiscations, plus spécifiquement la témoignages révélant la présence d’un peintre
répartition des manuscrits de la citadelle d’Alep91. ottoman dans le domaine syro-égyptien était un
Ce document classe les manuscrits en trois caté- manuscrit turc illustré du Hümāyūnnāma produit
gories : premièrement, les manuscrits luxueux au Caire en 156795. Le manuscrit de la collection
jugés dignes de rejoindre la bibliothèque impé- S., révèle la présence certaine, en 1537 à Damas,
riale ; deuxièmement, les ouvrages jugés “ordi- d’un ou plusieurs artisans du livre ottoman. Si le
naires” qui pourront éventuellement être acquis lieu de copie n’est pas explicitement mentionné,
par les membres de la cour ; enfin, les manuscrits on peut supposer au vu de l’identité du copiste,
incomplets ou jugés inintéressants, destinés à être célèbre intellectuel damascène, que le manuscrit y
vendus. a été copié et peint. L’étude des vêtements, en leur
Les sources mentionnent également la dépor- qualité de marqueurs spatio-temporels nous four-
tation de nombreux notables, artistes et artisans, nit à cet égard des indices précieux et plusieurs
dont des papetiers et des relieurs, du Caire vers coiffes rattachent indubitablement ce manuscrit
Istanbul92. Un auteur turc écrivant à la fin du xvie aux anciens domaines mamlouks. Une coiffe mas-
siècle des chroniques sur les artistes, mentionne culine visible aux feuillets 4v, 20v, 60v du volume
également la déportation d’artistes aleppins 1 est identique à celle que porte le personnage
vers Istanbul, évoquant en particulier un artiste d’une gravure de Vecellio intitulée « Maure noble
nommé Tāj al-Dīn et son fils Ḥusayn Balī93. En du Caire »96. Le personnage y porte « un turban
parallèle, Selīm Ier aurait favorisé l’immigration de toile semblable à celui des Turcs » agrémenté
d’artistes et artisans venus d’Istanbul pour aller d’un « mouchoir qui tombe du cou ». Une variante
que l’on trouve au feuillet 57v du volume 1 et aux
feuillets 15v et 17 du volume 2, consiste à enrou-
89 Hirschler, Medieval Damascus, 46‑53.
90 Gülru Necipoğlu, « The Spatial Organization of ler l’écharpe de toile autour du cou avant de la
Knowledge in the Ottoman Palace Library: An Ency- laisser retomber sur le torse. Ce couvre-chef est
clopedic Collection and Its Inventory », in Treasures of également représenté dans plusieurs tableaux
Knowledge: An Inventory of the Ottoman Palace Library italiens du début du xvie siècle dont une toile
(1502/3-1503/4), éd. Gülru Necipoğlu, Cemal Kafadar, et
Cornell H. Fleischer, vol. 1, 2 vol. (Leiden : Brill, 2019),
longtemps attribuée à Giovanni Bellini et décri-
note de bas de page n°120, 69. vant la réception d’une délégation vénitienne à
91 İsmail E. Erünsal, Ottoman Libraries: A Survey of the
History, Development and Organization of Ottoman
Foundation Libraries (Cambridge : Harvard University,
2008), 30. Le document est conservé à Istanbul, 94 Ibn Iyās, Badāʾiʿ al-zuhūr fī waqāʾiʿ al-duhūr, 188 ;
Topkapı Sarayı Arşivi n° D.9101. Behrens-Abouseif, « The Ottoman Conquest of Egypt
92 Ibn Iyās, Badāʾiʿ al-zuhūr fī waqāʾiʿ al-duhūr, éd. and the Arts », 305.
Muḥammad Muṣṭafā (Fīsbādin : Frānz Shtāynar, 1975), 95 Güner Inal, « Kahire’de yapılmış bir Hümâyünnâme’nin
vol. 5, 230, 398. minyatürleri », Türk Tarih Kurumu Belleten 40, no 159
93 Voir Esra Akın-Kıvanç, éd. et trad., Mustafa ʿÂli’s Epic (1976) : 439‑65.
deeds of artists: A critical edition of the earliest Ottoman 96 Cesare Vecellio, Costumes anciens et modernes, 4e éd.
text about the calligraphers and painters of the Islamic (Paris : Typographie de Firmin Didot frères, 1860), vol.
world (Leiden : Brill, 2011), 273-274. 2, n°430.
Entre deux mondes 349
(qāḍī al-quḍāt). Ces majālis semblent avoir gran- termes ce manuscrit doit-il être défini et à quel
dement favorisé les échanges et la circulation des champ culturel et artistique appartient-il ? Doit-il
manuscrits, dont les dignitaires ottomans étaient être considéré comme appartenant à la peinture
particulièrement friands. Les contemporains nous arabe ou à la peinture ottomane ? La peinture
relatent ainsi que lors de son séjour de quatre ans arabe est habituellement définie comme la pein-
à Damas, Kınalızade ʿAli aurait acheté plus de ture des manuscrits en langue arabe mais l’étude
500 manuscrits106. de ce manuscrit souligne combien cette définition
C’est sans doute au sein de ce tissu social fait linguistique, trop restrictive, ne suffit pas à rendre
d’accointances culturelles et intellectuelles qu’il compte des interactions artistiques complexes qui
faut replacer ce manuscrit. Son idiosyncrasie se font jour au xvie siècle entre des domaines tels
résulte probablement d’une volonté de se confor- que l’Anatolie, l’Égypte, le Syrie et l’Iran.
mer aux goûts voire aux demandes d’un comman-
ditaire. L’absence des pages originelles de titre
initiale et finale du manuscrit nous prive hélas des Bibliographie
indices précieux qui y sont habituellement consi-
gnés – telles que les marques de possesseurs ou Akın-Kıvanç, Esra, éd. et trad. Mustafa ʿÂli’s Epic deeds
notes de lectures – tandis que les trois colophons of artists: a critical edition of the earliest Ottoman text
restent muets quant à l’identité des anciens pro- about the calligraphers and painters of the Islamic
priétaires. Néanmoins, l’association unique des world. Leiden : Brill, 2011.
trois versifications en arabe qui le composent, Anselmi, Chiara, Paola Ricciardi, David Buti, Aldo
pointent résolument vers une clientèle lettrée tan- Romani, Patrizia Moretti, Kristine R. Beers,
dis que les peintures, effectuées par un artiste otto- Brunetto G. Brunetti, Costanza Miliani, et Antonio
man, suggèrent possiblement un commanditaire Sgamellotti. « MOLAB® Meets Persia: Non-Invasive
issu des rangs des hauts fonctionnaires ottomans. Study of a Sixteenth-Century Illuminated Manus-
cript ». Studies in Conservation 60, no 1 (2015) : 185-92.
Arnold, Thomas W., et Adolf Grohmann. The Islamic
11 Conclusion Book: A Contribution to its Art and History from the
VII-XVIII Century. Paris : Pegasus Press, 1929.
Ce manuscrit est un témoin exceptionnel de son Aslanapa, Oktay. Turkish art and architecture. Londres :
temps, qui a trop longtemps échappé à l’attention Faber and Faber, 1971.
des historiens de l’art. Son étude témoigne de la
complexité du paysage artistique damascène après
la conquête ottomane et soulève des questions s’accordent à en questionner les fondements, il n’existe
complexes qui ont trait aux interactions cultu- pas à ce jour d’article qui en fasse un traitement en
relles et artistiques. Plus encore, il soulève avec interrogeant directement la production illustrée. Voir
à ce sujet Anna Contadini, « The Manuscript as a
acuité des problèmes de définition et de termino-
Whole », in Arab Painting: Text and Image in Illustrated
logie qui sous-tendent aujourd’hui tout le champ Arabic Manuscripts, éd. Anna Contadini (Leiden :
de l’histoire de l’art islamique et qui ont trait au Brill, 2007), 3‑16 ; Oleg Grabar, « What Does “Arab
concept problématique de l’“identité”107. En quels Painting” Mean? », in Arab Painting, 17‑22 ; Nourane
Ben Azzouna, « The challenge of identity: the so-called
school of Baghdad reassessed » (The genius loci in
106 Pfeifer, « Encounter after the Conquest », 250. Islamic art: historical explorations in topological aes-
107 La question de l’identité de la peinture arabe demeure thetics, Université de Vienne, Institut d’Histoire de
problématique et si de nombreuses publications l’art, 2016), communication non publiée.
352 El Khiari et al.
Atıl, Esin. Kalila Wa Dimna: Fables from a Fourteenth- Blois, François de. Burzōy’s Voyage to India and the
Century Arabic Manuscript. Washington, D.C.: Origin of the Book of Kalīlah Wa Dimnah. Londres :
Smithsonian Institution Press, 1981. Royal Asiatic Society, 1990.
Atıl, Esin. « Mamluk Painting in the Late Fifteenth Brend, Barbara. « The Little People: Miniature
Century ». Muqarnas 2 (1984) : 159. Cityscapes and Figures in Persian and Ottoman
Atıl, Esin. « Ottoman Miniature Painting under Sultan Painting ». In Proceedings of the Third European
Mehmed II ». Ars Orientalis, 1973, 103-20. Conference of Iranian Studies, Part 2: Medieval and
Atıl, Esin. Renaissance of Islam: Art of the Mamluks. Modern Persian Studies. Édité par Charles Melville,
Washington D.C. : Smithsonian Institution Press, 367-83. Wiesbaden : Verlag, 1999.
1981. Brentjes, Burchard. Chane-Sultane-Emire: der Islam
Atıl, Esin. Turkish art. Washington D.C. : Smithsonian vom Zusammenbruch des Timuridenreiches bis zur
Institution Press, 1980. europäischen Okkupation. Leipzig : Koehler und
Bağcı, Serpil, Filiz Çağman, Günsel Renda, et Zeren Amelang, 1974.
Tanındı. Ottoman Painting. 2e éd. Ankara : Ministry Brockelmann, Carl. « Kalīla wa-Dimna » In Encyclo
of Culture and Tourism Publications, 2010. paedia of Islam, Second Edition. Édité par Peri
Bağcı, Serpil, Filiz Çağman, Günsel Renda, et Zeren Bearman, Thierry Bianquis, Clifford E. Bosworth,
Tanındı. « The Art of Painting in Ottoman Provinces ». Emeri van Donzel, Wolfhart P. Heinrichs. Consulté
In Ottoman Painting, 2e éd., 246-61. Ankara : Ministry en ligne le 22 novembre 2018.
of Culture and Tourism Publications, 2010. Buchthal, Hugo. « “Hellenistic” Miniatures in Early
Barrucand, Marianne. « Les représentations d’archi- Islamic Manuscripts ». Ars Islamica 7, no 2 (1940) :
tectures dans la miniature islamique en Orient du 125-50.
début du XIIIe au début du XIVe siècle ». Cahiers Buchthal, Hugo. « Three Illustrated Hariri Manuscripts
archéologiques 34 (1986) : 119-41. in the British Museum ». The Burlington Magazine
Barrucand, Marianne. « Un manuscrit arabe illustré for Connoisseurs 77, no 452 (1940) : 144-52.
de Kalila wa Dimna du XIIIe siècle et sa copie otto- Burgio, Lucia, Robin J.H. Clark, Vânia S.F. Muralha,
mane ». Archéologie islamique, no 2 (1991) : 81-95. et Tim Stanley. « Pigment Analysis by Raman
Behrens-Abouseif, Doris. Egypt’s Adjustment to Ottoman Microscopy of the Non-Figurative Illumination in
Rule: Institutions, Waqf and Architecture in Cairo 16th- to 18th-Century Islamic Manuscripts ». Journal
(16th and 17th Centuries). Leiden : Brill, 1994. of Raman Spectroscopy 39, no 10 (2008) : 1482-93.
Behrens-Abouseif, Doris. « Sultan al-Ghawri and the Çağman, Filiz. « The Miniatures of the Divan-i
Arts ». Mamluk Studies Review 6 (2002) : 69-75. Huseyni and the Influence of their Style ». In Fifth
Behrens-Abouseif, Doris. « The Ottoman Conquest International Congress of Turkish Art. Édité par Géza
of Egypt and the Arts ». In Conquête ottomane de Fehér, 231-59. Budapest : Akadémiai kiadó, 1978.
l’Égypte (1517) : arrière-plan, impact, échos. Édité Chaplin, Tracey D., Robin J.H. Clark, Alison McKay, et
par Benjamin Lellouch et Nicolas Michel, 303-26. Sabina Pugh. « Raman Spectroscopic Analysis of
Leiden : Brill, 2013. Selected Astronomical and Cartographic Folios from
Ben Azzouna, Nourane. « The challenge of iden- the Early 13th Century Islamic ‘Book of Curiosities of
tity: the so-called school of Baghdad reassessed ». the Sciences and Marvels for the Eyes’ ». Journal of
Communication présentée lors du workshop inti- Raman Spectroscopy 37, no 8 (2006) : 865-77.
tulé « The genius loci in Islamic art: historical explo- Clark, Robin J.H., et Sigrid Mirabaud. « Identification
rations in topological aesthetics », Université de of the Pigments on a Sixteenth Century Persian Book
Vienne, Institut d’Histoire de l’art, 25-26 novembre of Poetry by Raman Microscopy ». Journal of Raman
2016. Spectroscopy 37, no 1-3 (2006) : 235-39.
Entre deux mondes 353
Çollak, Fatih, et Cemil Akpınar. « Gazzi, Bedreddin ». In Manuscripts. Édité par Anna Contadini, 17-22.
Türkiye Diyanet Vakfı İslâm Ansiklopedisi, 13:537-39. Leiden : Brill, 2007.
Istanbul : Türkiye Diyanet Vakfı, 1996. Grube, Ernst. « Notes on Ottoman Painting in the
Contadini, Anna. « Ayyubid Illustrated Manuscripts 15th century ». In Essays in Islamic art and archi-
and their North Jaziran and ʿAbbasid Neighbours ». tecture: in honor of Katharina Otto-Dorn. Édité par
In Ayyubid Jerusalem. The Holy City in Context 1187- Abbas Daneshvari, 51-61. Malibu : Undena Publica-
1250. Édité par Robert Hillenbrand et Sylvia Auld, tions, 1981.
179-94. Londres : Altajir Trust, 2009. Grube, Ernst J. « Prolegomena for a Corpus Publication
Contadini, Anna. « The Manuscript as a Whole ». In of Illustrated Kalilah wa Dimna Manuscripts ».
Arab Painting: Text and Image in Illustrated Arabic Islamic Art 4 (1990-91) : 301-481.
Manuscripts. Édité par Anna Contadini, 3-16. Leiden : Grube, Ernst J. « Some Observations Concerning the
Brill, 2007. Ottoman Illustrated Manuscripts of the Kalilah wa
Déroche, François, éd. Manuel de codicologie des manus- Dimnah: Ali Çelebi’s Humayun-Name ». In Milletle-
crits en écriture arabe. Paris : Bibliothèque nationale rarası Türk Sanatları Kongresi, 2 : 195-205. Ankara :
de France, 2000. Kültür Bakanlığı, 1995.
D’Hulster, Kristof. « “Sitting with Ottomans and Grube, Ernst J. « The Date of the Venice Iskandar-
Standing with Persians”: The Šāhnāma-i Türkī as nama ». Islamic Art 2 (1987) : 187-202.
a Highlight of Mamluk Court Culture ». In Egypt Haldane, Duncan. Mamluk Painting. Warminster : Aris
and Syria in the Fatimid, Ayyubid and Mamluk Eras & Phillips, 1978.
VI. Proceedings of the 14th and 15th International Hayder, Ali. « Ibn Makanis (sa vie, son œuvre) sous
Colloquium at the Katholieke Universiteit Leuven. les Mamlouks ». Thèse de doctorat, Université
Édité par Urbain Vermeulen et Kristof D’Hulster, Bordeaux 3, 1981.
229-56. Louvain : Peeters Publishers, 2010. Hillenbrand, Robert. « Mamlūk and Īlkhānid Bestiaries:
Elger, Ralf. « Badr al-Dīn Muḥammad al-Ġazzī ». In Convention and Experiment ». Ars Orientalis 20
Essays in Arabic Literary Biography, 1350-1850. Édité (1990) : 149-87.
par Joseph E. Lowry et Devin J. Stewart, 2 : 98-106. Hirschler, Konrad. Medieval Damascus: Plurality and
Wiesbaden : Harrassowitz, 2009. Diversity in an Arabic Library: The Ashrafiya Library
Erünsal, İsmail E. Ottoman Libraries: A Survey of Catalogue. Edinburgh : Edinburgh University Press,
the History, Development and Organization of 2016.
Ottoman Foundation Libraries. Cambridge : Harvard Holter, Knut. « Die frühmamlukische miniaturmale-
University, 2008. rei ». Die graphischen Künste, no 2 (1978) : 1-14.
Fetvacı, Emine. « Ottoman Author Portraits in the Early- Hunt, Lucy-Anne. « Manuscript Production by Chris-
Modern Period ». In Affect, Emotion, and Subjectivity tians in 13th-14th Century Greater Syria and Meso-
in Early Modern Muslim Empires: New Studies in potamia and Related Areas ». Aram Periodical 9-10
Ottoman, Safavid, and Mughal Art and Culture. Édité (1997) : 289-326.
par Kishwar Rizvi, 66-94. Leiden : Brill, 2017. Hunt, Lucy-Anne. « Christian-Muslim Relations in
Fuess, Albrecht. « Sultans with horns: The political Painting in Egypt of the Twelfth to Mid-Thirteenth
significance of headgear in the Mamluk Empire ». Centuries: Sources of Wall Painting at Deir es-Suriami
Mamluk Studies Review 2, no 12 (2008) : 71-94. and the illustration of the New Testament ms Paris,
Gelder, Geert J. van, éd. Classical Arabic Literature: A Copte-Arabe 1/Cairo, Bible 94 ». Cahiers archéolo-
Library of Arabic Literature Anthology. New York : giques 33 (1985) : 111-55.
New York University Press, 2013. Ibn Iyās, Muḥammad b. Aḥmad, Badāʾiʿ al-zuhūr fī
Grabar, Oleg. « What Does “Arab Painting” Mean? » In waqāʾiʿ al-duhūr. Edité par Muḥammad Muṣṭafā.
Arab Painting: Text and Image in Illustrated Arabic Fīsbādin : Frānz Shtāynar, 1961-1975.
354 El Khiari et al.
Ibn Khallikān, Aḥmad b. Muḥammad Šams al-Dīn. Ibn Meredith-Owens, George M. Turkish miniatures.
Khallikan’s Biographical Dictionary. Traduit par Mac Londres : Trustees of the British Museum, 1963.
Guckin De Slane. Paris : Printed for the Oriental Milstein, Rachel, et Bilha Moor. « Wonders of a
translation fund of Great Britain and Ireland, 1843. Changing World: Late Illustrated ʿAjaʾib Manuscripts
Inal, Güner. « Kahire’de yapılmış bir Hümâyünnâme’nin (Part I) ». Jerusalem Studies in Arabic and Islam 32
minyatürleri ». Türk Tarih Kurumu Belleten 40, no 159 (2006) : 1-48.
(1976) : 439-65. Muralha, Vânia S.F., Lucia Burgio, et Robin J.H. Clark.
Jurado-López, Alicia, Ornela Demko, Robin J.H. Clark, et « Raman spectroscopy analysis of pigments on
David Jacobs. « Analysis of the Palette of a Precious 16-17th c. Persian manuscripts ». Spectrochimica Acta
16th Century illuminated Turkish Manuscript by Part A: Molecular and Biomolecular Spectroscopy 92
Raman Microscopy ». Journal of Raman Spectroscopy (2012) : 21-28.
35, no 2 (2004) : 119-24. Muslu, Cihan Yüksel. « Ottoman-Mamluk Relations
Kafescioǧlu, Çiğdem. « “In The Image of Rūm”: Ottoman and the Complex Image of Bāyezīd II ». In Conquête
Architectural Patronage in Sixteenth-Century Ottomane de l’Égypte (1517). Arrière-Plan, Impacts,
Aleppo and Damascus ». Muqarnas 16 (1999) : 70-96. Échos. Édité par Benjamin Lellouch et Nicolas
Knipe, Penley, Katherine Eremin, Marc Walton, Agnese Michel, 49-76. Leiden : Brill, 2013.
Babini, et Georgina Rayner. « Materials and Tech- Muslu, Cihan Yüksel. « Patterns of Mobility between
niques of Islamic Manuscripts ». Heritage Science 6, Ottoman and Mamluk Lands ». In The Mamluk
no 55 (2018). Sultanate from the Perspective of Regional and
Lellouch, Benjamin. Les Ottomans en Égypte : historiens World History. Édité par Reuven Amitai et Stephan
et conquérants au XVIe siècle. Paris : Peeters, 2006. Conermann, 391-430. Göttingen : V&R unipress, 2019.
Lellouch, Benjamin, et Nicolas Michel. « The Ottoman Muslu, Cihan Yüksel. The Ottomans and the Mamluks:
Conquest of Egypt and the Arts ». In Introduction : les Imperial diplomacy and warfare in the Islamic world.
échelles de l’événement. Édité par Benjamin Lellouch Londres : I.B. Tauris, 2014.
et Nicolas Michel, 1-50. Leiden : Brill, 2013. Necipoğlu, Gülru. « The Spatial Organization of
Lellouch, Benjamin et Nicols Michel, éd. Conquête otto- Knowledge in the Ottoman Palace Library: An
mane de l’Égypte (1517) : arrière-plan, impact, échos. Encyclopedic Collection and Its Inventory ». In
Leiden : Brill, 2013. Treasures of Knowledge: An Inventory of the Ottoman
Loiseau, Julien. « La ville démobilisée : Ordre urbain et Palace Library (1502/3-1503/4). Édité par Gülru
fabrique de la ville au Caire avant et après 1517 ». In Necipoğlu, Cemal Kafadar, et Cornell H. Fleischer,
Conquête ottomane de l’Égypte (1517) : arrière-plan, 1 : 1-77. Leiden : Brill, 2019.
impact, échos. Édité par Benjamin Lellouch et O’Kane, Bernard. Early Persian Painting: Kalila and
Nicolas Michel, 269-84. Leiden : Brill, 2013. Dimna Manuscripts of the Late Fourteenth Century.
Loveday, Helen. Islamic Paper: A Study of the Ancient Londres : I.B. Tauris, 2003.
Craft. Londres : The Don Baker Memorial Fund, 2001. Ökten, Ertuğrul. « Scholars and Mobility: A Preliminary
Majda, Tadeusz. « lllustracje do ʿAdzaʾib al-mahlukat Assessment from the Perspective of al-Shaqāyiq
Al-Kazwiniego. Tureckie malarstwo miniaturowe w al-Nuʿmāniyya ». Osmanlı Araştırmaları 41 (2013) :
Kairze, Bagdadzie i Damaszku w XVI-XVIII wieku ». 55-70.
In Z Mekki do Poznania: materiały 5. Ogólnopolskiej Papoutsakis, Nefeli. « Ibn al-Habbāriyya ». In Encyclo-
Konferencji Arabistycznej, Poznań, 9-10 czerwca paedia of Islam, THREE. Édité par Kate Fleet, Gudrun
1997. Édité par Henryk Jankowski, 177-83. Poznań : Krämer, Denis Matringe, John Nawas, Everett Row-
Uniwersytet im. Adama Mickiewicza w Poznaniu. son. Consulté en ligne le 2 décembre 2018.
Katedra Orientalistyki i Bałtologii, 1998. Petry, Carl F. « Travel Patterns of Medieval Notables in
the Near East ». Studia Islamica, no 62 (1985) : 53-87.
Entre deux mondes 355
Pfeifer, Helen. « Encounter after the Conquest: In At the Crossroads of Empires: 14th-15th Century
Scholarly Gatherings in 16th-Century Ottoman Eastern Anatolia. Proceedings of the International
Damascus ». International Journal of Middle East Symposium Held in Istanbul, 4th-6th May 2007, 221-38.
Studies 47 (2015) : 219-39. Istanbul : Institut Français d’Études Anatoliennes,
Raby, Julian. « Between Sogdia and the Mamluks: 2012.
A Note on the Earliest Illustrations to Kalila wa Tanındı, Zeren. « The Illustration of the Shahnama
Dimna ». Oriental Art 33, no 4 (1991) : 381-98. and the Art of the Book in Ottoman Turkey ». In
Raby, Julian. « Pride and Prejudice: Mehmed the Shahnama Studies II. The Reception of Firdausi’s
Conqueror and the Italian Portrait Medal ». Studies Shahnama. Édité par Charles Melville et Gabrielle
in the History of Art 21 (1987) : 171-94. van der Berg, 141-58. Leiden : Brill, 2012.
Raby, Julian. Venice, Dürer, and the oriental mode. Tanındı, Zeren. « Two Bibliophile Mamluk Emirs:
Londres : Islamic Art Publications, 1982. Qansuh the Master of the Stables and Yashbak
Raymond, André. Artisans et commerçants au Caire au the Secretary ». In The Arts of the Mamluks in
XVIIIe siècle. Le Caire : Institut français d’archéolo- Egypt and Syria: evolution and impact. Édité par
gie orientale, 1999. Doris Behrens-Abouseif, 267-81. Göttingen : Bonn
Richard, Francis. « Forgeries IV. Of Islamic Manus- University Press, 2012.
cripts ». In Encyclopædia Iranica, 2011, consulté en Titley, Norah. « Early Ottoman Miniature Painting:
ligne le 20 avril 2020. Two Recently Acquired Manuscripts in the British
Ross, Edward D. « An Arabic and a Persian metrical Library ». The British Library Journal 9, no 2 (1983) :
version of Burzoe’s autobiography from “Kalila 124-39.
and Dimna” ». Bulletin of the School of Oriental and Uluç, Lâle. Turkman Governors, Shiraz Artisans and
African Studies 4, no 3 (1927) : 441-72. Ottoman Collectors: Sixteenth Century Manuscripts.
Rührdanz, Karin, éd. Orientalische illustrierte Hand- Istanbul : Türkiye İş Bankası, 2006.
schriften: aus Museen und Bibliotheken der Deutschen Vecellio, Cesare. Costumes anciens et modernes. 4e éd.
Demokratischen Republik Ausstellung im Islamischen Paris : Typographie de Firmin Didot frères, 1860.
Museum der Staatlichen Museen zu Berlin, Haupt- Walz, Terence. « The Paper Trade of Egypt and the
stadt der DDR. Berlin : Staatliche Museen zu Berlin, Sudan in the Eighteenth and Nineteenth Centuries
1984. and its Re-Export to the Bilād as-Sūdān ». The
Soudavar, Abolala. « The Concepts of ‘al-Aqdamo Trans-Saharan Book Trade. Manuscript, Culture,
Aṣaḥḥ’ and ‘Yaqin-e Sābeq’, and the Problem of Arabic Literacy and Intellectual History in Muslim
Semi-fakes ». Studia Iranica 28, no 2 (1999) : 255-73. Africa, 8 (2010) : 73-107.
Sprengling, Martin. « Kalīla Studies ». American Journal Walzer, Sofie. « An Illustrated Leaf from a Lost Mamluk
of Semitic Languages and Literatures, no 40 (1924) : Kalilah wa Dimnah Manuscript ». Ars Orientalis 2
81-97. (1957) : 503-5.
Tanindi, Zeren. « Additions to Illustrated Manuscripts Walzer, Sofie. « The Mamlûk Illuminated Manuscript of
in Ottoman Workshops ». Muqarnas 17 (2000) : 147. Kalîla wa-Dimna ». In Aus der Welt der islamischen
Tanındı, Zeren. « Osmanlı Yönetimindeki Eyaletlerde Kunst: Festschrift für Ernst Kühnel zum 75 Geburstag
Kitap Sanatı ». In Ortadoğu’da Osmanlı Dönemi am 26-10-1957. Édité par Richard Ettinghausen, 195-
Kültür İzleri Uluslar Arası Bilgi Şöleni Bildirileri (25- 206. Berlin : Gebr. Mann, 1959.
27 Ekim 2000 – Hatay, 28 Ekim 2000 – İskenderun), Winter, Michael. « Chapter Five. Ottoman Qāḍīs in
2 : 501-8, 767-70. Ankara : Atatürk Kültür Merkezi Damascus in the 16th-18th Centuries ». In Law,
Başkanlığı Yayınları, 2001. Custom, and Statute in the Muslim World. Studies
Tanındı, Zeren. « The Arts of the Book: Patrons and in Honor of Aharon Layish. Édité par Ron Shaham,
Interactions in Erzincan between 1365 and 1410 ». 87-109. Leiden : Brill, 2007.
356 El Khiari et al.
Witkam, Jan J. « Increasing the Value or Hiding Anatolia in the Late Fourteenth Century ». Journal of
the Defects? The Case of the Stuttgart Hātifī Islamic Studies 25 (2014) : 263-97.
Manuscript ». Journal of Islamic Manuscripts 8, no 2 Yoltar-Yıldırım, Ayşi̇n. « A 1498-99 Khusraw va Shīrīn:
(2017) : 240-53. Turning the Pages of an Ottoman Illustrated
Wust, Efraim. Catalogue of the Arabic, Persian and Manuscript ». Muqarnas 22 (2005) : 95-109.
Turkish Manuscripts of the Yahuda Collection of the Yoltar-Yıldırım, Ayşi̇n. « The Role of Illustrated
National Library of Israel. Leiden : Brill, 2020. Manuscripts in Ottoman Luxury Book Production:
Yıldız, Sara Nur. « From Cairo to Ayasuluk: Haci Pasa 1413-1520 ». Thèse de doctorat, New York University,
and the Transmission of Islamic Learning to Western 2003.
12
Une version illustrée de Kalīla wa-Dimna en turc ottoman
Francis Richard
Le grand mouvement de traduction en turc otto- La version persane du Kalīla revue par Kāshifī
man des œuvres persanes classiques qui a atteint semble avoir été très tôt connue dans la capitale
son apogée au cours du xvie siècle ne pouvait ottomane. La bibliothèque Süleymaniye d’Istan-
laisser de côté le Kalīla wa-Dimna, classique de la bul possède en tout cas un manuscrit persan, n°
prose persane et modèle de rédaction bien connu 4215 du fonds Ayasofya, qui est une copie datée de
de tous les lettrés. Du reste le texte venait de faire 897H/1492 du texte de Kāshifī, réalisée du vivant
l’objet d’une nouvelle réécriture en persan à la cour même de l’auteur, qui mourut en 910H/15052.
timouride de Hérat dans un style plus moderne, Mais les bibliothèques turques conservent
durant les toutes dernières années du xve siècle. un certain nombre d’exemplaires3 d’une traduc-
Cette nouvelle version persane, les Lumières de tion turque ottomane des Anvār-i Suhaylī, qui
Canope (Anvār-i Suhaylī), qui sera d’ailleurs par- est dédiée à Soliman le Magnifique (Süleymān
tiellement connue en France dès 1640 grâce à Kānūnī, 1520-1566). Cette version, qui est inti-
David Sahid d’Ispahan et à Gilbert Gaulmin, est tulée Hümāyūnnāme, connut elle aussi, une
l’œuvre de Ḥusayn Vāʿiz Kāshifī. Dédiée à Aḥmad grande célébrité. Elle est due à ʿAlā al-Dīn ʿAlī
Suhaylī, vizir du sultan Ḥusayn Mīrzā Bayqarā, elle Çelebī b. Sāliḥ al-Felibevī al-Rūmī al-ʿUthmānī
suit un plan un peu différent de la version médié- al-Faqīh al-Ḥanafī al-Qāzī al-Müderris al-Adīb
vale et est enrichie de quelques anecdotes. Il en al-shāʿir al-Khattāt, dit Sālihzādeh, qui mourut en
existera une copie illustrée très célèbre, réalisée 950H/1543 à Bursa. La date exacte de la traduction
en Inde un siècle plus tard, en 1588, pour la biblio- ne semble pas connue4. C’est ce Hümayūnnāme –
thèque de l’empereur Akbar, copie qui témoigne qui conserve tels quels les distiques persans
du succès de cette version revue1. insérés dans la version persane – qui fut ensuite
largement utilisé comme modèle de turc ottoman
en prose ornée. Il n’est donc pas étonnant qu’il ait
1 Le manuscrit, qui date des alentours de 1575, comporte aussi été largement diffusé en Europe auprès de
plus de 230 illustrations et a été réalisé sans doute sous la tous ceux qui voulaient apprendre le turc. Ainsi
direction de ʿAbd al-Samad. Conservé à Poona dans la col-
lection Elphinsone, il s’est trouvé démembré et un certain
nombre de feuillets sont conservés dans l’ancien Musée à Qazvin ou à Ispahan, peut être à l’intention de Shāh
Prince of Wales de Mumbai ; voir Karl Khandalavala ʿAbbās Ier.
et Kalpana Desai, « Indian Illustrated Manuscripts of 2 Adnan Karaismailoğlu, « Hüseyn Vâiz-i Kâşifî », in Türkiye
the Kalilah Wa Dimna, Anvar- Suhaili, Iyar-i Danish Diyanet Vakfi islâm Ansiklopedisi (Istanbul : Türkiye
and Humayun-Nameh », in A Mirror for Princes from Diyanet Vakfı Yayınları Çamuroğlu, 1999), 16-18.
India: Illustrated Versions of the Kalilah Wa Dimnah, 3 Ali R. Karabulut, Istanbul ve Anadolu kütüphanelerinde
Anvar-i Suhayli, Iyar-i Danish, and Humayun Nameh, éd. mevcut el yazması eserler ansiklopedisi (Kayseri : Akabe
Ernst J. Grube (Bombay : Marg Publications, 1991), 128- Kitabevi, 2005), n° 3144 (3), 997.
144 ; Som P. Verma, « Anwar-i Suhaili – An Illustrated 4 Un manuscrit non illustré dont la copie date de 948H/1542
Manuscript of Akbar’s Court in the Collection of Bharat a été vendu chez Christie’s à Londres le 7 octobre 2011 :
Kala Bhavan, Varanasi », Roopa Lekha 14, no 1-2 (1977) : Art & Textiles of the Islamic & Indian Worlds Including
17‑22. Il s’agit là d’une commande impériale, de même que Works from the Collection of the Late Simon Digby, Christie’s
la très belle copie conservée dans la collection de l’Aga 7 October 2011, London, South Kensington (Londres :
Khan et qui a été réalisée sans doute par Sādiqī en 1593 Christie’s, 2011), n° 36.
peut-on en trouver plusieurs copies manuscrites « ex cubiculo Orientalium 1761 ». On trouve dans
dans la bibliothèque de l’École des Langues orien- le manuscrit nombre des signatures d’anciens
tales de Paris (Bibliothèque universitaire des lan- élèves drogmans ou Jeunes de langues qui l’eurent
gues et civilisations, BULAC, depuis 2011). Parmi entre leurs mains : Flourat, Deval, Ardouin, Ybary,
ces copies du Hümayūnnāme, on peut citer un Digeon « de l’isle de Chio », à nouveau Deval en
manuscrit non daté (coté Turc 111) de la fin du xvie 1772, puis Fonton en 1780 (à Smyrne puis Alep) et
siècle, qui possède un beau frontispice enluminé Fornetti en 1782. Ce volume possède une autre par-
et qui avait figuré dans la bibliothèque royale safa- ticularité : outre son frontispice enluminé – que
vide. À son feuillet 10 figurent plusieurs marques l’on voit au f.1v – il aurait dû être illustré d’environ
de ʿarz-dīdeh5 de bibliothécaires iraniens : l’une quatre-vingt-quinze peintures dont les emplace-
d’elles datée de 1117H/1705-6 qualifie le papier de ments ont été laissés vides par le copiste7 et qui
samarqandī et stipule que le manuscrit provient n’ont, pour une raison inconnue, jamais été exé-
des biens (amvāl) d’Allāh-vardī Khān, ancien cutées. On ignore dans quel atelier ce manuscrit
beglerbegī (gouverneur) de Shirvân. Le volume a aurait été illustré.
ensuite appartenu à divers dignitaires ottomans, Mais on connaît d’autres copies illustrées du
parmi lesquels un certain Meḥmet Çelebī b. même Hümayūnnāme. Un des plus importants
Mustafā Çelebī. Une des autres copies, le ms. Turc de ces manuscrits est le manuscrit Revan 843 de
114, terminé en 1003H/1594-5 par le calligraphe la bibliothèque du Palais de Topkapı, qui com-
Mustafā b. Ibrāhīm, sans doute pour un person- porte la version de ʿAlī Sālihzādeh accompagnée
nage nommé ʿAlī b. ʿAbd al-Laṭīf (son cachet daté de très belles peintures, sans doute un exemplaire
de 999H/1590-1 figure au f.1) et qui comporte aussi de dédicace réalisé dans les ateliers du sultan8.
un frontispice enluminé, moins soigné, mérite À Londres, à la British Library9, le manuscrit
d’être citée car elle provient de l’École des Jeunes Add. 15 153, qui est du xvie siècle et dont les dix
de langue d’Istanbul et était destinée à l’usage des derniers feuillets ont été ajoutés en 997H/1589,
élèves. Antoine Ducaurroy, directeur de l’École a comporte cent soixante-cinq peintures. Ernst
inscrit au contreplat « acheté par Mr Ducaurroy Grube, qui a publié une étude sur un manuscrit
avec autorisation de Mr Ruffin pour l’École des illustré du Hümayūnnāme de ʿAlī Sālihzādeh10,
Jeunes de langue de Constantinople, sur les épar- remarque qu’il ne semble avoir existé en Turquie
gnes faites pendant les années 1803, 1804, 1805 et
1806 »… La copie aujourd’hui cotée Turc 112 à la 7 Ainsi aux ff.18, 23v, 28, 28v, 30v, 33, 36, 38v, 41v, 43, 47,
BULAC n’est également pas datée mais paraît de 52v, 55v, 58v, 59v, 65v, 71, 76v, 78, 80v, 81v, 88, 89v, 91, 92,
l’extrême fin du xvie siècle. Elle est aussi liée à 95, 99, 104, 107v, 111, 115, 118, 118v, 122v, 123v, 128v, 129v,
l’histoire de l’enseignement du turc6. Elle figurait 133, 138, 142, 147v, 149v, 150v, 158v, 162, 170v, 173, 173v, 184v,
185v, 188v, 189v, 190v, 191v, 196, 197, 199, 201v, 203v, 209,
dans la bibliothèque de la « classe des Orientaux » 213v, 214, 219v, 225v, 230v, 239v, 244, 246v, 247v, 249v,
du collège Louis le Grand à Paris et on lit au f.1 255, 256v, 260v, 264, 267, 269, 271v, 274v, 280v, 284, 288v,
291, 294v, 296, 297v, 299v, 302v, 307v, 319, 322v, 326v,
332v, 339v et 348.
5 Iraj Afshar, « ʿArz dans la tradition bibliothéconomique 8 Plusieurs de ces peintures sont reproduites par Ömer
irano-indienne », Scribes et manuscrits du Moyen-Orient, F. Akün, « ʿAlâeddin Ali Çelebi (ö. 950/1543) » in Türkiye
éd. François Déroche et Francis Richard (Paris : Biblio- Diyanet Vakfi Islâm Ansiklopedisi (Istanbul : Türkiye
thèque nationale de France, 1997), 331-343. Diyanet Vakfı Yayınları Çamuroğlu, 1989), 315-318.
6 L’orientaliste Antoine Galland possédait aussi une copie 9 Charles Rieu, Catalogue of the Turkish Manuscripts
du Hümayūnnāme de ʿAlī Sālihzādeh. Cette copie appar- in the British Museum (Londres : British Museum,
tint ensuite à Silvestre de Sacy et est aujourd’hui conservée 1888), 227.
à Manchester, voir Jan Schmidt, A Catalogue of the Turkish 10 Ernst J. Grube, « Some Observations Concerning the
Manuscripts of the John Rylands Library (Leiden : Brill, Ottoman Illustrated Manuscripts of the Kalilah Wa
2011), 112-113, n° 59. Dimnah: Ali Çelebi’s Humayun-Name », in Milletlerarası
Une version illustrée de Kalīla wa-Dimna en turc ottoman 359
que fort peu de copies illustrées du texte, toutes du céramique d’Iznik14. On y voit des motifs d’acanthe
xvie siècle, souvent dans un style provincial, tel le avec sur certaines pages de grandes feuilles allon-
ms. Hazine 359 de Topkapı11. gées, ainsi que d’amples volutes. Ce décor, bien
La découverte récente de quelques feuil- que relativement moins élaboré, rappelle un peu
lets venant d’un autre Kalīla va-Dimna en turc les marges du recueil des poèmes de Soliman le
conservé à l’École des Langues orientales dans les Magnifique (Dīvān-i Muhibbī), conservé à Topkapı,
fonds de la BULAC démontre pourtant le succès qui ont été réalisées par Kara Memi, disciple de
que ce texte a pu rencontrer dans l’Empire otto- Shāhqulī (actif à partir de 1545), en 156515. On peut
man à cette époque. Probablement ce succès est-il aussi évoquer d’autres manuscrits comme la copie
à mettre en rapport avec le développement d’une cotée Revan 804 du Dīvān de Navāʾī faite pour la
langue littéraire pouvant rivaliser avec les modèles bibliothèque de Soliman16. Cela permet de dater
persans. Ces quelques feuillets, qui ont reçu la cote les feuillets de la BULAC avec une certaine préci-
ms. Turc 205, sont d’un grand intérêt. Il y a en tout sion du règne de ce sultan ou du début de celui
quatorze feuillets à la BULAC, dont la plupart ne se de son successeur. Outre ce décor des marges, on
suivent pas12, et ni le début ni la fin du texte ne sont trouve quatre peintures, qui paraissent contempo-
présents. La version turque du Kalīla, avec – repro- raines de la copie. Ce sont respectivement :
duites telles quelles – les brèves pièces poétiques – au f.6v « le peintre et la jeune fille » (81 ×
en persan insérées dans le texte, est calligraphiée 79 mm) (Fig. 12.1). Les deux personnages s’en-
dans une belle écriture naskhī ottomane. Le texte tretiennent, assis dans une petite salle couverte
turc est vocalisé. Ce type d’écriture vocalisée n’est d’un dôme. Le décor est figuré assez sommaire-
pas rare dans les manuscrits de luxe ottomans de ment et une fenêtre ouvre sur un jardin planté
la seconde moitié du xvie siècle13. Chaque page d’œillets rouges. Le personnage masculin rap-
comporte dix-neuf lignes et la surface d’écriture pelle par son costume certains personnages
est de 80 × 175 mm, la page mesurant 157 × 250 d’un Gulistān de 1565 de la Freer Gallery17, mais
mm. Certains titres sont dorés, d’autres sont en surtout du Şehnāme-i Selim Hān de 1581 de
rouge. Une des caractéristiques de ces feuillets la bibliothèque de Topkapı18. Le personnage
est leur riche décor : le centre de la page est sablé féminin à la coiffe bien caractéristique peut être
d’or et comporte un double encadrement de deux
bandes dorées, l’une plus fine et l’autre plus large.
14 Il s’agit ici du type de décors « à la grande feuille saz »
Les marges sont décorées de motifs floraux d’au qui se rencontre sur de très nombreuses pièces de
moins trois types différents, peints en doré avec céramique d’Iznik entre 1560 et 1600, type décrit par
des rehauts de rouge et de bleu, dans un style otto- exemple dans Frédéric Hitzel et Mireille Jacotin, Iznik :
man qui évoque certains motifs présents sur la l’aventure d’une collection. Les céramiques ottomanes
du Musée national de la Renaissance, Château d’Écouen
(Paris : Réunion des musées nationaux, 2005), 89-103.
15 Nurhan Atasoy, Kara Memi (vol. 1) ; Muhibbî divani
Türk Sanatları Kongresi (Ankara : Kültür Bakanlığı, (vol. 2) (Istanbul : Masa Yayinlari, 2016).
1995), 195-205. 16 Ivan Stchoukine, La peinture turque d’après les manus-
11 Güner Inal, « Kahire’de yapılmış bir Hümâyünnâme’nin crits illustrés. Ière partie de Sulaymān Ier à Os̱mān II,
minyatürleri », Türk Tarih Kurumu Belleten 40, no 159 1520-1622 (Paris : Geuthner, 1966), recense aux pp. 51-62
(1976), 439-465. cinq copies de Navāʾī illustrées sous son règne (n° 10
12 Ainsi les ff.2 et 3 se font-ils suite, ainsi que les ff.6 et 5, à 13 et 23). Pour lui, cinq des peintures du ms. R.804,
puis les ff.8 à 12 et les ff.13 et 14. Les ff.4,5, 6 et 8 avaient commencé à Tabriz, ont été achevées à Istanbul dans
été maladroitement remontés et sont à l’envers. Le f.6 l’atelier du sultan entre 1535 et 1540.
est mutilé. 17 Serpil Bağcı et al., Osmanlı resim sanatı (Istanbul :
13 À titre d’exemple les Kırk Vezir Hikayesi de l’Istanbul T.C. Kültür ve Turizm Bakanligi Yayinlari, 2006), 115,
University Library (ms. T.7415), des environs de 1580, n° 78.
sont d’une écriture assez comparable. 18 Bağcı et al., Osmanlı resim sanatı, 116, n° 79.
360 Richard
figure 12.1 Fable du peintre et de la jeune fille, version turque ottomane de Kalīla wa-Dimna, Paris,
BULAC, ms. Turc 205, f.6v
Une version illustrée de Kalīla wa-Dimna en turc ottoman 361
daté de la même période19. Très simple, la repré- Certaines de ces peintures sont malheureuse-
sentation du dôme argenté correspond à ce que ment un peu abîmées, mais elles n’ont pratique-
l’on peut trouver dans nombre de manuscrits20. ment pas été retouchées. Les feuillets de ce Kalīla
– au f.7 « le renard et le tambour » (78 × 110 mm) va-Dimna ottoman sont entrés à la bibliothèque
(Fig. 12.2). Cette peinture, d’une composition de l’École des Langues orientales à une date
simple, montrant le renard en train de frapper inconnue, sans doute au xixe siècle. Ils étaient
le tambour suspendu à un arbre avec son visage, maladroitement cousus et couverts d’une feuille
est extrêmement proche d’une des illustrations manuscrite portant des notes de grammaire
du manuscrit persan de Kalīla copié en 1429 à turque. On peut supposer qu’ils figuraient dans
Hérat pour le sultan timouride Baysunghur par les papiers d’un professeur ou d’un ancien élève
Shams al-Dīn Baysunghurī21. Or ce manuscrit se de l’école. Séparé du précédent à une date que
trouve conservé à Istanbul dans la bibliothèque nous ignorons, un autre fragment, plus important,
de Topkapı sous la cote Revan 1022. Il est extrê- de ce même manuscrit est conservé aujourd’hui
mement vraisemblable que le peintre qui a à la Bibliothèque nationale de France. Acquis le
illustré cette version turque ait pu y avoir accès 5 février 1906 auprès du libraire Cornuau, boule-
et l’utiliser comme modèle, puisque le manus- vard Hausmann, sa provenance antérieure n’est
crit se trouvait déjà au xvie siècle dans le trésor malheureusement pas connue23. Il porte actuel-
impérial ottoman. lement la cote Supplément turc 1243 et le cata-
– au f.8 « le singe sur une planche » (83 × 79 mm) logue indique qu’il contient la fin du troisième
(Fig. 12.3). chapitre (bāb), la totalité du quatrième et le début
– au f.11v « Kalila et Dimna » (78 × 102 mm) (Fig. du cinquième24. Le fragment de la BnF possède
12.4). Sur cette peinture on doit noter particu- les mêmes caractéristiques codicologiques et les
lièrement la manière dont l’artiste représente mêmes décors marginaux. Il ne renferme ni le
les montagnes, que l’on retrouve dans nombre début ni la fin du texte et comporte trente-cinq
de manuscrits ottomans issus des ateliers impé- feuillets. Quelques feuillets sont mutilés. Sa reliure
riaux du dernier quart du xvie siècle, notam- en demi-parchemin est moderne et a été mise
ment le Zübdetüt-tevārih de Seyyed Lokman ou après son entrée à la Bibliothèque nationale. Le
le Firāsetnāme de Talikizāde (1583 et 1590)22. Les fragment de la BnF a conservé cinq peintures :
représentations d’arbres sont assez soignées – – au f.18, « le médecin ignorant et la fille du roi »
comme sur les autres images du manuscrit – et (Fig. 12.5). On serait volontiers tenté de com-
le tronc est figuré avec soin ; il s’agit d’une tradi- parer le turban un peu surélevé du médecin
tion différente de celle de la peinture persane à des turbans comparables que l’on voit dans
de la même époque. un manuscrit de 1584 du Nüsretnāme de Āli
(Topkapı, H. 1365)25. La coupole argentée du
19 Dans le Siyer-i Nabi de 1594-5, les costumes féminins
sont légèrement différents (Bağcı et al., Osmanlı resim
Sanatı, 158, n°122). Mais le contexte n’est pas tout à fait
le même. 23 Quelques feuillets non illustrés qui paraissent aussi
20 Par exemple, toujours dans Bağcı et al., Osmanlı resim venir de ce manuscrit sont conservés dans une collec-
sanatı, n° 82 et 85 (datant de 1579 et 1581). tion privée [renseignement aimablement communi-
21 Le manuscrit, qui porte la date de 833H/1429, a reçu la qué par Madame Éloïse Brac de la Perrière].
cote R. 1022 dans les collections de Topkapı. Voir par 24 Edgar Blochet, Catalogue des manuscrits turcs, tome II,
exemple Ivan Stchoukine, Les peintures des manus- Supplément nos 573-1419 (Paris : Bibliothèque nationale,
crits tîmûrides (Paris : Imprimerie nationale, 1954), 54, 1933), 209, où le texte est identifié à tort comme étant le
n° XXXV. Hümayūnnāme. Le manuscrit est intégralement repro-
22 Bağcı et al., Osmanlı resim sanatı, 137 et 177, n° 100 et duit sur le site Gallica.
145. 25 Bağcı et al., Osmanlı resim sanatı, 166-167, n° 132-133.
362 Richard
figure 12.2 Fable du renard et du tambour, version turque ottomane de Kalīla wa-Dimna, Paris, BULAC,
ms. Turc 205, f.7
Une version illustrée de Kalīla wa-Dimna en turc ottoman 363
figure 12.3 Fable du singe et du menuisier, version turque ottomane de Kalīla wa-Dimna, Paris, BULAC, ms.
Turc 205, f.8
364 Richard
figure 12.4 Kalila et Dimna discutent, version turque ottomane de Kalīla wa-Dimna, Paris, BULAC,
ms. Turc 205, f.11v
Une version illustrée de Kalīla wa-Dimna en turc ottoman 365
figure 12.5 Fable du médecin ignorant et de la fille du roi, version turque ottomane de Kalīla
wa-Dimna, Paris, BnF, Supplément turc 1243, f.18
366 Richard
bâtiment évoque la coupole de la scène du semblent se trouver dans les mains de collection-
peintre et de la jeune fille. neurs français, montre qu’il s’agit d’une version dif-
– au f.22v, « les deux perroquets et l’épouse du férente de celle du Hümayūnnāme. Or on connaît
souverain » (Fig. 12.6). Les trois personnages au moins deux autres traductions turques otto-
masculins qui s’entretiennent auprès de la cage manes des Anvār-i Suhaylī, différentes de celle de
ont une grande ressemblance, par leur costume ʿAlī Sālihzādeh. L’une est attestée par un manuscrit
et leur style, avec les personnages masculins de deux cent six feuillets orné de deux peintures
du manuscrit du Şehnāme-i Selim Hān de 1581 qui a figuré dans la vente Christie’s du 27 avril
(Bibliothèque de Topkapı, A. 3595)26. On peut 1993 à Londres. Cette version, que le catalogue
peut-être rattacher nos feuillets au même intitule Hümayūnnāme ou Sharaf al-insān, serait
atelier. due à Mahmūd Hamīdī. Elle a été copiée – dans ce
– au f.26 « la colombe au collier », où sont repré- manuscrit qui pourrait être l’original (?) – en 1526-
sentés les oiseaux pris dans les rets du chasseur 1527, du vivant du traducteur. Ce dernier, Mahmūd
(Fig. 12.7). Hamīdī, réalisa ce travail pour l’empereur Soliman
– au f.29v « la colombe au collier », où sont repré- (1520-1566) la sixième année de son règne (1526-
sentés le corbeau et le rat (Fig. 12.8). 7). Une traduction différente, elle aussi faite dans
– au f.33 « le rat, le corbeau, la tortue et l’anti- la première moitié du xvie siècle, est connue par
lope » : le chasseur git à terre avec le loup, le cerf une copie datée de 1068H/1657 conservée à la
et le sanglier (Fig. 12.9). Staatsbibliothek de Berlin, le ms. or. Oct. 2507,
De facture semblable aux précédentes, ces pein- Terjüma-i Kalīla va-Dimna29. Le traducteur, qui
tures se rattachent à l’école impériale ottomane de a suivi lui aussi le texte des Anvār-i Suhaylī, est
la seconde moitié du xvie siècle. Les costumes des Mustafā b. Ḥusāmī, qui était actif à la Madrasa
personnages pourraient encore évoquer ceux que Mürādiya d’Edirne. Il se peut fort bien que ces
l’on voit sur les peintures des manuscrits jusque deux traductions aient été faites antérieurement à
vers 1580. Le manuscrit nous semble certainement la rédaction du Hümayūnnāme.
antérieur au Siyer-i Nabi Hazine 1221 à 1223 de la Le manuscrit qui est aujourd’hui partiellement
Bibliothèque de Topkapı qui date, lui, de 159427. conservé à la BULAC et à la BnF pourrait se ratta-
Dans les différents volumes de ce Siyer-i Nabi, illus- cher à l’une de ces versions. Ce pourrait être aussi
trés sous la direction du naqqāsh-bāshi par plu- une version plus tardive, un peu postérieure à
sieurs artistes différents, on ne retrouve déjà plus celle de ʿAlī Sālihzādeh. Il s’agit, en tout cas, sinon
la manière de peindre de notre illustrateur. d’un exemplaire de dédicace destiné au sultan, au
On ignore tout de la façon dont ce manuscrit, moins d’une copie de luxe sortie de l’atelier impé-
ou ces fragments de manuscrit, sont parvenus en rial. Son iconographie semble reprendre assez
France. Tant les feuillets de la BnF que ceux de la fidèlement un ou plusieurs modèles persans anté-
B.U.L.A.C possèdent quelques marques ou gloses, rieurs connus, auquel l’artiste pouvait avoir accès
peu nombreuses, dans les marges28, ou quelques à Istanbul à son époque, les réinterprétant dans
chiffres d’une main occidentale. Le manuscrit a le style en vogue à la fin du règne de Soliman, ou
certainement été entre les mains d’un Français qui dans les années qui ont suivi. Quoi qu’il en soit,
en a lu le texte turc, peut-être un Jeune de langue. ce manuscrit est un témoin supplémentaire de
Mais son écriture n’a pu être identifiée jusqu’ici. l’intérêt porté au texte de Kalīla va-Dimna dans la
L’examen du texte de ce Kalīla va-Dimna otto- Turquie du xvie siècle et de la volonté de l’adapter
man jadis démembré, dont d’autres feuillets aux besoins du public lettré ottoman, pour qui la
figure 12.6 Fable des deux perroquets et l’épouse du souverain, version turque ottomane de
Kalīla wa-Dimna, Paris, BnF, Supplément turc 1243, f.22v
368 Richard
figure 12.7 Fable de la colombe au collier, version turque ottomane de Kalīla wa-Dimna, Paris,
BnF, Supplément turc 1243, f.26
Une version illustrée de Kalīla wa-Dimna en turc ottoman 369
figure 12.8 Fable de la colombe au collier, version turque ottomane de Kalīla wa-Dimna, Paris, BnF,
Supplément turc 1243, f.29v
370 Richard
figure 12.9 Fable du chasseur, version turque ottomane de Kalīla wa-Dimna, Paris, BnF,
Supplément turc 1243, f.33
Une version illustrée de Kalīla wa-Dimna en turc ottoman 371
tradition littéraire en prose jouait un rôle essen- national de la Renaissance, Château d’Écouen. Paris :
tiel, et de disposer ainsi d’un modèle de « beau Réunion des musées nationaux, 2005.
style » ne le cédant en rien au modèle timouride Inal, Güner. « Kahire’de yapılmış bir Hümâyünnâme’nin
de Hérat. minyatürleri ». Türk Tarih Kurumu Belleten 40, no 159
(1976) : 439-465.
Karabulut, Ali R. Istanbul ve Anadolu kütüphanele-
Bibliographie rinde mevcut el yazması eserler ansiklopedisi. vol. 2.
Kayseri : Akabe Kitabevi, 2005.
Afshar, Iraj. « ʿArz dans la tradition bibliothécono- Karaismailoğlu, Adnan. « Hüseyn Vâiz-i Kâşifî ». In
mique irano-indienne ». In Scribes et manuscrits du Türkiye Diyanet Vakfi islâm Ansiklopedisi, vol. 19 :
Moyen-Orient. Édité par François Déroche et Francis 16-18. Istanbul : Türkiye Diyanet Vakfı Yayınları
Richard, 331-343. Paris : Bibliothèque nationale de Çamuroğlu, 1999.
France, 1997. Khandalavala, Karl et Kalpana Desai. « Indian Illustrated
Akün, Ömer F. « ʿAlâeddin Ali Çelebi (ö. 950/1543) ». Manuscripts of the Kalilah wa Dimna, Anvar- Suhaili,
In Türkiye Diyanet Vakfi Islâm Ansiklopedisi, vol. 2 : Iyar-i Danish and Humayun-Nameh ». In A Mirror
315-318. Istanbul : Türkiye Diyanet Vakfı Yayınları for Princes from India: Illustrated Versions of the
Çamuroğlu, 1989. Kalilah wa Dimnah, Anvar-i Suhayli, Iyar-i Danish,
Atasoy, Nurhan. Kara Memi (vol. 1) – Muhibbī divani and Humayun Nameh. Édité par Ernst J. Grube, 128-
(vol. 2). Istanbul : Masa Yayinlari, 2016. 144. Bombay : Marg Publications, 1991.
Bağcı, Serpil, Filiz Çağman, Renda Günsel et Zeren Rieu, Charles. Catalogue of the Turkish Manuscripts in
Tanındı. Osmanlı resim sanatı. Istanbul : T.C. Kültür the British Museum. Londres : British Museum, 1888.
ve Turizm Bakanligi Yayinlari, 2006. Schmidt, Jan. A Catalogue of the Turkish Manuscripts of
Blochet, Edgar. Catalogue des manuscrits turcs, tome II, the John Rylands Library. Leiden : Brill, 2011.
Supplément nos 573-1419. Paris : Bibliothèque natio- Stchoukine, Ivan. La peinture turque d’après les manus-
nale, 1933. crits illustrés. Ière partie de Sulaymān Ier à Osmān II,
Götz, Manfred. Türkische Handschriften. Teil 2. Wiesba- 1520-1622. Paris : Geuthner, 1966.
den : Steiner, 1968. Stchoukine, Ivan. Les peintures des manuscrits tîmû-
Grube, Ernst J. « Some Observations Concerning the rides. Paris : Imprimerie nationale, 1954.
Ottoman Illustrated Manuscripts of the Kalilah Verma, Som P. « Anwar-i Suhaili – An Illustrated
wa Dimnah : Ali Çelebi’s Humayun-Name ». In Manuscript of Akbar’s Court in the Collection of
Milletlerarası Türk Sanatları Kongresi, vol. 2 : 195-205. Bharat Kala Bhavan, Varanasi ». Roopa Lekha 14, no
Ankara : Kültür Bakanlığı, 1995. 1-2 (1977) : 17-22.
Hitzel, Frédéric et Mireille Jacotin. Iznik : l’aventure
d’une collection. Les céramiques ottomanes du Musée
13
The Patron and the Author: An Illustration from the Iyār-i Dānish
Reveals the Relationship between Emperor Akbar and Abū’l Fażl
Mika Natif
Bridging peoples, places, and times, the book of text and of its circulation, adaptations, and addi-
Kalīla wa-Dimna in its various renditions is an tions, as is the case in Ibn al-Muqaffaʿ’s Arabic
astounding testimony to cultural connections translation and Kamāl al-Dīn Ḥusayn Vāʿiẓ Kāshifī’s
between civilizations.1 With its various prefaces Anvār-i Suhaylī. By the end of the sixteenth cen-
and introductions, the fables tell stories that tury, the text had come full circle, resulting in the
uncover different layers of narration. They con- rendition composed by Abū’l Fażl for the Emperor
tain anecdotes related to life lessons and moral- Akbar in northern India.
ity, enacted by animals and humans. However, it Based on the Sanskrit text of the Pañcatantra,
is of even more interest to note the ways in which the stories became well known in their Arabic and
the text delineates narratives about the transmis- Persian translations, and many of these manu-
sion of knowledge, and about translations and scripts were also illustrated. Among these medi-
adaptations. Many of the subsequent authors eval texts, Abū al-Maʿālī Naṣrallāh Munshī’s Persian
who translated or rewrote the fables comment on rendition, written in the middle of the twelfth
and acknowledge the history of the text and how century, and Kāshifī’s Anvār-i Suhaylī, from the
it had traveled from India to Iran. These nods to late fifteenth century, also became popular on the
the past disclose a fascinating chain of transmis- Indian subcontinent. It appears that some mem-
sion from culture to culture, from author to author, bers of the Mughal elite were not content with the
and from patron to patron. This essay examines Timurid version of the text written by Kāshifī.2 In
the close relationship between the Mughal Indian the late 1580s, looking to simplify Kāshifī’s florid
ruler-patron Emperor Akbar (r. 1556–1605) and his Persian language, the third Mughal emperor, Jalāl
author-vizier Abū’l Fażl b. Mubārak (1551–1602), al-Dīn Muḥammad Akbar, commissioned his min-
and its visual manifestation in the opening illus- ister and historiographer, Abū’l Fażl, to rewrite
tration to the Iyār-i Dānish, Abū’l Fażl’s version of the Kalila and Dimna tales.3 The new version
the Kalīla wa-Dimna stories. The painting reveals
Abū’l Fażl and Akbar’s awareness of the processes 2 The Anvār-i Suhaylī, a late fifteenth-century recension of
of translation, assimilation, and the Mughal for- the famous animal fables, was composed by the Timurid
mation of self-identity in a multi-faceted society. theologian and preacher Kamāl al-Dīn Ḥusayn Vāʿiẓ
The stories of Kalila and Dimna, which were Kāshifī. It became very popular among the Mughal elite
during Akbar’s reign, and at least five manuscripts were
translated into different languages over the centu-
copied and illustrated. On the text of the Anvār-i Suhaylī,
ries, have a long history in the Muslim world. In see Christine van Ruymbeke, Kashefi’s Anvar-e Sohayli:
their prefaces, some of the later author-translators Rewriting Kalila and Dimna in Timurid Herat (Leiden:
express an awareness of the Indian origin of the Brill, 2016).
3 Based on his study of MS Add 4945, now in the British
Library, Charles Rieu established that the Iyār-i Dānish
1 François de Blois, Burzōy’s Voyage to India and the Origin was completed in the thirty-third year of Akbar’s reign,
of the Book of Kalīlah wa Dimnah (London: Royal Asiatic 996 AH/1588 CE. Charles Rieu, Catalogue of the Persian
Society, 1990), iii. Manuscripts in the British Museum (London: British
was entitled the Iyār-i Dānish (The Touchstone of adds very little embellishment to his text. This
Wisdom) by Abū’l Fażl, who explains: “By order type of writing stands in stark contrast to his
of His Majesty, the author of this volume com- style in the official Mughal court histories, such
posed a new version of the Kalílah Damnah, and as the Akbar-nāma or the Āīn-i Akbarī. In the Āīn-i
published it under the title of ʿAyár Dánish.”4 It is Akbarī, Abū’l Fażl explains: “The original [Kalīla
worth noting that Abū al-Fażl’s version did not win wa-Dimna] is a master-piece of practical wisdom,
the popularity of the Anvār-i Suhaylī or Naṣrallāh but is full of rhetorical difficulties; and though
Persian version of the Kalīla va Dimna. Naçrullah i Mustaufi and Maulana Husain i Waʿiz
Although Kāshifī’s recension was well known had translated it into Persian, their style abounds
on the subcontinent, the core stories in Abū’l Fażl’s in rare metaphors and difficult words.”6 But who
text were more closely related to Ibn al-Muqaffaʿ’s would have had difficulty understanding the high-
eighth-century medieval Arabic narrative. Unlike flown language of Naṣrallāh and Kāshifī, especially
Naṣrallāh and Kāshifī, Abū al-Fażl parted from when numerous copies of the Anvār-i Suhaylī
much of the Persian poetry that was so important and the Kalīla va Dimna exist from this period?7
in the Anvār-i Suhaylī, and minimized the use of Christine van Ruymbeke suggests that Abū’l Fażl’s
Arabic and Qurʾanic quotations, which were pres- simple version of the Kalila and Dimna stories
ent in Kalīla wa-Dimna. Adhering more closely to was probably aimed at non-native Persian speak-
the basic narrative, Abū’l Fażl removed the layers ers, who may not have been able to understand
of commentary and sophistication that his prede- Kāshifī’s florid language. Calling Abū’l Fażl’s Iyār-i
cessors had added to the text.5 Dānish a “summary Mughal rewriting” of Kāshifī’s
Abū’l Fażl’s Iyār-i Dānish is by no means elegant text, Van Ruymbeke further hypothesizes that the
or artistic. It is unlike his other compositions for Mughal author created this recension of the sto-
Akbar; here, he is not playful with words and he ries for his own intellectual purposes, so that he
could easily compare his Persian version with the
Museum, 1879–83), 2:756–57. This date has been accepted
original Sanskrit Pañcatantra.8
by most scholars. See Hermann Ethé, Catalogue of Persian One of the most surprising aspects is that the
Manuscripts in the Library of the India Office (Oxford: spare, skeletal Iyār-i Dānish was lavishly illus-
Printed for the India Office by H. Hart, 1903), vol. 1, 512–13; trated in Akbar’s royal Mughal atelier in Lahore.9
Carl Brockelmann, “Kalila Wa-Dimna,” in Encyclopaedia
Recognized as one of the masterpieces produced
of Islam, 2nd ed. (Leiden: Brill), vol. 4, 503–506; Saiyid
Athar Abbas Rizvi, Religious and Intellectual History of by Akbar’s workshops, the Iyār-i Dānish manu-
the Muslims in Akbar’s Reign, with Special Reference to script demonstrates the sophistication and com-
Abu’l Fazl (1556–1605) (New Delhi: Munshiram Manoharlal plex tastes of the upper-class Mughal society of
Publishers, 1975), 219–20; Linda York Leach, Mughal and
Other Indian Paintings from the Chester Beatty Library
(London: Scorpion Cavendish, 1995), vol. 1, 75; Christine 6 Āīn-i Akbarī, vol. 1, 106.
van Ruymbeke, “Authorship, Ownership and Rewriting: 7 For further information regarding the surviving illus-
Vāʿiẓ Kāshifī and Abū al-Fażl b. Mobārak within the Noble trated manuscripts, see Ernst J. Grube, “Prolegomena for
Hereditary Line of Kalīla wa-Dimna Authors,” Jerusalem a Corpus Publication of Illustrated Kalilah wa Dimnah
Studies in Arabic and Islam 45 (2018): 181–210, 1, forthcom- Manuscripts,” Islamic Art 4 (1991): 301–453.
ing. I am very grateful to Christine van Ruymbeke for shar- 8 Van Ruymbeke remarks: “Prosimetrum’s use and impor-
ing her invaluable article with me. tance is remarkably highlighted in absentia: Abu’l-Fazl’s
4 Abu’l Fazl, Ain-i Akbari, trans. H. Blochmann (Calcutta: voice is silent, his text is bare, lacking in truly authorial
Printed for the Asiatic Society of Bengal, 1927–1949), activity. The readers are not given a key to decode the sto-
vol. 1, 106. Ayar means a strong stone. ʿAlī Akbar Dihkhudā, ries, which have returned to the state of a blank canvas.”
Lughatnāmah, Tālīf-i ʿAlī Akbar Dihkhudā (Tehran: Van Ruymbeke, Kashefi’s Anvar-e Sohayli, 247–48.
Sāzmān-i Lughat-nāmah, 1337–1352 [1959–1974]), 3:27. 9 Between 1570 and 1611, at least six Anvār-i Suhaylī manu-
5 See the in-depth analysis by Van Ruymbeke in Kashefi’s scripts were illustrated in Mughal India (and only one
Anvar-e Sohayli. Iyār-i Dānish).
374 Natif
the 1590s and its aristocratic literati.10 Illustrated scene at the court of the Sassanian King Khusraw
manuscripts of the Kalila and Dimna stories have Anūshirvān (r. 531–79), with Burzūya the physician
often been considered to function as “textbooks” (Fig. 13.1).15 Abū’l Fażl’s text tells us how the sto-
for the education of princes.11 In the Āīn-i Akbarī, ries of Kalila and Dimna came from India to Iran,
Abū’l Fażl reports that “every boy ought to read and were then translated into Pahlavi. He writes
books on morals, arithmetic, agriculture, geom- that King Khusraw Anūshirvān heard a story from
etry, astronomy, physiognomy, household matters, a man about a special plant, said by the Greeks to
the rules of government, medicine, logic, sciences, grow in the mountains of Hindustan. This won-
and history …”12 He also mentions in his preface to drous plant had the ability to keep people alive
the Iyār-i Dānish that the book was intended for and make them immortal. The man recounted
the education of kings and the people.13 However, that after searching in vain for the magical plant,
the illustrations in the Iyār-i Dānish seem too he came across an old Brahman sage, who revealed
sophisticated and complex to be aimed solely at to him that the mysterious plant was actually a
the education of the young elite. Furthermore, book of wisdom kept in the treasuries of the kings
their subject matter, as well as their intricate com- in India. The name of the book was “Kalīla wa-
positions, cannot be understood in such a con- Dimna.” When Khusraw Anūshirvān heard this,
strained way. The manuscript’s text and images he decided to send Burzūya the physician to find
place the exotic stories within more familiar ter- the book and bring it to the court.16 With the sup-
ritory: the Mughalization of all figures, including port of his patron, Burzūya went to Hindustan and
their physical settings, and the geography, flora, approached the king, asking if he could see the
and fauna the recording of Mughal court rituals books of wisdom that he possessed. After obtain-
and of historical individuals of the Akbari milieu. ing access to this hidden knowledge, Burzūya
The illustration cycle in the Varanasi Iyār-i read the books every day, committed them to
Dānish commences with a painting on the fifth memory, and copied them out in secret. Once he
folio made by the artist Basawan.14 It shows a had finished reading and transmitting the texts,
he brought them back to the court of Anūshirvān.
10 Dated by its colophon to Lahore, 1596–97, the Iyār-i One of the books was Kalīla wa-Dimna. Upon
Dānish manuscript Ms.9069 is housed in the Bharat
Kala Bhavan, Hindu University of Benares (Varanasi).
11 Som Prakash Verma, “Anvar-i Suhayli – an Illustrated century, to the original volume. Most of the Mughal
Manuscript of Akbar’s Court in the Collection of Bharat paintings in this manuscript had formal attributions
Kala Bhavan, Varanasi,” Roopa-Lekha 44, no. 1–2 (1977): written below the illustrations. When the manuscript
17–22; Grube, “Prolegomena for a Corpus Publication,” was re-margined, these attributions were cut out and
315; Karl Khandalavala and Kalpana Desai, “Indian pasted onto the back of the pictures, while the text
Illustrated Manuscripts of the Kalilah wa Dimnah, pages were separated from the illustrations. For the
Anvar-i Suhayli, and Iyar-i Danish,” in A Mirror for whereabouts of the manuscript from the eighteenth
Princes from India: Illustrated Versions of the Kalilah wa century onward, see Krishnadasa, Anwar-e-Suhaili
Dimnah, Anvar-i Suhayli, Iyar-i Danish, and Humayun (Iyar-i-Danish), 34–35.
Nameh, edited by Ernst J. Grube (Mumbai: Marg 15 Published: Bonnie C. Wade, Imaging sound: an eth-
Publications, 1991), 128–144, here 137; Rai Krishnadasa, nomusicological study of music, art, and culture in
Anwar-e-Suhaili (Iyar-i-Danish) (Varanasi: Bharat Kala Mughal India (Chicago: University of Chicago Press,
Bhavan, Banaras Hindu University, reprinted in 1999). 1998), fig. 110; Stuart Cary Welch, India: Art and Culture
12 Ain-i Akbari, vol. 1, 288–9. 1300–1900 (New York: Metropolitan Museum of Art,
13 Abū al-Fażl ibn Mubarak, ʿIyār-i Dānish, matn-i ʿilmī va 1985), no. 110 (color); Welch, 1960, ii, pl. A; Krishnadasa,
intiqādī bā taḥqīq va taṣḥīḥ-i Amrīzdān ʿAlīmardānuf; Anwar-e-Suhaili (Iyar-i-Danish), pl. A.
muḥarrirān-i masʾūl, Aṣghar Jānfidā va Jābilqā 16 Abū al-Fażl, ʿIyār-i Dānish, 2. De Blois claims that this
Dādʿalīāshāyif (Dushanbah: Dānish, 1988), 2. version of the introduction is found in Abū al-Maʿālī
14 Technically, the first painting in the manuscript is a Nasrullah’s Persian translation of the Kalila and Dimna.
garden scene that was added later, in the seventeenth François de Blois, Burzōy’s Voyage to India, 41.
The Patron and the Author 375
figure 13.1 Burzoy presented before King Anushirvan, by Basawan, Iyar-i Danish,
fol. 5, Lahore, 1596–7, Mughal India. Varanasi, Bharat Kala Bhavan,
Ms. 9069
376 Natif
the physician’s return, the king prepared a grand indicated the scenes to be painted.”19 Whether
reception and offered Burzūya whatever he would or not the Emperor himself specified which pas-
like from the royal treasury. To the surprise of the sages in each of these texts should be illustrated,
grandees at the court and of the king himself, the the sentence gives a sense of the importance given
physician chose only a robe of honor (khilaʿat), to the images produced in royal manuscripts. We
but asked that the story of his life, his journey to may thus allow for an in-depth reading of the
Hindustan, and how he had obtained the text be Iyār-i Dānish illustration.
added to the introduction of Kalīla wa-Dimna. The scene in Figure 13.1 entices the viewer to
The task of writing his biography and adventurous draw interesting parallels between Abū’l Fażl and
story was granted to Anūshirvān’s famous vizier, Burzūya, as well as between Khusraw Anūshirvān
Wuzurjmihr.17 Abū’l Fażl further states in his pref- and Emperor Akbar. It further alludes to con-
ace that the fables can be traced back to an Indian nections between the translations of texts from
source, thus indicating his awareness of history Sanskrit to Pahlavi at the Sassanian court, and the
and place. Christine van Ruymbeke argues that, by translation of Indian texts into Persian at Akbar’s
rewriting and repeating older introductions, the court. Lastly, it enables us to consider the allegori-
Mughal author emphasizes the “ultimate Indian cal transmittal of knowledge from one just ruler
origin of the text.”18 to another, and the relationship depicted between
In the Varanasi Iyār-i Dānish, the painting Abū’l Fażl and Emperor Akbar. An educated audi-
conveying this story is a full-page composition ence, like many officials among the Mughal elite,
(Fig. 13.1). The scene takes place at court, in a set- would have been expected to pick up on these
ting defined by all the attributes of grandeur and allusions to the past and make the connection
luxury, with rich draperies, carpets, gold vessels, with the present.20
an entourage of high officials, a leopard, a hawk,
and a horse with grooms. The ruler, at the top cen-
ter of the page, is seated on his majestic throne; he 19 Choudhri Mohammed Naim, “Appendix C”, in Tūtī-
gestures toward the physician (clad in a blue and nāma: Tales of a Parrot = das Papageienbuch:
gold robe) and a courtier, who dominate the mid- vollständige Faksimile-Ausgabe im Originalformat der
dle part of the painting. In the lower part, there Handschrift aus dem Besitz des Cleveland Museum of
Art = Complete Colour Facsimile Edition in Original Size
are musicians, servants, and animals with their
of the Manuscript in Possession of the Cleveland Museum
attendants. of Art (Graz: Akademische Druck- u. Verlagsanstalt,
The decision to represent this scene in the first 1976), vol. 2, 184.
illustration holds multiple layers of meaning and 20 For a similar level of intellectual engagement regard-
engagement for the author, artists, and audience. In ing Mughal painting, see Mika Natif, “The Generative
Garden: Sensuality, Male Intimacy, and Eternity in
the Āīn-i Akbarī, Abū’l Fażl mentions the intention Govardhan’s Illustration of Saʿdī’s Gulistān,” in Eros
behind the way in which scenes were selected for and Sexuality in Islamic Art, edited by Francesca Leoni
painting in the royal ateliers: “The Chingiz-niama, and Mika Natif (Farnham, Surrey; Burlington, VT:
the Zafar-nama, this book, the Razm-nama, the Ashgate Publishing, 2013), 54. Paul Losensky discusses
the Safavid and Mughal poetic practice of including
Ramayan, the Nal-Daman, the Kalila wa Dimna,
quotations from past Persian poets or imitations of
the ʿIyār-i Dānish, and other books have been illus- their style. See Paul E. Losensky, Welcoming Fighânî:
trated (paikar-nigari), His Majesty himself having Imitation and Poetic Individuality in the Safavid-Mughal
Ghazal (Costa Mesa, CA: Mazda Publishers, 1998), esp.
chap. 3. Gülru Necipoğlu identifies a similar creative
process in the way Timurid artists and poets introduce
17 Abū al-Fażl, ʿIyār-i Dānish, 6–7. innovative elements into their creations, while at the
18 Van Ruymbeke, “Authorship, Ownership and Rewrit- same time quoting past tradition. Gülru Necipoğlu,
ing,” 3. The Topkapi Scroll: Geometry and Ornament in Islamic
The Patron and the Author 377
figure 13.2
Detail: Burzoy presented
before King Anushirvan,
by Basawan, Iyar-i Danish,
fol. 5, Lahore, 1596–7,
Mughal India. Varanasi,
Bharat Kala Bhavan,
Ms. 9069
The painting is fascinating on numerous levels. Grand carpets are spread on the ground, and lav-
Through the use of portraiture, it links three dif- ish textiles decorate the side areas of the royal
ferent courts that existed in separate places and at pavilion. Moreover, the luxurious gold objects on
distinct times – that of the Mughals, of the Iranian the ground, the various hunting animals, the musi-
King Khusraw Anūshirvān, and of the Indian King cians, and the numerous persons clad in brightly
Dabshalīm – while suggesting a keen awareness colored robes, animatedly gesturing and talking
of the transmission and mediation of knowledge. to one another, all come together to exemplify the
Through visual means, the scene of the Sassanian exhilaration of the moment and the abundance of
ruler receiving his physician is immediately trans- the court. Nevertheless, the scene is disorientat-
lated into Mughal terms: every detail in the set- ing, with the only stable element being the figure
ting reflects the environment of the Mughal court. of the ruler seated on his golden throne, elevated
Architecture is rendered in the customary red and somewhat isolated from the crowd.
stone, with detailed carving and masonry work. The physician, approaching the royal seat, offers
an insight into Akbari court etiquette (Fig. 13.2). In
Architecture (Santa Monica, CA: Getty Center for the the Iyār-i Dānish text, it is said that when Burzūya
History of Art and the Humanities, 1995), 214–15, 218. appeared before King Anūshirvān he “kissed
378 Natif
the ground and gave thanks” (zamīn bus kardah turban, and has a thick dark beard, while playing
shukrana ʿināyathai Anūshirvān ba-jāāvordah).21 the role of the “intellectual ambassador between
In the illustration, however, the physician per- Iran and India”, as François De Blois calls him.26
forms a different gesture – the kornish salutation. The Jalayirid illustration conveys the key
When doing the kornish, according to the Āīn-i moment of the transmission of knowledge and
Akbarī, the person has to have “the palm of the its assimilation into a new scheme of patron-
right hand placed upon the forehead and the head age or ownership. However, what is missing in
to be bent downwards.”22 In the painting, Burzūya this scene, and also in the Iyār-i Dānish paint-
follows Mughal court etiquette very closely. Just ing, is an actual depiction of the book itself – the
below him, wearing a light purple robe, another “Object of Translation.” A more tangible render-
official is shown performing the taslīm, which ing can be seen in a copy of the Anvār-i Suhaylī
is described in the Āīn-i Akbarī as “placing the made for Prince Salīm, Akbar’s son and the future
back of the right hand on the ground, and then Emperor Jahangir (r. 1605–1626). In this illustra-
raising it gently till the person stands straight.”23 tion (Fig. 13.4), the ruler holds a closed book in his
These particular gestures can be further identified left hand, most likely the newly written stories,
in other Akbar-period illustrations when a man and with his right, gestures toward the author-
approaches the emperor. These salutations are translator Burzūya, who stands in supplication,
closely linked to Akbar’s rule since, according to bowing gently with his hands together.27 The
Abū’l Fażl, the emperor himself is credited with complex, almost chaotic painting in the Iyār-i
inventing them.24 Dānish – with numerous figures wearing magnifi-
Following the text more closely, in earlier illus- cent clothes, beautiful carpets, gold vessels, pea-
trations of the same scene, such as that in the cocks on the roof, and an elaborate architectural
Jalayirid manuscript of Kalila and Dimna, Burzūya setting – and the dramatic diagonal lines of the
is depicted performing the zamīn bus – kissing the composition stand in sharp contrast to the stifled,
ground – in accordance with the Persian text.25 focused, somewhat somber illustration from the
As he prostrates himself in front of the Sassanian Anvār-i Suhaylī (compare Figs. 13.1 and 13.4).
king, Burzūya is shown in these illustrations as The lack of a visible book in the Varanasi paint-
having the characteristics of a Muslim religious ing is reminiscent of the silsila of wisdom. The
scholar (ʿulamaʾ). He wears a green robe and a entire painting becomes an allegorical scene of
subliminal transference and exchange. The chain
21 Abū al-Fażl, ʿIyār-i Dānish, 7. of transmission is not solely one of texts and
22 Āīn-i Akbarī, vol. 1, 158–59. images, but also relates to creators/writers and
23 Ibid. their patrons. While Akbar associates himself with
24 Ibid.
25 See “Burzoy returns from India, presented before
past rulers such as Anūshirvān and Dabshalīm,
King Anushirvan,” a folio from Kalila and Dimna, by Abū’l Fażl links himself to the glorious line of the
Nasrullah Munshi, Tabriz, Jalayirid, c. 1370–74. The page past writers of the stories. Christine van Ruymbeke
is mounted in an album. Istanbul University Library F claims that by agreeing to rewrite a well-known
1422, F228. Published in Bernard O’Kane, Early Persian
text such as Kalīla wa-Dimna, Abū al-Fażl inserted
Painting: Kalila and Dimna Manuscripts of the Late
Fourteenth Century (London: I.B. Tauris Publishers; himself into a prestigious list of writers, while
Cairo: AUC Press, 2003), 98. A related painting is found
in a later copy of the Jalayirid manuscript, now housed 26 De Blois, Burzōy’ Voyage to India, 33.
at the Raza Library in Rampur, fol. 11a. Published in 27 In an earlier illustration from the Anvar-i Suhayli
Barbara Schmitz and Ziyaud-Din A. Desai, Mughal made around the 1520s, probably in Shaybanid Herat,
and Persian Paintings and Illustrated Manuscripts in Burzūya is depicted presenting an actual book to
the Raza Library, Rampur (New Delhi: Indira Gandhi Khusraw Anūshirvān. See Fig. 13.7 here. MS 9109 IOS AS
National Centre for the Arts, 2006), pl. 260. Uzbek SSR, f. 5b.
The Patron and the Author 379
figure 13.3
Detail: Burzoy presented before
King Anushirvan, by Basawan, Iyar-i
Danish, fol. 5, Lahore, 1596–7, Mughal
India. Varanasi, Bharat Kala Bhavan,
Ms. 9069
sacrificing, to a certain degree, his own originality; physician. The figure of the ruler-patron, Khusraw
this idea may explain why Abū’l Fażl repeats the Anūshirvān, is highlighted in the Iyār-i Dānish
list of past authors/translators/rewriters several illustration (Fig. 13.3). In accordance with the
times, and why it “resonates as an isnād.”28 Kalīla wa-Dimna tradition, the patrons sponsor-
The concept of a silsila of knowledge is ing translations of the text, and variations on it,
also echoed in the images of the king and his are important rulers or political figures who have
acquired fame in the Muslim world. Their pres-
28 Van Ruymbeke, Kashefi’s Anvar-e Sohayli, 18–22; Van ence in the Kalīla wa-Dimna preface, associated
Ruymbeke, “Authorship, Ownership and Rewriting,”
with the silsila of transmitters, contributes to the
5–6. In his preface to the Iyar-i Danish, Abū al-Fażl
mentions Dabshalim and Bidpai, as well as Khusraw esteem of the recensions they have supported.
Anūshirvān and Burzūya (1–2), and again references Abū’l Fażl provides a detailed list of pre-Islamic
Ibn Muqaffa and Abū al-Maʿālī Naṣrallāh (3). He states and Muslim patrons, to which he adds his own
that Maḥmūd Ghaznī (the Ghaznavid) ordered the benefactor, Emperor Akbar. The list includes
poet Rūdakī to compose a versified version of Kalila
and Dimna and that Nasrullah was commissioned by
the Hindu King Dabshalīm, the Sassanian ruler
Bahrāmshāh to translate Ibn Muqaffaʿ’s Arabic version Anūshirvān, the Abbasid Caliph Abū al-Jaʿfar
into Persian. Abū al-Fażl, ʿIyār-i Dānish, 8–9. Manṣūr, the Ghaznavid Sulṭān Maḥmud of Ghaznī
380 Natif
figure 13.4 King Anushirvan and Burzoy, fol. 6a, Anvar-i Suhayli, by Kashifi, Allahabad, 1604–10,
Mughal India. London, British Library, Add. MS 18579
The Patron and the Author 381
and his grandson, Bahrāmshāh b. Masʿūd, and successfully conveys the special rapport between
the Timurids Amīr Suhaylī and Sulṭān Ḥusayn the patron and the author, as they form a diago-
Bāyqarā.29 All these historical and mythical per- nal visual axis from the top down in the pictorial
sonalities allegorically coalesce in the image of space (Fig. 13.1).
the Mughal-looking ruler, representing Khusraw When Abū’l Fażl wrote the Iyār-i Dānish in the
Anūshirvān in the text and Emperor Akbar in late 1580s, the Mughal court was well engaged
the image. in cross-cultural activities, employing a number
The firm link between patron and author/ of literati who had been given the task of trans-
advisor can be seen in the hand gestures that lating various texts into Persian, the official lan-
each of the characters performs in this paint- guage of the court. With the support of Akbar, the
ing (Fig. 13.2). Anūshirvān, seated on an elevated Bāburnāma (History of Babur, on Akbar’s grandfa-
throne encrusted with jewels, extends his left hand ther) was translated from Chaghatai into Persian,
and, with an open palm, signals the physician. along with Sanskrit texts such as the Indian epics
The emperor’s motion is echoed and amplified Mahabharata and Ramayana:
by the courtier, who stretches forward both arms
and points toward Burzūya, whose hands create Philologists are constantly engaged in trans-
an ocular continuum, a diagonal linkage between lating Hindí, Greek, Arabic, and Persian
patron and writer. books into other languages. Thus a part of
The presence of the patron in the illustra- the Zích i Jadíd i Mírzáí was translated under
tion is further related to his role in initiating the the superintendence of Amír Fathullah
acquisition/translation/rewriting of the text. Van of Shíráz, and also the Kishnjóshí, the
Ruymbeke suggests that we should consider the Gangádhar, the Mohesh Mahánand, from
various translations and recensions of Kalīla wa- Hindí (Sanskrit) into Persian, according to
Dimna as “collaborative duos” in which ownership the interpretation of the author of this book
extends to both patron and author, though it is the [i.e. Abū’l Fażl]. The Mahábhárat, which
benefactor who is considered the physical owner belongs to the ancient books of Hindústán,
of the work.30 This concept was well understood has likewise been translated from Hindí into
at the Mughal court, as Fayzi, Akbar’s poet laure- Persian, under the superintendence of Naqíb
ate and Abū’l Fażl’s brother, writes: “In that breath, Khán, Mauláná ʿAbdul Qádir of Badáon, and
which is in the poet / the poetry is mine and the Shaikh Sultán of T’hanésar.32
ambition the shah’s.”31 The Mughal artist Basawan
Poets at the Great Islamic Courts, edited by Amy Landau
(Baltimore: Walters Art Museum; Seattle: University of
29 Abū al-Fażl, ʿIyār-i Dānish, 7–8. For a discussion of the Washington Press, 2015), 142.
different approaches to this matter by Kāshifī and Abū 32 Ain-i Akbari, vol. 1, 106. Also see ʿAbd al-Qādir Badāʿūnī,
al-Fażl, see Van Ruymbeke, “Authorship, Ownership Muntakhab al-Tawarikh of ʿAbd al-Qādir Badāʿūnī,
and Rewriting,” 8–9. trans. George S.A. Ranking, Sir Wolseley Haig, and
30 She further argues that the patron is considered to be W.H. Lowe (Calcutta: Asiatic Society of Bengal,
the true owner of the work, since he is the one who 1884–1925), vol. 2, 344. Texts from different languages,
is able to keep it, while the author/translator/rewriter including European languages, were translated into
follows the patron’s lead in terms of stylistic changes, Persian as collaborative projects. See Muzaffar Alam,
additions, and omissions to the text. Van Ruymbeke, “The Pursuit of Persian: Language in Mughal Politics,”
“Authorship, Ownership and Rewriting,” 9. Modern Asian Studies 32, no. 2 (1998): 326–27; Qamar
31 From Fayzi’s translation of the Mahabharat, ms. BL IO Adamjee and Audrey Truschke, “Reimagining the ‘Idol
Islamic 761, fol. 186b. In Qamar Adamjee and Audrey Temple of Hindustan’,” 141–65; Carl W. Ernst, “Muslim
Truschke, “Reimagining the ‘Idol Temple of Hindustan’: Studies of Hinduism? A Reconsideration of Arabic and
Textual and Visual Translation of Sanskrit Texts in Persian Translations from Indian Languages,” Iranian
Mughal India,” in Pearls on a String: Artists, Patrons, and Studies 36, no. 2 (2003): 173–95; Muzzafar Alam and
382 Natif
Abū’l Fażl appears to have supervised many of Indian origin of the Kalīla wa-Dimna text. In the
these projects, as well as making contributions of preface to the Iyār-i Dānish, he clearly states that
his own, such as the preface that he composed for the Brahman Hakim Bidpai composed the book of
the Persian Mahabharata.33 Similar to the Iyār-i Kalīla wa-Dimna or Karataka and Damanaka for
Dānish, he considers the Mahabharata as part King Dabshalīm of Hindustan.38 It is tempting to
of the history of ancient India, which would fur- think that Abū’l Fażl took delight in the notion of
ther enrich the education of the ruler.34 Like the bringing the Indian stories full circle, back to their
book of Kalīla wa-Dimna, which is compared to a place of origin. Echoing this notion, the painting
miraculous plant that can bring the dead back to of Burzūya before Anūshirvān is filled with refer-
life, Abū’l Fażl sees a knowledge of other traditions ences to the land of Hindustan.
(philosophy, science, and literature) as a form of The Mughal assimilation of the ancient text is
“rescue from the idiocies of fools pretending to be manifested through the visual engagement with
wise.”35 Justifying his own creative efforts, Abū’l India, its landscape and plants, and the represen-
Fażl emphasizes the antiquity of the text of Kalīla tation of animals, such as the two peacocks above
wa-Dimna and its recognition as a source of advice the king’s throne, the trained leopard at the bot-
by a long and prestigious line of kings. He writes: tom, and the red sandstone architecture (Fig. 13.1).
“Truly, this book is a monument left by the ancient Such visual markers enhance the notions of an
kings, in which one learns the fundamental princi-
ples of monarchical rule; it is a register by eminent not seem to have been popular and only one manu-
mortals on the universal rules of psychology and script is known of the Persian text. See Tara Chand
and Saiyed A.H. Abidi, “Panchakhyana, a Unique and
civic education.”36 Unknown Persian Translation of the Panchatantra,”
As he was involved with the rewriting of both Islamic Culture 39 (1965): 32–33, 35, 37. Van Ruymbeke
the Mahabharata and Kalīla wa-Dimna, Akbar’s remarks that it was only during Akbar’s time that
minister must have noticed some overlap between a systematic comparison of the Kalila wa-Dimna
with the Pañcatantra took place. Van Ruymbeke,
the stories. The Mughal vizier also may have been Kashefi’s Anvar-e Sohayli, 18, n. 44. The link between
aware that the Pañcatantra was the origin of the Pañcatantra and Kalila wa Dimna was noticed
Kalīla wa-Dimna, since the royal library seems to in the tenth century by the philosopher and scien-
have had a copy of the text in its original Sanskrit tist al-Bīrūnī. In his Chronology of Ancient Nations,
al-Bīrūnī comments on the literary ties between the
language.37 Indeed, Abū’l Fażl was aware of the
two texts and criticizes Ibn al-Muqaffaʿ for modifying
and adding to the original. Abū al-Fażl used al-Bīrūnī’s
Sanjay Subrahmanyam, “Love, Passion and Reason in treatise when he was writing about Indian philosophy
Faizi’s Nal-Daman,” in Love in South Asia: A Cultural and cosmology in the Ain-i Akbari. Eduard Sachau,
History, ed. Francesca Orsini (Cambridge: Cambridge Alberuni’s India (London, 1888; reprint, London, 1910),
University Press, 2006), 109–41. vol. 1, 159; Ernst, “Muslim Studies,” 177.
33 For a discussion and translation of Abū al-Fażl’s intro- 38 Abū al-Fażl, ʿIyār-i Dānish, 1. The fact that Abū al-Fażl
duction, see Ernst, “Muslim Studies,” 173–95. mentions the names Karataka and Damanaka in con-
34 Abū al-Fażl refers to the Iyar-i Danish as “Pand-nama”, a junction with Kalila and Dimna is important. These
book of counsel or advice. Abū al-Fażl, ʿIyār-i Dānish, 2. are the names of a pair of jackals, the main characters
35 From Abū al-Fażl’s introduction to the Mahabharata. of the first chapters in the book. Both the authors of
Translated by Ernst, “Muslim Studies,” 180–181. the Pahlavi and Arabic versions, Burzūya and Ibn
36 Translated by Christine van Ruymbeke, in Van al-Muqaffaʿ, name their respective translations after
Ruymbeke, Kashefi’s Anvar-e Sohayli, 132; Abū al-Fażl, the two beasts. By using the Sanskrit names of the
ʿIyār-i Dānish, 3. jackals (Karataka and Damanaka), in addition to their
37 It was only after the completion of the Iyar-i Danish names in the Arabic and the Persian translations of
that Emperor Akbar ordered Mustafa ibn Khaliqdad the text (Kalila and Dimna), Abū al-Fażl demonstrates
al-Hashimi al-Abbasi to translate a Jain version of the his knowledge and awareness of the original Sanskrit
text from Sanskrit into Persian. This text is known as composition, either the Pañcatantra or its Jain version,
the Pancakhyana. However, the Pancakhyana does Pancakhyana.
The Patron and the Author 383
affinity and continuum between a mythical past Suhaylī picture that was made for Akbar’s son
and a glorious present. Carl Ernst claims: “The (Fig. 13.4), we may recognize a portrait of Prince
translation of the Sanskrit epics was not an aca- Salīm inserted into the composition. Manifested
demic enterprise comparable to the modern study in these Mughal illustrations are notions of dynas-
of religion; it was instead part of an imperial effort tic continuation and legitimacy, as well as classic
to bring both Indic and Persianate culture into father and son competition, here between Akbar
the service of Akbar.”39 It is in this light that we and the future Emperor Jahangir.41
ought to consider the painting of Burzūya and A similarly celebratory scene involving the com-
Anūshirvān. pletion of a book at the Akbari court was painted
Examining closely the facial features of the by Dawlat, ca. 1600–1603, only six years later
ruler seated on the throne, an interesting paral- (Fig. 13.6). In this illustration, the author, Abū’l
lel emerges with representations of Akbar him- Fażl, holds open a copy of his Akbarnāma (History
self. Based on visual and textual sources, we may of Akbar) and humbly offers it to the emperor. As
assume that Akbar had a wheaten complexion; in the Iyār-i Dānish scene, Akbar’s court in the
black, slightly slanted eyes; arched eyebrows; a Akbarnāma illustration appears lush and exuber-
straight nose, and a signature dark mustache. “His ant, while the author remains modest and grate-
countenance was radiant, and he had the build of ful, looking up toward his benefactor. The Iyār-i
a lion, broad of chest with long hands and arms”, Dānish and the Akbarnāma compositions follow a
as Jahangir described him.40 This description also similar spatial organization, one that emphasizes
resonates with another portrait of the young Akbar the close relationship between the ruler-patron
from ca. 1580 (Fig. 13.5). It seems clear that the and the author-vizier. Once again, the financial
ruler in the Iyār-i Dānish is the Mughal Emperor and intellectual authorities of the book meet in a
in the guise of Khusraw Anūshirvān, embodying painting portraying Emperor Akbar and Abū’l Fażl.
the Sassanian philosopher-king, whose reign was Depictions of the book’s author and its patron in
considered the highpoint of the Sassanian Empire Kalīla wa-Dimna manuscripts are not uncommon.
(compare Figs. 13.3 and 13.5). Both Anūshirvān Burzūya and Anūshirvān are shown numerous
and Akbar were known for their love of philoso- times, as part of the illustration cycle of the frame
phy (Indian philosophy), science, literature, and story.42 Likewise, other authors/translators/rewrit-
history, and for holding intellectual discussions at ers who inscribed themselves into the narrative by
court. They both commissioned the translation of writing a new preface, such as Naṣrallāh, Kāshifī
texts from Sanskrit, Greek, and Syriac into Pahlavi and Abū’l Fażl, are also depicted from time to time.
or Persian. Moreover, by ordering Abū’l Fażl to In a late fourteenth-century Jalayirid manuscript
rewrite the text, Akbar aligned himself with the of Kalīla wa-Dimna, Naṣrallāh, who translated the
Iranian king, following in his wise and prestigious text from Arabic into Persian, appears in an audi-
footsteps. By injecting Akbar’s representation into ence scene before his patron, Bahrāmshāh.43 The
the picture, the text and the illustration of the
Iyār-i Dānish place new material within familiar
41 For comparative portraits of Jahangir and their analysis,
frameworks of knowledge, science, and history.
see Elaine Wright, Muraqqaʿ: Imperial Mughal Albums
Similarly, in the figure of Anūshirvān in the Anvār-i from the Chester Beatty Library, Dublin (Alexandria, VA:
Art Services International; Hanover: Distributed by
39 Ernst, “Muslim Studies,” 182. University Press of New England, 2008), 168–69.
40 Nūr al-Dīn Muḥammad Jahāngīr, The Jahangirnama: 42 At least eight manuscripts depict this scene, according
Memoirs of Jahangir, Emperor of India, trans. and ed. to Grube, “Prolegomena for a Corpus Publication,” 406.
Wheeler M. Thackston (Washington, DC; New York: 43 Istanbul 1422, fol. 28a. At least four manuscripts have
Freer Gallery of Art, Arthur M. Sackler Gallery; Oxford illustrations of this subject. See Grube, “Prolegomena
University Press, 1999), 36. for a Corpus Publication,” 405.
384 Natif
figure 13.5 Akbar and a dervish, ca. 1580–90, inscribed to Abd al-Samad, Mughal India. Toronto, Aga Khan
Museum Collection, AKM14
The Patron and the Author 385
figure 13.6 Abu’l Fazl Presents Akbar with the second volume of the Akbar Nama,
by Govardhan, fol. 176v, ca. 1603–5, Mughal India. (Right hand side of a
double-page illustration) Dublin, Chester Beatty Library, ms. 3
386 Natif
Ghaznavid ruler is seated on a lavish throne, sur- he offers his book to the ruler, who elegantly ges-
rounded by richly patterned walls, curtains, and tures toward him with his hand. This image invites
carpets. In this lush setting, a few courtiers have the viewer to further meld Anūshirvān with the
gathered around the ruler; among them, a rather Timurid rulers, and by extension, to see in Burzūya
old official, adorned with white hair and a beard, a manifestation of Kāshifī. Eleanor Sims argues
sits next to Bahrāmshāh’s throne. Seated to the left, that the Tashkent Anvār-i Suhaylī manuscript
slightly separated from others, is the translator, was commissioned by Amīr Suhaylī himself, the
Naṣrallāh. Dressed in the fashion of the Ghaznavid patron of the text.47 The link between the manu-
court, with a short sleeve overcoat, a long tunic, a script and the court circle of the Timurid Sulṭān
belt, boots, and a turban, the author elegantly ges- Ḥusayn Mirzā is based upon the identification of
tures toward his patron, who gazes in his direction. the portraits in the illustration on fol. 5b.48 The
A similar conflation between past and present intentional blending of past and present patrons
authors and patrons can be seen in a sixteenth- and authors strengthens the notion of transfer-
century Anvār-i Suhaylī manuscript which has a ence and continuity among the different courts
painting depicting a ruler receiving a book from a linked to the Kalīla wa-Dimna tradition.
courtier (Fig. 13.7).44 It is worth noting that in the The first illustration in the Lahore Iyār-i Dānish
catalog of the Uzbek Academy of Sciences, which also consciously articulates the conflation of
houses the manuscript, the illustration is identi- Anūshirvān and Burzūya with Akbar and Abū’l
fied as Kāshifī presenting the Anvār-i Suhaylī to his Fażl through the use of portraiture (Fig. 13.1). In a
patron, Amīr Suhaylī.45 While the text surround- broader context, books symbolize the mobility and
ing the image narrates the story of Burzūya and longevity of knowledge. The addition of images of
Anūshirvān, the illustration resembles a simplified authors and patrons, as part of the manuscript
version of a Timurid reception scene, one taking body, “legitimizes” the process of transference and
place in a blooming garden. Anūshirvān’s image, connects the figures depicted to a prestigious sil-
wearing a green overcoat, a Central Asian crown, sila of men of learning and power.
short beard, and boots, brings to mind represen- Interestingly, the Iyār-i Dānish makes no men-
tations of Timur and his descendants.46 Burzūya, tion of a patron in its colophon or elsewhere in the
with a plain brown robe, a heavy beard, and a white book. In contemporary historical sources, Abū’l
turban, is rendered in the guise of a Timurid reli- Fażl specifically mentions that among the illus-
gious scholar. Kneeling at the foot of the throne, trated manuscripts in Akbar’s kitābkhāna were the
book of Kalila and Dimna and the Iyār-i Dānish – a
possible reference to this codex. Considering the
44 The manuscript was probably produced in Bukhara fact that Akbar’s portrait appears in the first illus-
or Herat in the first half of the sixteenth century. It is
currently in Tashkent, Uzbek Academy of Sciences,
tration, and the calligrapher’s in the last, along
Catalogue of Oriental Manuscripts 9109, cat. no. with the depiction of Mughal court etiquette
5767, fol. 5b. Aleksandr A. Semenov, ed., Sobranie and the choice of the new rendition of the text,
Vostochnykh Rukopisei Akademii nauk Uzbekskoi SSR I would like to propose that this manuscript should
(Tashkent: Izd-vo Akademii nauk UzSSR, 1952), vol. 8,
be seen as part of the Mughal effort to construct
189–91, no. 5767.
45 Galina Pugachenkova and Olimpiada Galerkina, an intellectual identity in a pluralistic society,
Miniatiury Sredneĭ Azii: v izbrannykh obraztsakh: iz one shaped by the Mughal view of transcultural
sovetskikh i zarubezhnykh sobraniĭ (Moskva: Izobra
zitel′noe iskusstvo, 1979), 90.
46 For comparison, see Timur’s image in the Zafarnama 47 Eleanor Sims, “16th-Century Persian and Turkish
(Book of Conquest), made in Herat, 1467–1480. Manuscripts of Animal Fables in Persia, Transoxiana,
Baltimore, The Johns Hopkins University, Garrett and Ottoman Turkey,” in A Mirror for Princes, 101.
Zafarnama. 48 Semenov, Sobranie Vostochnykh Rukopisei, vol. 8, 191.
The Patron and the Author 387
figure 13.7 Borzoy before Anushirwan, Anvar-i Suhayli, by Kashifi, first half of the sixteenth
century, Central Asia. Tashkent, Uzbek Academy of Sciences, Catalogue of
Oriental Manuscripts 9109
388 Natif
knowledge. More specifically, the first illustration Alam, Muzaffar. “The Pursuit of Persian: Language
(Fig. 13.1) casts Akbar and Abū’l Fażl in the light in Mughal Politics.” Modern Asian Studies 32, no. 2
of Khusraw Anūshirvān and Burzūya, linking the (1998): 317–349.
Mughal court to the just rule of enlightened past Brockelmann, Carl. “Kalila wa-Dimna.” In Encyclopaedia
kings, while reinforcing Akbar’s legitimacy as the of Islam, second edition. vol. 4, pp. 503–506. Leiden:
true ruler of all of India. Brill.
Chand, Tara and Sayied A.H. Abidi. “Panchakhyana, a
Unique and Unknown Persian Translation of the
Bibliography Panchatantra.” Islamic Culture 39 (1965): 29–39.
De Blois, François. Burzōy’s Voyage to India and the
Primary Sources Origin of the Book of Kalīlah wa Dimnah. London:
ʿAbd al-Qādir Badāʿūnī. Muntakhab al-Tawarikh of Royal Asiatic Society, 1990.
ʿAbd al-Qādir Badāʿūnī. Translated by George S.A. Ernst, Carl W. “Muslim Studies of Hinduism? A Recon-
Ranking, Sir Wolseley Haig, and W.H. Lowe. 3 vols. sideration of Arabic and Persian Translations from
Calcutta: Asiatic Society of Bengal, 1884–1925. Indian Languages.” Iranian Studies 36, no. 2 (2003):
Abū al-Fażl ibn Mubarak (ʿAlami). Ain-i Akbari. 173–195.
Translated by H. Blochmann. Calcutta: Printed for Ethé, Hermann. Catalogue of Persian Manuscripts in the
the Asiatic Society of Bengal, 1927–1949. Library of the India Office. 2 vols. Oxford: Printed for
Abū al-Fażl ibn Mubarak (ʿAlami). Akbar-nama (History the India Office by H. Hart, 1903.
of the reign of Akbar including an account of his Grube, Ernst J. “Prolegomena for a Corpus Publication
predecessors). Translated by H. Beveridge. Lahore: of Illustrated Kalilah wa Dimnah Manuscripts.”
Book Traders, Al-Abbass International, 1977. Islamic Art 4 (1991): 301–481.
Abū al-Fażl ibn Mubarak (ʿAlami). ʿIyār-i Dānish, matn- Khandalavala, Karl and Kalpana Desai. “Indian Illus-
i ʿilmī va intiqādī bā taḥqīq va taṣḥīḥ-i Amrīzdān trated Manuscripts of the Kalilah wa Dimnah,
ʿAlīmardānuf; muḥarrirān-i masʾūl, Aṣghar Jānfidā va Anvar-i Suhayli, and Iyār-i Dānish.” In A Mirror for
Jābilqā Dādʿalīāshāyif. Dushanbah: Dānish, 1988. Princes from India: Illustrated Versions of the Kali-
ʿAlī Akbar Dihkhudā. Lughatnāmah, Tālīf-i ʿAlī Akbar lah wa Dimnah, Anvar-i Suhayli, Iyar-i Danish, and
Dihkhudā. Tehran: Sāzmān-i Lughat-nāmah, 1337– Humayun Nameh, edited by Ernst J. Grube, 128–144.
1352 (1959–1974). Mumbai: Marg Publications, 1991.
Nūr al-Dīn Muḥammad Jahāngīr. The Jahangirnama: Krishnadasa, Ray. Anwar-e-Suhaili (Iyar-i-Danish).
Memoirs of Jahangir, Emperor of India. Translated Varanasi: Bharat Kala Bhavan, Banaras Hindu
and edited by Wheeler M. Thackston. Washing University, reprinted in 1999.
ton, DC; New York: Freer Gallery of Art, Arthur Leach, Linda York. Mughal and Other Indian Paintings
M. Sackler Gallery; Oxford University Press, 1999. from the Chester Beatty Library. 2 vols. London:
Scorpion Cavendish, 1995.
Secondary Literature Lefèvre, Corinne. “In the Name of the Fathers: Mughal
Adamjee, Qamar and Audrey Truschke. “Reimagining Genealogical Strategies from Bābur to Shāh Jahān.”
the ‘Idol Temple of Hindustan’: Textual and Visual Religions of South Asia 5, no. 1–2 (2011): 409–442.
Translation of Sanskrit Texts in Mughal India.” In Losensky, Paul E. Welcoming Fighânî: Imitation and
Pearls on a String: Artists, Patrons, and Poets at the Poetic Individuality in the Safavid-Mughal Ghazal.
Great Islamic Courts, edited by Amy Landau, 141–165. Costa Mesa, CA: Mazda Publishers, 1998.
Seattle: University of Washington Press, 2015. Naim, Choudhri Mohammed. “Appendix C.” In Tūtī-
Alam, Muzaffar and Sanjay Subrahmanyam. “Love, nāma: Tales of a Parrot = das Papageienbuch:
Passion and Reason in Faizi’s Nal-Daman.” In Love in vollständige Faksimile-Ausgabe im Originalformat
South Asia: A Cultural History, edited by Francesca der Handschrift aus dem Besitz des Cleveland
Orsini, 109–141. Cambridge University Press, 2006. Museum of Art = Complete Colour Facsimile Edition
The Patron and the Author 389
in Original Size of the Manuscript in Possession of the Ruymbeke, Christine van. Kashefi’s Anvar-e Sohayli:
Cleveland Museum of Art. Graz: Akademische Druck- Rewriting Kalila and Dimna in Timurid Herat. Leiden:
u. Verlagsanstalt, 1976. Brill, 2016.
Natif, Mika. “The Generative Garden: Sensuality, Male Sachau, Eduard, ed. and trans. Alberuni’s India. London,
Intimacy, and Eternity in Govardhan’s Illustration 1888; reprint, London: Kegan Paul, Trench, Trübner
of Saʿdī’s Gulistān.” In Eros and Sexuality in Islamic & Co., 1910.
Art, edited by Francesca Leoni and Mika Natif, Schmitz, Barbara and Ziyaud-Din A. Desai. Mughal and
43–64. Farnham, Surrey; Burlington, VT: Ashgate Persian Paintings and Illustrated Manuscripts in the
Publishing, 2013. Raza Library, Rampur. New Delhi: Indira Gandhi
Necipoğlu, Gülru. The Topkapi Scroll: Geometry and National Centre for the Arts, 2006.
Ornament in Islamic Architecture. Santa Monica, Semenov, Aleksandr A., ed. Sobranie Vostochnykh
CA: Getty Center for the History of Art and the Rukopisei Akademii nauk Uzbekskoi SSR. Tashkent:
Humanities, 1995. Izd-vo Akademii nauk UzSSR, 1952.
O’Kane, Bernard. Early Persian Painting: Kalila and Sims, Eleanor. “16th-Century Persian and Turkish
Dimna Manuscripts of the Late Fourteenth Century. Manuscripts of Animal Fables in Persia, Transoxiana,
London: I.B. Tauris Publishers; Cairo: AUC Press, and Ottoman Turkey.” In A Mirror for Princes from
2003. India: Illustrated Versions of the Kalilah wa Dimnah,
Pgachenkova, Galina and Olimpiada Galerkina. Min- Anvar-i Suhayli, Iyar-i Danish, and Humayun nameh,
iatiury Sredneĭ Azii: v izbrannykh obraztsakh: edited by Ernst J. Grube, 98–123. Mumbai: Marg
iz sovetskikh i zarubezhnykh sobraniĭ. Moscow: Publications, 1991.
Izobrazitel′noe iskusstvo, 1979. Verma, Som Prakash. “Anvar-i Suhayli – an Illustrated
Rieu, Charles. Catalogue of the Persian Manuscripts Manuscript of Akbar’s Court in the Collection of
in the British Museum. London: British Museum, Bharat Kala Bhavan, Varanasi.” Roopa-Lekha 44, no.
1879–83. 1–2 (1977): 17–22.
Rizvi, Saiyid Athar Abbas. Religious and Intellectual Wade, Bonnie C. Imaging sound: an ethnomusicologi-
History of the Muslims in Akbar’s Reign, with Special cal study of music, art, and culture in Mughal India.
Reference to Abu’l Fazl (1556–1605). New Delhi: Chicago: University of Chicago Press, 1998.
Munshiram Manoharlal Publishers, 1975. Wright, Elaine. Muraqqaʿ: Imperial Mughal Albums from
Ruymbeke, Christine van. “Authorship, Ownership the Chester Beatty Library, Dublin. Alexandria, VA:
and Rewriting: Vāʿiz Kāshifī and Abū’l-Fazl b. Art Services International; Hanover: Distributed by
Mobārak within the Noble Hereditary Line of Kalīla University Press of New England, 2008.
wa-Dimna Authors.” Jerusalem Studies in Arabic and
Islam 45 (2018): 181–210.
Catalogue des manuscrits illustrés de Kalīla wa-Dimna et textes
apparentés / Catalog of Illustrated Kalīla wa-Dimna Manuscripts
and Related Texts
Paris Bibliothèque Arabe 3465 Égypte ou c.1220 98 (dont De Slane 1883, 603 ;
nationale de Syrie 8 ajoutées Buchtal 1943 ; de
France postérieure- Blois 1990, P13, 70 ;
ment) Grube 1990-91, n° 1,
374 ; O’Kane 2003,
n° 1, 38-39, 218 ;
https://archiveset-
manuscrits.bnf.fr/
ark:/12148/cc31361k
Riyad King Faysal 2536 Égypte ou 26 ṣafar Muḥammad 65 (posté- Bin Junayd 2007 ;
Center for Syrie 747H/ 18 juin b. ʿAlī b. rieures ?) O’Kane (sous
Research and 1346 Sālim b. presse)
Islamic Studies Aḥmad
for Research and al-Ḥanafī
Islamic Studies
Paris Bibliothèque Arabe 3466 Égypte Avant 103 espaces De Slane 1883, 603 ;
nationale de ou Syrie 854H/1450a1 libres de Blois 1990, P1, 67 ;
France https://archiveset-
manuscrits.bnf.fr/
ark:/12148/cc31362t
Cambridge University T-S Ar.51.60 Égypte Début xive ? 1 fol. Walzer 1957 ; Grube
Library ou Syrie 1990-91, n° 6, 376 ;
O’Kane 2003, n° 3,
39-40, 221
Paris Bibliothèque Arabe 3467 Égypte xive 50 (dont De Slane 1883, 603 ;
nationale de ou Syrie 3 ajoutées Walzer 1959 ; Hal-
France postérieure- dane 1978, 95-96 ;
ment) de Blois 1990, P14,
70 ; Grube 1990-91,
n° 3, 375 ; O’Kane
2003, n° 5, 39, 222 ;
https://archiveset-
manuscrits.bnf.fr/
ark:/12148/cc313632
Londres British Library Add.24350 Égypte ? xive 90 espaces de Blois 1990, L6,
libres 69 ; Grube 1990-91,
n° 5A, 376 ; O’Kane
2003, n° 7, 39,
222-224
Cambridge University T-S Ar.40.9 Égypte ou Fin xive ? 1 fol. O’Kane 2003, n° 9,
Library Syrie 40-41, 225
Manchester The John MS Ar. 487 Égypte ? 1 shaʿbān 113 Mingana 1934,
Rylands 1083H/ 816-817 ; Grube
Library 22 novembre 1990-91, n° 79, 400 ;
1672 O’Kane 2003, 295
Paris Bibliothèque Arabe Égypte ? 13 rabīʿ I ʿAbd al-Rawf 180 De Slane 1883, 604 ;
nationale de 3475 1175H/12 al-Sanjalafī de Blois 1990, P10,
France octobre 1761 69 ; Grube 1990-91,
n° 82, 400 ; Barru-
cand 1991 ; https://
archiveset
manuscrits.bnf.fr/
ark:/12148/cc31371s
Londres British Library Or. 4044 Syrie ? xviie ? 76 Rieu 1894, n° 1156,
(ff.1-135) 731-732 ; de Blois
1990, L5, 69 ; Grube
1990-91, n° 75, 399
Paris Bibliothèque Arabe 3469 Égypte ? xviie ? 67 espaces De Slane 1883, 603 ;
nationale de libres de Blois 1990, P15,
France 71 ; https://archive
setmanuscrits.
bnf.fr/ark:/12148/
cc31365j
396 Catalogue des manuscrits illustrés de Kalīla wa-Dimna et textes apparentés
Paris Bibliothèque Arabe 3478 Égypte ? xviiie ? 109 espaces De Slane 1883, 604 ;
nationale de libres de Blois 1990, P18,
France 72 ; https://archive
setmanuscrits.
bnf.fr/ark:/12148/
cc31374h
b Authenticité douteuse.
Paris Bibliothèque Arabe 3511 Syrie ? xviie ? 58 peintures De Slane 1883, 609
nationale de (34 espaces
France libres)
Anthologie poétique
Collection Damas ? 1537 Badr al-Dīn 130 Kraus 1981, 83 ; Grube 1990-
privée Muḥammad 91, n° 40, 391
b. al-Ghazzī
al-ʿĀmirī
Berlin Museum für I. 6940 Damas ? 1537 Badr al-Dīn 1 fol. Aslanapa 1971, 313 ;
Islamische Muḥammad Rührdanz 1984, n° 39,
Kunst b. al-Ghazzī 122-124
al-ʿĀmirī
Légende
En gris foncé, manuscrits dont les figures n’ont pas été exécutées.
Atıl, Esin. Kalila wa Dimna: Fables from a Fourteenth- Kraus, Hans P. Illuminated Manuscripts from the
Century Arabic Manuscript. Washington, D.C. : Eleventh to the Eighteenth Centuries. New York : H.P.
Smithsonian Institution Press, 1981. Kraus, 1981.
Aumer, Joseph. Die arabischen Handschriften der K. Krek, Miroslav. A Catalogue of Arabic Manuscripts in the
Hof-und Staatsbibliothek in München. Munich : Palm, Oriental Institute of Chicago. New Haven : American
1866. Oriental Society, 1961.
Barrucand, Marianne. « Le Kalila wa Dimna de la Mac Guckin de Slane, William. Catalogue des manuscrits
Bibliothèque royale de Rabat : un manuscrit illustré arabes / Bibliothèque nationale. Paris : Imprimerie
il-khanide ». Revue des études islamiques, numéro nationale, 1883-1895.
spécial en l’honneur de Dominique Sourdel 54 (1986) : Melikian-Chirvani, Assadulah S. Sulwān al-muṭāʿ fī
17-51. ʿudwān al-atbāʿ: A Rediscovered Masterpiece of Arab
Barrucand, Marianne. « Un manuscrit arabe illustré Literature and Painting. Kuwait : Tri Publishing, 1985.
de Kalila wa Dimna du XIIIe siècle et sa copie otto- Mingana, Alphonse. Catalogue of the Arabic manus-
mane ». Archéologie islamique, no 2 (1991) : 81-95. cripts in the John Rylands Library, Manchester.
Bin Junaid, Yahya. « Kalila wa Dimna ». Ḥadīth al-Dār / Manchester : Manchester University Press, 1934.
Hadeeth ad-Dar: the Journal of Dar al-Athar al-Isla- O’Kane, Bernard. Early Persian Painting: Kalila and
miyyah 24 (2007) : 32-36. Dimna Manuscripts of the Late Fourteenth Century.
Blois, François de. Burzōy’s Voyage to India and the Londres : I.B. Tauris, 2003.
Origin of the Book of Kalīlah wa Dimnah. Londres : O’Kane, Bernard. « Mamluk Illustrated Kalila and
Royal Asiatic Society, 1990. Dimnas and the Riyadh Manuscript ». In Arts and
Bothmer, Hans-Caspar. Kalila und Dimna: Ibn al- Culture of the Book in the Islamic World. Edité par
Muqaffaʿ’s Fabelbuch in einer mittelalterlichen Bilder- Serpil Baǧcı et Aslıhan Erkmen. Sous presse.
handschrift: Cod. arab. 616 der Bayerischen München, Rieu, Charles. Supplement to the Catalogue of the
Staatsbibliothek. Wiesbaden : Reichert, 1981. Arabic Manuscripts in the British Museum. Londres :
Buchthal, Hugo. « “Hellenistic” Miniatures in Early Longmans, 1894.
Islamic Manuscripts ». Ars Islamica 7, n° 2 (1940) : Rührdanz, Karin, éd. Orientalische illustrierte
125-150. Handschriften: aus Museen und Bibliotheken der
Derenbourg, Hartwig. Les manuscrits arabes de l’Escu- Deutschen Demokratischen Republik Ausstellung
rial. Vol. 1. Paris : E. Leroux, 1884. im Museum, Islamischen der Staatlichen Museen
Ettinghausen, Richard. La peinture arabe. Genève : zu Berlin, Hauptstadt der DDR. Berlin : Staatliche
Skira, 1962. Museen zu Berlin, 1984.
Gacek, Adam. Arabic Manuscripts in the Libraries of Río Sánchez, Francisco del. Catalogue des manuscrits de
McGill University. Union catalogue. Montréal : McGill la fondation Georges et Mathilde Salem (Alep, Syrie).
University Libraries, 1991. Wiesbaden : Reichert, 2008.
Grube, Ernst J. « Prolegomena for a Corpus Publication Sbath, Paul. Bibliothèque de manuscrits Paul Sbath,
of Illustrated Kalilah wa Dimna Manuscripts ». prêtre syrien d’Alep. Vol. 3. Le Caire : H. Friedrich et
Islamic Art 4 (1990-91) : 301-481. Co., 1934.
Grube, Ernst J. « Some Observations Concerning the Walzer, Sofie. « An Illustrated Leaf from a Lost Mamluk
Ottoman Illustrated Manuscripts of the Kalilah wa Kalilah wa Dimnah Manuscript ». Ars Orientalis 2
Dimnah ». In Milletlerarası Türk sanatları kongresi, (1957) : 503-505.
II : 195-205. Ankara : Kültür Bakanlığı, 1995. Walzer, Sofie. « The Mamlûk Illuminated Manuscript of
Haldane, Duncan. Mamluk Painting. Warminster : Aris Kalîla wa-Dimna ». In Aus der Welt der islamischen
& Phillips, 1978. Kunst: Festschrift für Ernst Kühnel zum 75 Geburstag
Islamic Art and Manuscripts: sale LILAC-6713 : auction, am 26-10-1957. Édité par Richard Ettinghausen, 195-
Christie’s, Tuesday 29 April 2003. Londres : Christie’s, 206. Berlin: Gebr. Mann, 1959.
2003.
Catalogue des manuscrits illustrés de Kalīla wa-Dimna et textes apparentés 399
En gras la cote uniformisée, à côté entre parenthèses les formes non uniformisées
Kalīla wa-Dimna, Ibn al-Muqaffaʿ Arabe 3467 (A3467) 60–89, 109, 118, 164n80,
Berlin, Staatsbibliothek zu Berlin, Wetzstein II 218–220t, 237, 244f, 253, 269, 271–277, 281–283, 285f,
672 6n16, 7, 13t, 23, 40n76 289–291, 338nn.(46, 48), 339, 340nn.(60, 62)
Le Caire, al-Maktabat l-Markaziyya li-l-Makhṭūṭāt al- Arabe 3470 (A3470) 142–145, 152f, 154–165, 168–170,
Islamiyya, 1169 6n16, 33n72 270, 272–276, 281–282, 285f, 291, 293–294
Cambridge, Cambridge University Library, T-S Ar.51.60 Arabe 3472 (A3472) 269, 271–276, 280, 282–283, 285,
(TS-AR-051-060) 218–220t, 242, 340n60 291–293, 338n48
Cambridge, Corpus Christi College, Parker Library, MS Arabe 3473 (P3473) 6n16, 7, 10–21, 23–24, 29
Ar.578 (MS 578) 117n70, 118, 120f, 136n35, 164n80, Arabe 3475 (A3475) 5n12, 6n16, 13t, 172–188, 192–193,
218–220t, 340–342 195–201, 270–276, 280–282, 285, 291–293, 338nn.
Chicago, Oriental Institute, A 11991 5n12 (46, 48)
Collection privée, Vente Christie’s, avril 2003, vente 6713, Arabe 3476 32n65
lot 93 169 Arabe 3478 7, 29
Istanbul, Süleymaniye Kütüphanesi Arabe 3593 6n14
Ayasofia 4095 (A4095) 6, 10–17, 19n33, 20t, 21, Arabe 5881 (P5881/A5881) 6n16, 7, 9n30, 10–19,
23–25 21, 24–27, 29–33, 30n59, 33n72, 35–36f, 269–276,
Ayasofia 4213 (A4213) 5n12, 17n32 280–282, 285, 291–293
Ayasofia 4214 (A4214) 6n16, 30n59, 33n72 Rabat, Bibliothèque nationale du Royaume du Maroc,
Istanbul, Topkapı Sarayı Müzesi Kütüphanesi, Ahmet III Ms. 3655 (BRR3665/Rabat 3655/BRR 3655) 5n12,
3015 (Sultan Ahmet III 3015) 34n74 6n16, 32n67, 34n75, 172–200, 218–220t, 270n16, 301–323,
Londres, British Library 341n67
Add. 23466 32n65 Riyad, King Faysal Center for Research and Islamic
Or. 3900 (L3900) 34n74 Studies for Research and Islamic Studies
Or. 4044 (L4044) 7, 9–24, 27–29, 32–33, 39f, 40n76 2407 6n16, 32n65
Or. 8571 (L8571) 7, 13–21, 23–24, 29, 32n65 2536 (King Fayṣal 2536) 13t, 17n32, 34n74, 164n80,
Manchester, The John Rylands Library, MS Ar. 487 (Arabic 218–220t
MSS 487 [537]) 139n, 147f, 154n72 Tunis, Bibliothèque Nationale de Tunisie, 2281 6n16
Montreal, McGill University Library, Arabic 94 Khamsa, Niẓāmī
(117953) 6n16 Istanbul, Topkapı Sarayı Müzesi Kütüphanesi, Hazine 761
Munich, Bayerische Staatsbibliothek (H. 761) 172n2
Cod.arab.615 (Cod.Ar.615) 6n16, 169 Kırk vezir hikāyesi
Cod.arab.616 (Cod.Ar.616/ Cod. Arab.616/ Cod Arab Istanbul, Istanbul University Library, Ms T.7415 359n13
616) 17n32, 74n66, 97, 99, 102–104f, 109–113, 118, Kitāb al-diryāq, Pseudo-Galien
218–220t, 237, 245f, 253, 260–262, 264–265, 340n62, Paris, BnF, Arabe 2964 106
341n67, 344n72 Kitāb al-ḥashāʾish fī hayūlā ʿilāj al-ṭibb, Dioscoride
New Delhi, National Museum, Akhlaq 508 (dépôt de Bologne, Biblioteca universitaria, Cod. Arab. 2954 106,
l’Asafiya Library d’Hyderabad) 64 107f
New York, Metropolitan Museum of Art, 1981.373.51/ Istanbul, Süleymaniye Kütüphanesi, Ayasofia 3703 105
MMA, 1981.373.51 13t, 233–243, 247–250, 253, Kitāb al-ḥayawān, al-Jāḥiẓ
255–261, 263–266 Milan, Ambrosiana Library, Ar. 140 Inf. 118, 124
Oxford, Bodleian Library Kitāb al-ḥiyal, Banū Mūsā
E.D. Clarke.Or. 09 (Or. 9) 139n, 143, 146–150, 153–154, Doha, Museum of Islamic Art, Ms.683.2001
156–162t, 164–165, 167t, 169 (MS.683.2001) 209n20
Pococke 400 17n32, 19n32, 30n60, 34nn.(74–75), 99, Kitāb al-raml
109, 111–114, 118–119, 123n92, 164n80, 218–220t, 242, Paris, BnF, Arabe 2731 65
246–254, 258, 260–261, 264–265, 340n62, 341n67, Kitāb-i fālnāma, Jaʿfar al-Ṣādiq
344n72 Genève, Musée d’Art et d’Histoire, Pozzi Collection,
Paris, BnF 1971-107/121 64
Arabe 2789 32n65 Kitāb manāfiʿ al-ḥayāwān, Ibn Bakhtīshūʿ
Arabe 3465 (P3465/A3465) 6–7, 10–24, 34n75, 97–98, Paris, BnF, Arabe 2782 118–119
106–109, 111–118, 130–145, 150–151, 154–170, 218–220t, San Lorenzo del Escorial, Biblioteca Real, Ar. 898 114,
268–283, 285f, 287–289, 291–296, 338nn.(46, 48), 119, 121–122, 124, 220n25, 225, 228, 230n40
340nn.(61–62), 341nn.(64, 67), 342n68, 344 Kitāb naʿt al-ḥayawān, Ibn Bakhtīshū
Arabe 3466 (P3466) 6n16, 7, 9–21, 23–24, 26–29, Londres, British Library, Or. 2784 105–107, 114, 115, 221n25
31–33, 37–38f
402 Index des manuscrits / Index of Manuscripts
Par ville et institution / By City and Institution Dublin, Chester Beatty Library
Ms. 3, Abū’l Faẓl, Akbarnāma 383, 385f
Berlin, Museum für islamischen Kunst Ms. 1479, Coran 119n83
I.6940 (même manuscrit que collection privée), feuillet
isolé 331 Genève, Musée d’Art et d’Histoire, Pozzi Collection
Berlin, Staatsbibliothek zu Berlin 1971-107/121, Jaʿfar al-Ṣādiq, Kitāb-i fālnāma 64
Ms. or. fol. 2195 (Ms Or Fol 2195), Sīrat al-Iskandar
Dhū-l-Qarnayn 164n80 La Haye, Bibliothèque nationale des Pays-Bas
Ms. or. oct. 2507 (ms. or. Oct. 2507), Tarjuma-i Kalīla va- M10 B23, Bible 228n33
Dimna [traduction de Kāshifī, Anvār-i Suhaylī] 366
Wetzstein II 672, Ibn al-Muqaffaʿ, Kalīla wa- Istanbul, Istanbul University Library
Dimna 6n16, 7, 13t, 23, 40n76 F.1422, Naṣrallāh Munshī, Kalīla va-Dimna 342, 344
Beyrouth, Bibliothèque de l’Université Saint Joseph Mss 5964, Matrakçı Nasuh, Beyān-i menāzil-i sefer-i
00022, al-Tanūkhī, al-Faraj baʿda l-shidda 6n16, 32n65, Irākeyn-i Sultān Süleymān Hān 346–347
34n73 Ms T.7415, Kırk vezir Hikāyesi 359n13
Bologne, Biblioteca universitaria Istanbul, Süleymaniye Kütüphanesi
Cod. Arab. 2954, Dioscoride, Kitāb al-ḥashāʾish fī hayūlā Ayasofia 3703, Dioscoride, Kitāb al-ḥashāʾish fī hayūlā ʿilāj
ʿilāj al-ṭibb 106, 107f al-ṭibb 105
Ayasofia 4095 (A4095), Ibn al-Muqaffaʿ, Kalīla wa-
Le Caire, al-Maktabat l-Markaziyya li-l-Makhṭūṭāt Dimna 6, 10–17, 19n33, 20t, 21, 23–25
al-Islamiyya Ayasofia 4213 (A4213), Ibn al-Muqaffaʿ, Kalīla wa-
1169, Ibn al-Muqaffaʿ, Kalīla wa-Dimna 6n16, 33n72 Dimna 5n12, 17n32
Le Caire, Bibliothèque al-Azhar Ayasofia 4214 (A4214), Ibn al-Muqaffaʿ, Kalīla wa-
ms. 323 262, Ibn al-Habbāriya, al-Ṣādiḥ Dimna 6n16, 30n59, 33n72
wa-l-bāghim 332n33 Ayasofya 4215, Kāshifī, Anvār-i Suhaylī 357
Le Caire, Coptic Museum Esad Efendi 3638, Ikhwān al-Ṣafāʾ, Rasāʾil Ikhwān al-
MS Bibl. 90, Bible 122 Ṣafāʾ 302, 315, 319n47
Le Caire, Dār al-Kutub Kadizade Mehmed 333, al-Bisṭāmī, Rashḥ adhwāq
Adab Farsi 61, Naṣrallāh Munshī, Kalīla al-ḥikma al-rabbāniyya fī sharḥ awfāq al‐lumʿa
va-Dimna 156n72 al‐nūrāniyya 74n65
Cambridge, Cambridge University Library Istanbul, Topkapı Sarayı Müzesi Kütüphanesi
T-S Ar.51.60 (TS-AR-051-060), Ibn al-Muqaffaʿ, Kalīla wa- A. 3595, Seyyid Lokman, Shahnāme-i Selim Hān 366
Dimna, feuillet isolé 218–220t, 242, 340n60 Ahmet III 3015 (Sultan Ahmet III 3015), Ibn al-Muqaffaʿ,
Cambridge, Corpus Christi College, Parker Library Kalīla wa-Dimna 34n74
MS Ar.578 (MS 578), Ibn al-Muqaffaʿ, Kalīla wa- Ahmet III 3206, al-Mubashshir, Mukhtār al-ḥikam wa
Dimna 117n70, 118, 120f, 136n35, 164n80, 218–220t, maḥāsin al-kalim 315
340–342 Ahmet III 3519, Bible 122n86
Chicago, Oriental Institute Hazine 359, ʿAlā al-Dīn Sālihzādeh, Hümayūnnāme 359
A 11991, Ibn al-Muqaffaʿ, Kalīla wa-Dimna 5n12 Hazine 363, Naṣrallāh Munshī, Kalīla va-Dimna 319n48,
Collections privées 342nn.(68–69)
Anthologie poétique : Jalāl al-Dīn al-Naqqāsh, Miftāḥ Hazine 761 (H. 761), Niẓāmī, Khamsa 172n2
al-fiṭna fī naẓm Kalīla wa-Dimna, Ibn al-Habbāriya, Hazine 1221-1223, Darīr, Siyār-i Nabī 366
al-Ṣādiḥ wa-l-bāghim, Ibn Makānis, al-Laṭāʾim Hazine 1519 (H. 1519), Ḥusayn b. Ḥasan Sharīf,
wa-l-ashnāf 326–351 Shāhnāma-i Türkī 346
Vente Christie’s, avril 2003, vente 6713, lot 93, Ibn al- Hazine 1597/98 (H. 1597/98), Shūkri Bitlisī,
Muqaffaʿ, Kalīla wa-Dimna 169 Selīmnāma 349–350
Revan 804, Navāʾī, Dīvān 359
Doha, Museum of Islamic Art Revan 843, ʿAlā al-Dīn Sālihzādeh, Hümayūnnāme 358
Ms. 27.1999, Ibn Ẓafar, Sulwān al-muṭāʿ fi ʿudwān al-
atbāʿ 119, 121–125, 205–230 Jérusalem, L.A. Mayer Museum of Islamic Art
Ms. 647, al-Qazwīnī, ʿAjāʾib al-makhlūqāt wa gharāʾib ms. 39–69, Ibn Buṭlān, Daʿwat al-atịbbāʾ 118
al-mawjūdāt 118 Jérusalem, National Library of Israel
Ms.683.2001 (MS.683.2001), Banū Mūsā, Kitāb Yahuda Ms.Ar.1116, Shūkri Bitlisī, Selīmnāma 344, 345f
al-ḥiyal 209n20
404 Index des manuscrits / Index of Manuscripts
Kuwait, al-Sabah Collection, Dar al-Athar al-Islamiyyah New Delhi, National Museum
LNS 104 MS, Ibn Ẓafar, Sulwān al-muṭāʿ fi ʿudwān al- Akhlaq 508 (dépôt de l’Asafiya Library d’Hyderabad),
atbāʿ 104, 119n81, 207n11, 208, 215t Ibn al-Muqaffaʿ, Kalīla wa-Dimna, ʿAṭṭār, Manṭiq
al-ṭayr et Ḥāfiẓ, Dīvān 64
Londres, British Library New York, Metropolitan Museum of Art
Add 4945 (MS Add 4945), Abū’l-Fażl, Iyār-i 1981.373.51/ MMA, 1981.373.51, Ibn al-Muqaffaʿ, Kalīla wa-
Dānish 372n3 Dimna 13t, 233–243, 247–250, 253, 255–261, 263–266
Add. 7293, al-Ḥarīrī, al-Maqāmāt 118 New York, The Pierpont Morgan Library
Add. 15153, ʿAlā al-Dīn Sālihzādeh, Hümayūnnāma 358 S M.500, (Mss 500) al-Marāghī, Manāfiʿ-i
Add. 22406 à Add. 22412, Coran 119n83 Ḥayavān 199n29
Add. 23466, Ibn al-Muqaffaʿ, Kalīla wa-Dimna (ff. 1-70),
Pseudo-Ṣafadī, Lawʿat al-shākī 32n65 Oxford, Bodleian Library
Add. MS 18579, Kāshifī, Anvār-i Suhaylī 380f, 383 E.D. Clarke.Or.09 (Or.9), Ibn al-Muqaffaʿ, Kalīla wa-
IO Islamic 761, Fayẓī, Traduction du Dimna 139n, 143, 146–150, 153–154, 156–162t, 164–165,
Mahābhārata 381n31 167t, 169
Or. 2784, Ibn Bakhtīshū, Kitāb naʿt al-Ḥayawān 105–107, Huntington 264 (Ms. Huntington 264), al-Ṭarsūsī, al-
114, 115, 221n25 Tabṣira fī al-ḥurūb 118
Or. 3900 (L3900), Ibn al-Muqaffaʿ, Kalīla Marsh 458, al-Ḥarīrī, al-Maqāmāt 119, 121f, 230n40
wa-Dimna 34n74 Ouseley 133 (ms. Ouseley 133), al-Tabrīzī,
Or. 4044 (L4044), Ibn al-Muqaffaʿ, Kalīla wa-Dimna Dilsīznāma 345
(ff.1-135), Ibn Ẓafar, Sulwān al-muṭāʿ fi ʿudwān al-atbāʿ Pococke 400, Ibn al-Muqaffaʿ, Kalīla wa-Dimna 17n32,
(ff.136-207) 7, 9–24, 27–29, 32–33, 39f, 40n76 19n32, 30n60, 34nn.(74–75), 99, 109, 111–114, 118–119,
Or. 8571 (L8571), Ibn al-Muqaffaʿ, Kalīla wa-Dimna, Ibn 123n92, 164n80, 218–220t, 242, 246–254, 258, 260–261,
al-Habbāriya, al-Ṣādiḥ wa-l-bāghim 7, 13–21, 23, 24, 264–265, 340n62, 341n67, 344n72
29, 32n65
Or. 9718, al-Ḥarīrī, al-Maqāmāt 118, 119, 229n36 Paris, BnF
Londres, Khalili Collection Arabe 675, Mālik b. Anas, al-Muwaṭṭāʾ 281
MSS 727, Rashīd al-Dīn, Jāmiʿ al-tavārīkh 322 Arabe 2221, al-Idrīsī, Nuzhat al-mushtāq fī ikhtirāq
QUR 807, Coran 120f, 122 al-āfāq 281
Londres, SOAS Special Collection Library Arabe 2631, Pseudo-Idrīs, Traité de géomancie 69
MS 10102, Kāshifī, Anvār-i Suhaylī 110, 111f Arabe 2731, Anonyme, Kitāb al-raml 65
Arabe 2782, Ibn Bakhtīshūʿ, Kitāb manāfiʿ
Manchester, The John Rylands Library al-ḥayāwān 118–119
MS Ar. 487 (Arabic MSS 487 [537]), Ibn al-Muqaffaʿ, Kalīla Arabe 2789, al-Damīrī, Ḥayāt al-ḥayawān al-kubrā (ff.1-148),
wa-Dimna 139n, 147f, 154n72 Ibn al-Muqaffaʿ, Kalīla wa-Dimna (ff. 150v-242) 32n65
MS Ar. 646 (Arabic 646 (706)/ Arabic MSS 646 [706]/ Arabe 2964, Pseudo-Galien, Kitāb al-diryāq 106
MSS Arabic 646), Mille et une Nuits : les aventures Arabe 3210, Ibn Makānis, Diwān 332n33
de Khailakhān (ff. 24-45) ; le roi de Perse et deux Arabe 3465 (P3465/A3465), Ibn al-Muqaffaʿ, Kalīla wa-
vieillards (ff.46-56) ; l’histoire du roi Nuʿmān (ff.57- Dimna 6–7, 10–24, 34n75, 97–98, 106–109, 111–118,
263) 140–141, 165, 169 130–145, 150–151, 154–170, 218–220t, 268–283, 285f,
Milan, Ambrosiana Library 287–289, 291–296, 338nn.(46, 48), 340nn.(61–62),
Ar. 140 Inf., al-Jāḥiẓ, Kitāb al-ḥayawān 118, 124 341nn.(64, 67), 342n68, 344
Montreal, McGill University Library Arabe 3466 (P3466), Ibn al-Muqaffaʿ, Kalīla wa-
Arabic 94 (117953), Ibn al-Muqaffaʿ, Kalīla Dimna 6n16, 7, 9–21, 23–24, 26–29, 31–33, 37–38f
wa-Dimna 6n16 Arabe 3467, (A3467) Ibn al-Muqaffaʿ, Kalīla wa-
Munich, Bayerische Staatsbibliothek Dimna 60–89, 109, 118, 164n80, 218–220t, 237, 244f,
Cod.arab.464, al-Qazwīnī, ʿAjāʾib al-makhlūqāt wa 253, 269, 271–277, 281–283, 285f, 289–291, 338nn.(46,
gharāʾib al-mawjūdāt 106 48), 339, 340nn.(60, 62)
Cod.arab.615 (Cod.Ar.615), Ibn al-Muqaffaʿ, Kalīla wa- Arabe 3470 (A3470), Ibn al-Muqaffaʿ, Kalīla wa-
Dimna 6n16, 169 Dimna 142–145, 152f, 154–165, 168–170, 270, 272–276,
Cod.arab.616 (Cod.Ar.616/ Cod. Arab.616/ Cod Arab 616) 281–282, 285f, 291, 293–294
Ibn al-Muqaffaʿ, Kalīla wa-Dimna 17n32, 74n66, Arabe 3472 (A3472), Ibn al-Muqaffaʿ, Kalīla wa-
97, 99, 102–104f, 109–113, 118, 218–220t, 237, 245f, 253, Dimna 269, 271–276, 280, 282–283, 285, 291–293,
260–262, 264–265, 340n62, 341n67, 344n72 338n48
Index des manuscrits / Index of Manuscripts 405
Arabe 3473 (P3473), Ibn al-Muqaffaʿ, Kalīla wa- 2407, Ibn al-Muqaffaʿ, Kalīla wa-Dimna (ff.1-146v), Jalʿād
Dimna 6n16, 7, 10–21, 23–24, 29 wa-Shimās (ff.150-215) 6n16, 32n65
Arabe 3475 (A3475), Ibn al-Muqaffaʿ, Kalīla wa- 2536 (King Fayṣal 2536) Ibn al-Muqaffaʿ, Kalīla wa-
Dimna 5n12, 6n16, 13t, 172–188, 192–193, 195–201, Dimna 13t, 17n32, 34n74, 164n80, 218–220t
270–276, 280–282, 285, 291–293, 338nn.(46, 48)
Arabe 3476, Ibn al-Muqaffaʿ, Kalīla wa-Dimna (ff.1-253), Saint Petersbourg, Academy of Science
Ibn ʿAbd al-ʿĪd al-Mālikī’s, Ḥilyat al-kubarāʾ wa-bahjat D228, Lettres et Actes des Apôtres 119
al-nudamāʾ (ff.255-340) 32n65 San Lorenzo del Escorial, Biblioteca Real
Arabe 3478, Ibn al-Muqaffaʿ, Kalīla wa-Dimna 7, 29 Ar. 528 (n° 528/ Ms. arabe 528), Ibn Ẓafar, Sulwān al-muṭāʿ
Arabe 3511, Ibn Ẓafar, Sulwān al-muṭāʿ fi ʿudwān al- fi ʿudwān al-atbāʿ 211n23
atbāʿ 150f, 156 Ar. 898, Ibn Bakhtīshūʿ, Kitāb manāfiʿ al-ḥayāwān 114,
Arabe 3593, Ibn al-Muqaffaʿ, Kalīla wa-Dimna 6n14 119, 121–122, 124, 220n25, 225, 228, 230n40
Arabe 3612, (Paris BnF 3612) Mille et Une Nuits 32n65
Arabe 3929, al-Ḥarīrī, al-Maqāmāt 132n26 Tashkent, Bibliothèque A. Biruni
Arabe 4481 : Inventaire des manuscrits de la bibliothèque N°1620, Rashīd al-Dīn, Jāmiʿ al-tavārīkh 310n38
d’Asselin de Cherville, en arabe et en français 143n52 Tashkent, Uzbek Academy of Sciences
Arabe 5847, al-Ḥarīrī, al-Maqāmāt 132n26, 289n62 MS 9109, Kāshifī, Anvār-i Suhaylī 386, 387f
Arabe 5881 (P5881/A5881), Ibn al-Muqaffaʿ, Kalīla wa- Téhéran, Reza Abbasi Museum
Dimna 6n16, 7, 9n30, 10–19, 21, 24–27, 29–33, 30n59, M. 570, Ibn al-Ṣūfī (?), Risālat al-Ṣūfī fī al-kawākib 105
33n72, 35–36f, 269–276, 280–282, 285, 291–293 Toronto, Aga Khan Museum
Arabe 6094 (A6094), al-Ḥarīrī, al-Maqāmāt 117, 131, AKM12, Ibn Ẓafar, Sulwān al-muṭāʿ fi ʿudwān al-
132n26, 268, 270–276, 282, 284–289, 338n48 atbāʿ 119n81, 207n11, 208, 212t
Copte 13, Bible 107f, 108, 115, 116f, 131 AKM14, Akbar et un derviche, feuillet isolé 383, 384f
Copte 114, Lectionnaire des dimanches de carême 281 M.220, Déploration de la mort d’Abaqa Khan, feuillet
Supplément persan 205, al-Juvaynī, Tarīkh-i isolé 322n57
Jahāngushāy-i Juvaynī 310, 319n45 Tübingen, Universitätsbibliothek
Supplément persan 1113, Rashīd al-Dīn, Jāmiʿ al- M.a VI.32, Mille et une Nuits, Histoire du roi
tavārīkh 305, 322 Nuʿmān 140–142, 165, 169
Supplément turc 1243, Version turque ottomane de Kalīla Tunis, Bibliothèque Nationale de Tunisie
va-Dimna 361, 365–371 2281, Ibn al-Muqaffaʿ, Kalīla wa-Dimna 6n16
Paris, BULAC
ms. 579, Muḥammad ʿImrān al-Munajjim al-Jabsūyī al- Varanasi (ou Benares), Hindu University of Benares, Bharat
Faylasūf (attribué à), Texte de géomancie 62f Kala Bhavan
Turc 111, (Ms. turc 111), ʿAlā al-Dīn Sālihzādeh, Ms. 9069, Abū l-Fażl, Iyār-i Dānish 374–379, 381–383,
Hümayūnnāme 358 386
Turc 112, ʿAlā al-Dīn Sālihzādeh, Hümayūnnāme 358 Vatican, Biblioteca Apostolica Vaticana
Turc 114, ʿAlā al-Dīn Sālihzādeh, Hümayūnnāme 358 Vat. Ar. 1177, al-Yamanī, Muḍāhāt Kalīla wa-Dimna, ff.25r-
Turc 205, version turque ottomane de Kalīla va- 52v 2, 10–12t, 24
Dimna 359–364, 366, 371 Venise, Biblioteca Nazionale Marciana
Paris, Institut catholique Inv. Cod. Marc. Or. 90 (=57) (Cod. Or.XC = 57), Aḥmadī,
Copte-arabe 1, Bible 117 İskendernāma 344n74
Vienne, Nationalbibliothek
Rabat, Bibliothèque nationale du Royaume du Maroc A. F. 9, al-Ḥarīrī, al-Maqāmāt 119, 121–122, 124, 225,
Ms. 3655 (BRR3665/Rabat 3655/BRR 3655), Ibn al- 230n40
Muqaffaʿ, Kalīla wa-Dimna 5n12, 6n16, 32n67, 34n75,
172–200, 218–220t, 270n16, 301–323, 341n67 Washington, Freer Gallery of Art
Rampur, Raza Library F1954.1, Ibn Ẓafar, Sulwān al-muṭāʿ fi ʿudwān al-
MS 2982, Naṣrullāh Munshī, Kalīla va-Dimna 111, 113f atbāʿ 119n81, 207n11, 208, 213t, 223n26
Riyad, King Faysal Center for Research and Islamic Studies F1954.2, Ibn Ẓafar, Sulwān al-muṭāʿ fi ʿudwān al-
for Research and Islamic Studies atbāʿ 119n81, 207nn.(11, 13), 208, 215t
Index des noms propres / Index of Names
Les noms sont classés selon l’alphabet latin. L’article al- n’est pas pris en compte dans le classement. Les personnages et auteurs
connus sont listés sous leur appellation la plus commune. En raison de sa forte prévalence au sein du volume, le nom d’Ibn al-
Muqaffaʿ n’est pas indexé.
Bahrām Shāh, sultan 45, 106, 379n28, 381, 383, 386 Ibn ʿAbd al-ʿĪd al-Mālikī 32n65
al-Bakrī, al-Aṣamm (al-Aṣamm al-Bukayr) 134 Ibn Abū al-Minā (Nasīm b. Yuḥannā b. Abū al-Minā) 141,
Balʿamī 43n4 169
Bayezid II, sultan 326n6, 345 Ibn Bakhtīshūʿ 105–107, 114, 115f, 118–119, 121f, 122
Bayqarā Ḥusayn, sultan (Ḥusayn Mirzā) 44, 357, 381, 386 Ibn Buṭlān 118
Baysunghur, sultan 361 Ibn al-Dumayna 134
Baysunghurī, Shams al-Dīn 361 Ibn al-Durayhim al-Mawṣilī 122
Bellini, Gentile 349n100 Ibn al-Habbāriyya 7, 32n65, 327, 331
Bellini, Giovanni 348–349 Ibn Ḥajar, Abu’l-Qāsim b. Ḥammūd 206
al-Bisṭāmī 74 Ibn al-Ḥātim 72
al-Bukayr, al-Aṣamm (al-Aṣamm al-Bakrī) 134 Ibn Makānis 327, 331, 332n33
al-Bukhārī, Muḥammad b. ʿAbd Allāh 44n4, 101 Ibn al-Nadīm 45, 100–101
al-Bukhārī, Muḥammad b. Ismāʿīl 63 Ibn al-Ṣūfī 105
al-Būnī 73 Ibn Ẓafar al-Ṣiqillī 7, 32, 115n65, 119, 121–123, 150f, 205–208,
212t
Camboust, Henri-Charles du 70 al-Idrīsī 281
Cherville, Asselin de 143 Ikhwān al-Ṣafāʾ 56n34, 108n55
Clarke, Edward Daniel 147, 154, 156
Jahāngīr, empereur (prince Salīm) 378, 383
al-Damīrī 32n65 al-Jāḥiẓ 118, 124
Dawlat 383 Juvaynī, ʿAṭā Malik 310, 319n25, 320
Diez, Heinrich Friedrich von 265
al-Dimashqī, Ghāzī b. ʿAbd al-Raḥmān 119 Kamāli Chela 322
al-Dimashqī, Muḥammad b. Aḥmad Sakīkū 207n14 Kara Memi 359
Index des noms propres / Index of Names 407
Kāshifī 43n1, 44–49, 52, 55, 57–58, 101, 110–111, 357, 372, 380, al-Rabbāṭ, Aḥmad 142, 154
383, 386–387 al-Rabbāṭ, Ḥasan 4
al-Kawākibī, Aḥmad 207n14 Rashīd al-Dīn 302, 305, 320, 322
Kesu Kālān 322 Rūdakī 43n4, 45–46, 106, 379n28
Khālid b. Barmak 172 Rūmī 322
Kınalızade ʿAli 350–351 al-Ruqayyāt, ʿAbd Allāh b. Qays 134
Kraus, Hans Peter 327, 330
Kubilaï Khān 304, 322 Sacy, Silvestre de 5, 50n22, 131, 206n8, 358n6
al-Ṣādiq, Jaʿfar 61n6, 64
La Fontaine, Jean de 4 Sahid, David 357
Saladin, sultan 118
Maillet, Benoît de 131, 142n46, 150 Ṣalāḥ al-Dīn 207n14
al-Malik al-Ashraf, sultan 105 Sālihzādeh, ʿAlā al-Dīn 357–358, 366
al-Manṣūr, calife 45, 172, 379 Salīm, prince (empereur Jahāngīr) 378, 383
Maslama al-Qurṭubī (pseudo-Abū al-Qāsim Maslama b. Salīm Aghā 207n14
Aḥmad al-Faraḍī al-Majrīṭī) 74nn.(63–64), 77n78, 78, Ṣandal 119
79n82, 82nn.(86–87), 84n93, 87 al-Sanjalafī, ʿAbd al-Raʿūf / ʿAbd al-Rawf 270
al-Masʿūdī 99 Sayyid Lokman 361
Matrakçı Nasuh 346 Selīm Ier, sultan 325, 327, 347–348
Maurepas, comte de 131 Soliman / Süleyman, sultan 332, 344, 347, 357, 359, 366
Meḥmet Çelebī b. Mustafā Çelebī 358 Sūryāl b. Abū al-Minā 169
al-Mubashshir 315 al-Suyūtī 331
Muḥammad b. Mubādir 119
Mustafā b. Ḥusāmī 366 al-Ṭabarī 108, 130n2, 172
Mustafā b. Ibrāhīm 358 Tabbagh, George 331
al-Mutanabbī 144n57 al-Tabrīzī, Badīʿ al-Dīn 345
Talikizāde 361
al-Naqqāsh, Jalāl al-Dīn al-Ḥasan b. Aḥmad 326, 327n15, al-Tanūkhī 32n65, 34n73
331–333 Ṭūsī, Abū Manṣūr Aḥmad b. ʿAlī Asadī 43n4
Nasīm b. Yuḥannā b. Abū al-Minā (Ibn Abū al-Minā) 141,
169 al-ʿUmarī, ʿAbdallāh b. Muḥammad 6
al-Nāṣir, calife 105
Naṣrallāh Munshī 34n74, 43n4, 44–46, 55, 101, 111, 113f, Vansleb, Johann Michael 169
319n48, 372–373, 374n16, 379n28, 383, 386 Varsy, Jean-Georges 140–141
Naṣr II b. Aḥmad, émir 106
Naṣūḥ Pasha 207, 208n14 al-Wāsiṭī, Yahyā b. Maḥmūd 187, 199–201
Navāʾī 359
al-Yamanī, Abū ʿAbdallāh Muḥammad b. al-Ḥusayn b.
Pococke, Edward 261 ʿUmar 8, 9n30, 34, 50n22, 105n38, 144n58, 334
Polo, Marco 323 al-Yaʿqūbī 30n60
Pseudo-Idrīs 69n23
Pseudo-Ṣafadī 32n65 al-Zanātī, Muḥammad 61, 65