The Shahnamah of Muhammad Juki, a masterpiece of the Islamic book, is the most famous manuscript in the collection of the Royal Asiatic Society. While it has features in common with two other slightly earlier Timurid royal Shahnamahs, those of Baysunqur Mirza and Ibrahim Sultan, it is a work of extreme refinement, being devoid of the slightly rebarbative steely brilliance of the former and the mannered, archaic effect of the latter. It has never had the monograph it merited, and this it now receives in Dr Brend's sumptuous, but also very useful, publication.

The colophon is lost, but the script shares certain idiosyncracies with that of the Shah Rukh Khamsah in the Hermitage Museum, dated Rabi‘ ii ah 835 (December ad 1431) by a certain Mahmud who was a pupil of Ja‘far al-Baysunquri and may plausibly be attributed to him. Though the quality of the Muhammad Juki illustrations is more consistent and generally much higher, the fact that the same copyist was employed is an important link with Shah Rukh's own library. The manuscript is unfinished: two spaces for miniatures were left blank and were filled in Mughal India (nos 29 and 31). The former, showing the battle of Gav and Talhand (no. 29, folio 430b), which is pasted in, ingeniously crowds the figures into four groups, leaving a central space filled with a series of single combats, with the brothers symmetrically opposed, and was, Dr Brend suggests, very probably the work of a Bukhara painter at the Mughal court.

A number of pages in the second part, in the same hand, which make lavish use of text on the diagonal, led B.W. Robinson to suggest that to hasten its completion they were inserted to replace as many as nine pages with space for an illustration on one side, when Muhammad Juki's health began to fail, and Dr Brend (p. 46) lists the subjects of the miniatures they may have supplanted. Robinson's collation of the text with the Bombay lithographed edition of 1849, also indicated that there are lacunae between pp. 7 and 8; between pp. 41 and 42; and between pp. 425 and 426. These could account for a further three full-page illustrations and one half-page. At some point in its history the manuscript was remargined, and some catch-words and the marginal projections were rather clumsily cut out and glued to the new mounts.

Dr Brend's close reading lays emphasis on the expressionism of the paintings, conveyed as much by pose and facial expression as by gesture. It is also worth remarking that, as on folio 67b, showing the death of Suhrab and folio 243a, showing the Paladins in the snow, both composition and colour scheme deliberately exploit the pathetic fallacy. In the former the scarlet flowers dotting the landscape are like spots of blood. The tragedy is heightened by the somewhat wooden stance of their mounts. The Paladins in the Snow, is a brilliantly realistic depiction of drifting snow and the resignation of the frozen warriors. And, technically a complete innovation, the purplish-grey of the fallen snow combines to exceptional dramatic effect with the lowering clouds. Another example is the Fire Ordeal of Siyavush (no. 8, folio 76a) shows not only the flames but a dense surrounding aura of grey smoke. Unlike the worked-up version in Berlin, tentatively attributed to Shiraz circa 1420–25Footnote ¹ with a more richly clad horseman facing to the right, his horse springing out of the flames, however, the tranquillity of Siyavush's pose and his horse not even at full gallop lends this scene dramatic irony.

The use of iridescent colour-contrasts and coral-like needles of rock to represent receding planes, a feature of the 1410 Miscellany for Iskandar Sultan, is much more developed even than in Baysunqur's Shahnamah, and could be compared to the ingenious use of coloured flats to group figures and divide up space, as on a stage. This comparison is not far-fetched, since scholars have noted a certain theatricality in Timurid painting, and in literature too, as in Michele Bernardini's close reading of Hatifi's Haft Manzar. More exceptionally, the illustrations are distinguished by touches of humour, such as the gleeful expression of the div Akvan who is about to toss Rustam into the sea (no. 14, folio 165b), or the infant Rustam and the absurdly dwarfed mad elephant (no. 4, folio 32b).

In their present state the three brothers’ Shahnamahs have only four subjects in common. Muhammad Juki's shares seven with that of Ibrahim Sultan and only one with Baysunqur’s, and their choice was obviously in part reflected by their personal preferences. Could this have gone as far as portraying the patron? The Battle between Gushtasp and Arjasp (no. 22, folios 269b-270a), a double-page composition, repeats a schema in the Shahnamah of Ibrahim Sultan of circa 1430, which is not an illustration to the text but panegyric, and the white horse of the commanding figure on the right-hand page copies that of the hunting frontispiece of the Shahnamah of Baysunqur. To take this as a portrait of Muhammad Juki raises problems in the interpretation of the text and Dr Brend considers that a more plausible suggestion would be the figure of Gushtasp, whose features are highly individualised, playing polo before the Qaysar (no. 21, folio 252a). Observing the facial resemblance of the commanding figure to that of Khusraw in the 1431 Khamsah for Shah Rukh, moreover, she ingeniously suggests that he is the figure shown here.

The dominant influences on the illustrations were the 1410 Miscellany of Iskandar Sultan (British Library, Add. 27261), the Shahnamah of Baysunqur, and, to a lesser extent, the Shahnamah of Ibrahim Sultan. The importance of the second is unsurprising, since Muhammad Juki's painters were almost certainly recruited from the Herat studio, but, even so, their lyrical colourism and uncrowded compositions set it apart even from Baysunqur's manuscripts. The influence of these earlier manuscripts raises the question whether they had come into Muhammad Juki's possession (though this may appear unlikely in view of the location of his estates in Khuttalan on the Upper Oxus) or whether Shah Rukh transformed the manuscripts and the studio left by Baysunqur on his early death into a royal library where works were available for reference or consultation.

Dr Brend distinguishes seven painters on stylistic grounds, each painting evidently the responsibility of a single individual, from preliminary drawing to the finished work. There is too little illumination to decide which, if any, of them did that. Of these, the most important are the three principal painters A, B and C. A is especially marked by the influence of the Baysunqur Shahnamah. B's work is also marked by the influence of Baysunquri manuscripts, though also possibly by the Hermitage Khamsah for Shah Rukh: typical of his work are the Ordeal of Siyavush and the Paladins in the Snow. The compositions of painter C, probably the greatest virtuoso of the group, include Tahminah and Rustam (no. 6, folio 56b), and the Brazen Fortress (no. 23, folio 278a). Dr Brend stresses the continuity between the Shahnamah and a Khamsah of Nizami for the Karakoyunlu princess, ‘Ismat al-Din, Muhammad Juki's widow, dated 849 (1445–6), the year after his deathFootnote ² . Its colophon states that it was copied by Yusuf al-Jami and illuminated and illustrated by Khwajah ‘Ali al-Tabrizi, though of the fourteen illustrations one was never executed and two pages are Ottoman work of circa 1500. Five of the remainder are excellent copies of earlier works, in which she sees the hand of painter B of the Muhammad Juki Shahnamah, and identifies him as Khwajah ‘Ali al-Tabrizi; and two are by the painter in the Muhammad Juki Shahnamah whom she identifies as painter C. These are of Bahram Gur hunting gazelles, which she parallels with Firdawsi and the court poets of Ghazni (no.1, folio 7a), and Majnun at the Ka‘bah (H. 781, folio 111b), which she parallels with its most brilliant composition, the Brazen Fortress (no. 23, folio 278a) and the escape of Qubad (no. 28, folio 394a).

The dearth of extant manuscripts between the death of Muhammad Juki and the rise of the court scriptorium of Sultan Husayn Bayqara makes the place of his Shahnamah in later Timurid painting difficult to assess. But the visit of Iskandar to the Hermit in the Khamsah of 900/1494–5 (British Library Or. 6810, folio 273a) is a reworking, to even greater dramatic effect, of the Brazen Fortress. The influence of the manuscript on Mughal painting also appears to have been limited. However, a double page illustration in the V&A, from ‘Abd al-Rahim Khan-i Khanan's translation of the Baburnamah of 1589, showing Babur's return from Herat to Kabul makes use of various of its compositions, including the Paladins in the Snow, and Dr Brend suggests that it may have been this patron who ordered the painting of Gav and Talhand. It is unclear when the manuscript returned to the Royal Library, but the Brazen Fortress evidently influenced some pages in the Akbarnamah of c. 1590–95.

The documentary importance of two folios, 3a and 536a, covered with inspection- and transfer-notes, valuations and seal-impressions had long been appreciated and conservation of the manuscript revealed a third, 536b. These combine to give the subsequent history of the manuscript, which takes its place with other famous manuscripts, such as the Khamsah of 1494–5 (British Library Or. 6810) and the Haft Awrang of Jami in the Freer Gallery, among the most highly valued of the Mughals’ collection. These have all been deciphered, translated and judiciously set in context by The Late Dr A.H. Morton (pp. 165–177). The task was arduous for the notes generally give dates in the form of regnal years, without any specification of whose reign it was, and inconsistencies in calculating them from ruler to ruler further complicate the dating. Possibly the most interesting revelation is that in the later sixteenth century the manuscript was in the hands of two private owners, which prompts a revision of the received view of the Mughals’ library. Dr Morton concludes that before Akbar it is not possible to think of the Royal Library as a permanent collection, and even then successive rulers saw it as their private property rather than an independent institution. It did not, moreover, survive the mid-eighteenth century and its organisation may have seriously weakened after Awrangzib's death.

There are five Appendices. Appendix A gives the subjects and the break lines that is, the line immediately preceding the space for a miniature and upon which the illustration depends. Appendix B gives comparative tables of the miniatures in the Shahnamahs of Ibrahim Sultan, Baysunqur and Muhammad Juki. Appendix C groups the attributions of the miniatures by style. Appendix D gives readings for the notes and the seal-impressions on folios 3c, 536a and 536b. And, as a final bonne bouche Appendix E reviews the graphic documentation of the visit to Lucknow in 1814–15 or 1818 of Lord Moira, later Marquess of Hastings, when the Navvab Vazir of Oudh presented the manuscript to him.