The Paris Kitâb al-Diryâq manuscript, arabe 2964 at the Bibliothèque nationale de France must have been an extraordinary work of art when it was completed in rabi‘ alawwal 595 A.H., January 1199 C.E., by “the weakest of the slaves of God,” as he calls himself in the colophon. His name is Muhammad ibn Abi al-Fath ‘Abd al-Wahid and he came from a family of pious and learned men, and somehow connected with the organization of the pilgrimage to Mekkah. As it exists today, the manuscript is incomplete and its remaining folios have been assembled into a book at a much later time without consideration for their original sequence. Nothing is said in the preserved folios about where the manuscript was copied, illustrated, and decorated. Nor is there anything about place of origin in the companion manuscript to the Paris one, (a Kitâb al-Diryâq in the Nationalbibliothek in Vienna (AF 10), which is considered to be somewhat later but still within the same period. A couple of unillustrated examples of the same text are equally silent about their place of manufacture.

All that can reasonably be assumed is that the Paris manuscript, like most others, came from the primarily Arabic speaking area of the Muslim world. But even this point must be made with caution, since Arabic was, at least until the fourteenth century, the pan-Islamic language for learning and scientific information. My concern in this essay will be primarily to comment on the significance of this manuscript for the historian of art and to elaborate on the esthetic merits of its features. I shall conclude with a few hypotheses on the unresolved problems it poses to the historian, and the geographical, historical, or cultural setting which can be proposed for its creation. These remarks are only preliminary views and judgements which could certainly be refined and modified after a thorough indepth analysis of all the elements found in the decoration of the manuscript1.


The sixth century of the hijrah or the twelfth century in the common era was one of momentous changes within the Islamic world. Politically, power was, from al-Andalus to northern India, in the hands of newly arrived Berber, Kurdish, or Turkish military rulers who established many new dynasties (Almohads, Ayyubids, Ortoqids, Seljuqs, Ghorids, Khorezmshahs). They eliminated the heterodox shi’ite regime of Fatimid Egypt which, from Cairo, its brilliant new center, had dominated much of the Near East for nearly two centuries. They eventually defeated the imported state of the Crusaders from the West and transformed dozens of hitherto small cities into dynastic capitals. They fostered an international commerce which enriched the rulers as well as the artisans and merchants of the whole area, especially in whatever was at the core of the Islamic world, between Egypt and Central Asia.

Starting early in the twelfth century and with growing intensity after the middle of the century, an explosion of artistic activities engulfed theMuslim world. It affected architecture, as often alien rulers as well as a number of prominent local families sponsored mosques for regular prayer, madrasahs for the training of legal and theological elites, palaces and citadels for the families and courts of the rulers, commercial monuments like bazaars and caravanserais for merchants and artisans from many lands and many religious allegiances, schools and hospitals for the needs of all Muslims, and funerary memorials for themselves and for revered holy men and women. It was the time when the muqarnas, that ubiquitous geometric composition used in constmction as well as decoration, spread from east to west, even into Christian Armenia or Sicily, as a unique form of exclusively Islamic origin. And, in a more general way, this is the time of triumph for geometry in ornament.

In the arts of ceramics, glass, and metalwork, probably also of textiles which have not been as well preserved, new techniques were developed which allowed for two novelties. One was the possibility to elaborate very small details and therefore to create complicated patterns of design. The other one was the opportunity to multiply representations of people and of animals. The latter had never completely disappeared from Islamic art, but their presence, already restricted by religious convictions and social pressures, was further hampered by an artisanal technology that was not well adapted to precision of detail in the making of lines and to the preservation of many colors on objects. Now representations appeared almost everywhere (with a partial exception for North Africa).

The art of ceramics was revolutionized, especially in Iran, through the appearance of several competing techniques of decoration. Glass and metalwork were equally transformed everywhere. These revolutionary changes eventually affected the making of books which acquired illustrations in the thirteenth century, as with a number of manuscripts of Dioscorides’ Herbal, al-Hariri’s celebrated Maqamat, the stories of Kalilah and Dimnah, and any number of technological or pseudo-scientific texts. The earliest dated of these illustrated books is the Paris Kitâb al-Diryâq. Small fragments have remained of probably earlier, although not dated, illustrated books, but they are few in number and it is almost impossible to relate them to any specific cultural or geographic setting. We do not know, for instance, whether the fact that so many of them came from Egypt is the result of a peculiarly Egyptian artistic creativity or of the greater ease with which old documentsare preserved in the dry environment of the valley of the Nile. Should the Kitâb al-Diryâq be therefore considered as the first step in a newly developing art which will florish in the following century? Or is it the only remaining example in Arabic and around 1200 of a long tradition of illustrated books going back to the Late Antique and exemplified by Greek as well as Arabic, and possibly Latin as well as Persian, manuscripts dealing with a common source of knowledge? Or is it an exceptional and unique manuscript which no doubt used themes and motifs from an existing older vocabulary of available forms and approaches to design, but which was meant for a very specific patron, a specific taste, or a specific occasion?

An answer to these questions is made difficult by two regrettable features. One is that, in contradiction to works of literature like the Maqamat or even to semi-scientific ones like the herbals of Dioscorides, only a few, five known to me up to now, manuscripts of the Kitâb al-Diryâq exist and only two are illustrated. This makes one wonder about the popularity of the work and, as a consequence, about the social, cultural, or even professional range of its audience; it restricts how much one can use it to define the art of a time or of an area. The other regrettable feature is that the manuscript has been severely damaged by time and especially by men. Whole sections are missing and it is possible that some additional information about the patron of the manuscript or the circumstances of its making had been included. It is probable, however, that the missing sections consisted exclusively of text, perhaps with a few instances of calligraphic designs; but missing images cannot be excluded.. What remains of the manuscript may well have been a conscious selection of the manuscript’s most original pages preserved by some later amateur-collector from a codex which had been damaged already, as is clear from its first and last folios.

In other words, someone would have already in the past recognized the esthetic merits of the manuscript and kept its pages in the ways in which later Persian albums will deal with writing and with images, as a sequence of pages remarkable for their visual qualities rather than for their narrative accounts. Whether such an esthetic intent was also the point of the original manuscript remains to be seen. But the following question can legitimately be raised: was it from the beginning that this manuscript was meant to be seen rather than read Much work on many details is needed before this and other questions raised by the manuscript can really be answered. My objective in this essay is simply to raise the questions posed by this manuscript for the historian of art and to entice readers to develop questions and queries for scholars to answer. I shall discuss four features of the manuscript: its introductory pages and frontispieces, its calligraphic pages, the “portraits” of authors, and the narrative illustrations. I will leave aside for others to pick up the herbal and the snakes which require comparisons and analyses which are beyond my competencies. In conclusion, I shall try to sketch out a hypothetical image of the patron-maker-user of the manuscript, three functions which, as we shall see, may well have been exercised by the same person or at least by the same family.

Introductory folios

Unusually for a relatively small codex, the Paris Kitâb al- Diryâq comprises three separate units. There are three pages (pp. 1, 2, and 3, according to the sequential numbering we are proposing in this volume) containing what appears to have been meant as a table of contents with 66 horizontal cases for titles of chapters or of lists of various kind. The content of the list itself has been erased or was never completed, a rather curious feature for a fancy manuscript. The sets of cases are surrounded by a thick arabesque border containing an inscription which repeats the colophon and gives the name and patronymics of the scribe (katib), Muhammad ibn ‘Abi al-Fath; the text contains the word sahib next to katib, which could but does not necessarily have to be interpreted as “owner”. The letters of this inscription, originally in gold, may well have been redrawn on two of the folios, thereby indicating that, at least for a while, the manuscript remained in good shape.

The second unit (pp. 4 and 5) follows a pattern favored by makers of fancy books since the early part of the twelfth century: a sparkling medallion on an octagonal field framed by thick bands of arabesques and of writing. One page contains the title of the book and the other one identifies its destination, the library (literally the treasury of books) belonging to a learned man, one Muhammad ibn Jamal al-Din Muhammad, who turns out to be, in all likelihood, the nephew of the copyist and presumed decorator of the manuscript. The third introductory unit consists of a two pages frontispiece (pp.6 and 7) with two medallions of celestial figures inscribed within a rectangular frame whose upper and lower borders repeat the name of the scribe and add to his identification as the scribe that he was also the owner or patron (sahib) of the manuscript.

Although rare in twelfth century manuscripts, none of these features are unique or really new. I shall return to their artistic merits, but the important and seemingly unique feature of this manuscript is the repetition of the same information about the scribe, the recipient, and the donor, a sort of publicity about the people involved with the making and usage of the book rather than about the content of the book itself. It is a formal acknowledgment, almost a proclamation, of the pride the artist felt for the high esthetic quality of his work and for the pious, moral, and learned qualities of those or the one for whom the book was made. Like luxury publications today, the Kitab al- Diryaq repeats several times the title of the book and how wonderful are its user and patron. In short, these sets of preliminary pages, while not unusual by themselves when seen separately, are remarkable for being all there, illustrations or reflections of some particularly unusual setting for the book, as though someone was trying to publicize his work and to be sure than the user/owner will not forget him.

Calligraphic pages

The most unexpected features of the Kitâb al-Diryâq are the range and the composition of the scripts used in it. The anecdotal narrative is expressed in an elegant cursive with careful attention given to diacritical marks and to a rich array of rosettes seemingly serving as means of punctuation. A similarly clear cursive is used for the factual information provided in the frontispieces. But titles, captions, and repeated identifications of patrons and of authors and copyists are given in magnificently crafted versions of a so-called “oriental” Kufic angular script, often displayed together with a rich background of arabesques and of vegetal ornament. The presence of two scripts throughout a manuscript was usually limited to a few fancy Koranic manuscripts and should be considered in our case as an attempt to give an esthetic dimension to a text of relatively pedestrian importance. This esthetic impulse is even more obvious in the composition of individual pages.

Not only do we have carefully thought out horizontal bands of varying heights for texts, at time switching from one style of writing to another. But also some of the pages (16,18, 19, 29, 22-28, 32) with recipes for the making of antidotes are arranged in a variety of ornamental geometric patterns: triangles of oblique lines, rectangles with’ diagonal lines of writing, circular patterns, or even a design spread over two pages (pp. 40-41) creating assort of large rug or silk covering the pages of the manuscript. Why was this done? What was accomplished by it?

First, it made each page into a separate object, into a work of art, a phenomenon better known several centuries later with Persian manuscripts and albums. But there are already twelfth century, and possibly earlier, Koranic pages with the same concern for the design of a page and with the same use of several scripts. It may eventually be possible to discover an iconographic significance in the patterns used, as on the pages where the word for “honey,” the binding used for so many antidotes, appears in the center of a pattern of words identifying other ingredients. At this stage, not enough thought has been given to the organization of pages in Arabic manuscripts to justify such a meaning, however tempting it may be.

Whatever answer may eventually be given to the iconography of writing in early manuscripts, this particular codex became, like later albums, an assemblage of individual pages, each one a work of carefully crafted design with very few repetitions from page to page. Second, these arrangements did not consistently make visually clear the difference between a narrative text describing the elements of an antidote or retelling an anecdote, technical information on recipes or on snake types, and information about the title and producers of the book. They do not reflect the content of the book. Their purpose is primarily to please and to attract. In fact, the small size of some scripts, the unusual feature that some words or sentences are set vertically or even upside down in order to satisfy the needs of a design, and the difficulty of deciphering the words transcribed in Kufic script, were all features that suggest that the point of the book was not really to be read (although one could do so if one really wished and if one had the right competencies) but to be seen, to be looked at, to be appreciated sensually. It could even be understood as a game with a succession of boards proposing different ways of encoding and deciphering a text or simply creating a soothing object, attractive to contemplate. Through these pages, this manuscript becomes a wonderful example, one of the earliest ones in manuscripts (in contrast to ceramics or architectural decoration), of the transformation of writing into art for its own sake rather than maintaining it exclusively as a vehicle for communication of information. And it is in order to broadcast his own success in making this sort of statement as well as to honor the designated owner of the book that our copyist spent so many lines advertising himself and his purpose.

Portraits of authors

Nine Greek physicians, properly identified by name, are shown on the pages sitting in 6 rectangular cubicles provided with common signs (curtains) for the depiction of interior spaces and with various implements (lamps, stools, jugs, pen boxes) associated with writing. In three instances , a single bird flies into the space or fills an empty area within the frames of the cubicles. In two instances, two birds are shown symmetrically in the upper comers of the cubicle. Three physicians are alone and reading or writing and six sit with a smaller companion, a junior colleague, a student, or simply a devoted follower. All the personages wear a turban and are dressed in heavily decorated robes typical of the clothes worn by the wealthier upper classes in thirteenth century painting. Their feet are either bare or covered with simple slippers, another sign that the images were meant to show interiors. The heads of all the personages are surrounded by a halo and some of the haloes may well have had the rare feature of a fiery edge. Although comparable in type, the poses of the doctors vary in many details and it is clear that the artist tried to avoid the automatic sameness of such groups of people sharing the same function as they appear in other manuscripts.

As with the writing, it is possible that someday we will be able to leam to read such images and to recognize details with iconographic significance. But at this stage of knowledge and sensitivity to these images, we cannot do it, we cannot say whether a given pose or gesture signifies something different from another gesture or pose, whether the presence or absence of birds is an important clue or not, whether the specific activities of the doctors -reading, writing, talking- identify persons and events or are simply variants of standard poses meant only to fascinate the viewer with the talents of the painter. What can, however, be said is that there seems to be a clash between the very Greek names of the ancient doctors, many of whom are known from medical history, and the very contemporary Arab look they were given. But maybe it was not a clash at all in the mind of the artist or in the reaction of the viewer of the time. To them, these images were like a stage setting for the performance of a play. The text of the play is written down in the sections written in small cursive script and it is contemporary figures that play the part of ancient doctors. Just as the writing discussed earlier appears as a form of publicity for the book and its maker, so these actors are part of a visual performance to be seen and appreciated by their own contemporaries, even if they have taken names from ancient history.

Illustration of narratives

There are six illustrations in the manuscript (pp. 17, 19, 21, 30, 31, 33) dealing, in a very loose sense, with the making of antidotes against snake bites. They were all planned carefully as elongated rectangles designed as parts of the composition of six pages, always with a broad band o f Kufic writing above and several small lines of cursive text below. The last two (pp. 30 and 33) occupy a larger share of the page and the very last one, on p. 33, is almost a square. Once again, as he did with the writing, the designer of the manuscript composed each page separately and it is possible that the length of every story was even determined by the space available for it. Five of the stories deal with events connected to snake bites. Even if the images themselves do not really tell a story, the snakes in them are always clearly visible and I suppose that stories could be invented by just looking at pictures which are crowded with many details.

The exception occurs on p. 17 where we see a scene of pharmacists or physicians weighing ingredients for the preparation of some medicine; the composition and its activities will be standard in thirteenth century Dioscorides manuscripts and it is just possible that a visual type, for this topic already existed in the previous century and possibly in Byzantine models. For the other five pictures and to some extent with the one on p. 17 as well, once a story is identified precisely, almost every part of the illustration becomes clear or at least can be provided with an explanation. It is then relatively easy to apply to the analysis of these miniatures standard procedures used for the discussion of texts and their illustration, as has been done in the descriptive pages of this volume.

The painter included as much information as possible into every image and, as a result, the latter looked crowded and tight. On p. 30, two superimposed ground levels were indicated, which allowed for a particularly rich inventory of illustrated details. Elsewhere, the selected elements of a story seem to have been simply pasted on a blank but carefully framed area The compositional pattern or syntax of the images is relatively simple and, for the most part, two dimensional, but the vocabulary used in them is strikingly rich. The variety of types of personages, the range of the clothes they wear or of the activities in which they engage, the many animals (especially horses), the plants, objects, and architectural fragments found on these pages are truly extraordinary and without parallel in the twelfth century. Many of them still pose problems of exact definition and need much investigation. They illustrate a fascinating contrast between very different and almost incompatible modes of representation.

For instance, the schematic and heavy-handed representation of the tray carried by a servant on the upper left comer of p. 30 is in stark contrast with the sophistication of the representation of the four horses which frame the picture on pp. 15 and 31 and which almost foreshadow the horses of fifteenth century Persian miniatures. This is just one example, as every one of the six miniatures illustrates the contrast between extreme sophistication in formal presentation and almost child-like simplicity. Plants or objects and the bare legs of field workers or of personages being bitten are shown crudely, but the agricultural activities on p. 30 or the caricaturized figures of p. 21 are all drawn with a fine mix of humor and observation.

How was all this possible in the waning years of the twelfth century somewhere in the Arabophone world extending from Egypt to Iran? At this stage of our knowledge, there are no first steps in the art preceding the 1199 Kitâb al-Diryâq and, while much in its vocabulary will be found in thirteenth century Arab or Persian painting, its sophistication and complexity recall the Persian painting of that century, especially on ceramics, glass, or metalwork, and even the book illustrations of later times in Iran. Similarly, there is, as far as I know, nothing in the Christian art of the twelfth century, in Byzantium or in the various Christian communities which were then flourishing in the Near East, with clear or even remote parallels to our paintings. We are left then with two possibilities. One, in line with the classical evolutionary doctrine of the history of artistic processes, is that this manuscript implies the existence throughout the twelfth century of artistic developments that would have found their first fully visible expression in its creation.

These preliminaries have disappeared altogether or are still to be discovered in unexplored collections of manuscripts and of fragments of paper or papyrus. Chronologically speaking, and always within an evolutionary theory of the arts, the next stage would lie in the strange frontispieces of the Kitâb al-Aghânî spread between several collections, frontispieces which are compositionally and thematically quite closely related to the Kitâb al-Diryâq. And then there would follow the Vienna manuscript of the same text, several Dioscorides manuscripts, several illustrated scientific and technological codices, and the better known volumes of the Maqamat and of the fables of Kalilah and Dimnah. The Persian codex of Warqah and Gulshah belongs to this series, even though it contains many peculiarities of its own. Seen over a century of changes and evolution, it was a rich and original tradition, but it is rather strange hat its earliest work would have been its visually most completed and most complex one. Perhaps all that is needed to resolve the problem is a large number of searches in the appearance and growth of the components of the miniatures, dozens of details or principles of compositions. But we may also propose an alternate hypothesis, perhaps a more unusual one, but one which seems to fit with some at least of the peculiarities of the book as a whole.

The argument begins with the fact that almost everything about this manuscript is unusual: its size, its many frontispieces, its several repetitions of the names of copyist and owner, its use and manipulation of the newly developed art of calligraphy, its studied assemblage of stories, its arrangement of each page, and its tight images. The conclusion we may draw is that this was not a work to be set within a historical or social development, but that it was the original, maybe even eccentric, creation by an original artist, Muhammad ibn Abi al-Fath Abd al-Wahid, for his uncle, the imam Abu al-Fath Mahmud ibn Jamal al-Din. Nothing is apparently known about these personages, but the book with their names illustrates two features of the Abi al-Fath Abd al- Wahid, for his uncle, the imam Abu al-Fath Mahmud ibn Jamal al-Din. Nothing is apparently known about these personages, but the book with their names illustrates two features of the ways of art in the twelfth century in the Muslim world, ways for which we have other documents.

One is that many objects were made by and for restricted family circles, not, like in the following centuries, for a broad public, possibly for a real art market, or only for the ruling military elite. While increasing our estimation of the sophistication of elite families of the time the ashrafof so many historical sources), the point makes it difficult to assess the actual cultural importance of such works. The sense of my observations suggests that the Kitâb al-Diryâq was made for the private enjoyment of a learned religious and social leader who may not have cared much for the subjects depicted but who enjoyed the rich succession of beautiful pages offering surprise after surprise and who felt that a fancy pharmaceutical book is appropriate for his collection, perhaps because he was interested in 7 popular science, but not necessarily so.

The whole manuscript is and was meant to be a work of art rather than a work of science or of learning. And we may be allowed to fantasize that some of the personages represented are dressed in the manner of members of a family or of a group, one of whom may have been bitten by a snake. The most telling parallel we may have for it is a pen-case in the Hermitage Museum in St. Petersbourg made by a nephew for his uncle for the latter’s return to Herat in Afghanistan from the hajj, the pilgrimage to Mekkah, a trip that must have taken at least a year. The other feature of twelfth century art well illustrated by this manuscript is the wealth of visual material available to a patron and the quality of technical training for an artist, even an amateur. An understanding of the full scope of these models requires lengthy scholarly efforts, but what is important is to realize that sponsor or artist could make choices from a large range of sources and had access to many different ideas about how to design and illustrate a page.

The greater quality of the writing when compared to the painting may well be the result of a longer experience with the practice of calligraphy than with the art of representation. But here once again we are in the realm of hypotheses. Where could a family or a group such as the one outlined here have existed? It has to have been a large city with an administrative and religious tradition, probably connected with the Iranian world, because of so many parallels with later Iranian examples of an urban art like that of ceramics, but also aware of the Byzantine-inspired Christian world which maintained the memory of learning and which could furnish examples of illustrated pseudo-scientific manuscripts. For these reasons, as has been proposed before, a city in the rich and tumultuous zone involving Azerbayjan, Western Iran, the Jazirah, even Iraq or Anatolia, may indeed be a preferred locale for the making and appreciation of this manuscript. But in the mercantile world of the time, when artisans and merchants moved easily and members of the same family settled in different cities of a Muslim cultural world in full expansion, the specific cultural milieu for this manuscript can in fact be imagined in many urban centers from Cairo in Egypt to Heart in Afghanistan.

In collaboration with: