KALILA WA DEMNA

iii. Illustrations

Kalila wa Demna is a collection of didactic animal fables, with the jackals Kalila and Demna as two of the principal characters. The story cycle originated in India between 500 and 100 BCE, and circulated widely in the Near East. The fables were translated into many languages, undergoing significant changes in both form and content. The Sanskrit original and several significant early translations have not survived. In Sanskrit literature the story cycle is known as Pañcatantra, while it was often called Fables of Bidpai in early modern Europe.

The popularity of the text of Kalila wa Demna and its various translations is reflected not only in the number of medieval illustrated manuscripts, but even in reports of much earlier ones that have not survived. For example, in the preface to the Arabic translation, Ebn al-Moqaffaʿ (d. ca. 757) states that the text was to be illustrated. One of the most telling pieces of evidence is Ṭabari’s report of one of the charges at the trial of the Persian general Afšin at the ʿAbbasid court, who was accused of owning a heretical and lavishly illuminated manuscript. In his defense the general accused the judge of having a copy of Kalila wa Demna at his house (O’Kane, 2003, p. 28). This is significant, not only in showing the popularity of the text, but also for inferring the possibility that perhaps there were illustrated manuscripts at this early date.

Although Ebn al-Moqaffaʿ added several tales to the Sasanian recension of Borzuya (fl. 531-79), an analysis of the illustrations in later manuscripts shows that the five original tales of the Pañcatantra were the ones most frequently illustrated. The literary device of nesting—that is, the narrator of a frame story has a character break off to tell a new story—is familiar from the Thousand and One Nights. Its frequent employment in the original tales of the Kalila wa Demna added to the variety of choices for the illustrator. This may explain why within these five tales there is also a bias, in terms of the rate of illustration, towards those frame stories that comprise a greater number of subsidiary stories (O’Kane, 2003, pp. 24-25).

The earliest illustrated Arabic manuscripts that have survived date from the beginning of the 13th century, while the earliest illustrated Persian example (Istanbul, Topkapi Saray Library, MS H. 363) is from the last quarter of that century. The Arabic version is second only to the Maqāmāt by Ḥariri (d. 1122) in terms of surviving medieval illustrated texts. In the middle of the 14th century an atelier, probably in Damascus, seems to have been responsible for a group of six closely related Arabic manuscripts, five of which were illustrated (O’Kane, 2003, Appendices 3-7; for the recently published sixth manuscript in the King Faisal Centre for Islamic Studies and Research, Riyadh, Saudi Arabia, see Jenaid).

The earliest illustrated manuscripts in Persia are provincial productions. It is therefore surprising that in the third quarter of the 14th century a manuscript (see the preserved leaves in the Shah Tahmāsp Album, Istanbul University Library, F. 1422; see Figure 1) was produced probably under the patronage of the Jalayerid ruler Šayḵ Ovays (1356-74). It broke new ground in its treatment of pictorial space in Persian painting, using the previously neglected marginal spaces as extensions to the blocks inserted within the space previously left for the text. The visual complexity was accompanied by a concomitant increase in psychological dynamism, with the margin itself becoming the location of the main protagonists of the story in some instances (O’Kane, 2003, Appendix 15). These characteristics, together with its newfound skill in the depiction of landscape perspective and texture, mark it as one of the most important painted manuscripts of its time.

Royal interest in the work continued with two manuscripts from the atelier of Bāysonḡor b. Šāhroḵ (1397-1433; q.v.) in the Topkapi Saray Library (MSS H. 362 and R. 1022). One of these (MS H. 362; cf. O’Kane, 2003, Appendix 17; see Figure 2) is particularly interesting because its illustrations are probably Jalayerid paintings from around 1375-85, cut from their original manuscript and pasted into a new manuscript. This is evidence of the esteem with which Bāysonḡor’s atelier regarded these paintings, which also served as a model for several paintings in the new manuscript. Although the artist of Bāysonḡor’s version followed the stiffer depiction of figures that was characteristic of the school, the clarity of details and formalized setting sometimes transcends this limitation, giving the images an intense graphic power.

One more outstanding illustrated version of the work from the 15th century is known (Tehran, Golestān Library, MS 827, cf. Robinson; see Figure 3). Long misattributed to Bāysonḡor’s atelier, it was probably produced for the Qarā Qoyonlu governor of Baghdad, Pir Budaq (d. 1466). It is outstanding for the almost ethereal delicacy of the color palette of its landscape settings, particularly noticeable in its depiction of rocks.

When Ḥosayn Wāʿeẓ Kašefi (d. 1504-5) reworked Naṣr-Allāh’s text in the Anwār-e Sohayli at the end of the 15th century, his text accomplished several things: it eliminated the Arabic verses, it added several new stories, it introduced a more florid style, and it supplanted the popularity of the older text. However, it also signaled the waning of interest in illustrated manuscripts of the work in Persia, since only one fine illustrated example (1002/1593) was subsequently produced for a Persian patron (Geneva, Sadruddin Agha Khan Collection; cf. Welch, pp. 125-43). But since the patron in question was Ṣādeqi Beg Afšar (d. 1609-10), the Safavid court painter and chronicler who was for a period head of the royal library, the paintings, perhaps all by the hand of Ṣādeqi Beg himself, are both numerous and of high quality.

Another fine copy of the Anwār-e Sohayli was begun in Central Asia, but the bulk of its paintings were added at the Mughal court sometime in 1585-90 (London, SOAS Library, MS 10102; cf. Natif). By this date attention had shifted to another recension, that of Abu’l-Fażl ʿAllāmi (d. 1602), entitled ʿEyār-e dāneš, made for the Mughal court in 996/1587-88, which simplified the style but added yet more stories of Indian origin, although it did not entirely supplant the Anwār-e Sohayli. A number of fine illustrated versions were made for Akbar (r. 1556-1605) and Jahāngir (r. 1605-27), of which the Chester Beatty Library ʿEyār-e dāneš (MS 4, ca. 1595) is one of the finest and most extensively published (Leach, pp. 74-105). Its paintings display the naturalism that is characteristic of the mature Mughal style, although sometimes this comes at the expense of the drama present in earlier Arabic and Persian manuscripts.

Some illustrated versions are also known to have been made in Ottoman Turkey (Grube, 1993, pp. 391-93), but they do not approach the Iranian or Indian examples in quality or quantity.

Why was the work so popular as a choice for illustration? Although its moralizing aspects are an ostensible raison d’être, the lively storytelling, particularly embodied in the nested stories, must have most attracted readers and artists, especially since they gave an opportunity to illustrate risqué scenes while simultaneously deploring their moral laxity. This explains why the nested stories were illustrated more frequently than the frame stories (O’Kane, 2003, p. 25, Table 1).