JONATHAN M. BLOOM
THE INTRODUCTION OF PAPER TO THE ISLAMIC LANDS AND
THE DEVELOPMENT OF THE ILLUSTRATED MANUSCRIPT
Scholars have long noted that there was a sudden
explosion in the production of illustrated books in
the Islamic lands during the late twelfth and early
thirteenth centuries, when many different types of
books were illustrated with pictures, apparently for
the first time. In addition to scientific and technical
works, which had long been illustrated with diagrams
and maps, works of poetry and belles-lettres began
to be produced with pictorial frontispieces and illustrations that were not necessary to understand the
text. I
Scholars have offered various explanations for this
phenomenon. Some, adducing pictorial evidence from
the other arts, such as ceramics, have argued that
manuscript illustration had been practiced for a long
time in the Islamic lands, although no illustrated
manuscripts actually survive from early times. 2 Others
have traced this explosion of illustrated books to
external influences, such as artists in the Islamic lands
copying Middle Byzantine or Syrian Jacobite painting, 3 while still others have argued that the appearance of illustrated manuscripts was the result of an
internal development, specifically the emergence of
the bourgeoisie as patrons of this new art form. 4 Oddly
enough, few if any scholars have linked the emergence of the illustrated book to the introduction and
increasing use of paper.5
Paper was introduced to the Islamic lands from
Central Asia in the eighth century and was quickly
adopted for use in government offices.6 The oldest
surviving book on "Arab" or "Islamic" paper is generally thought to be a Greek manuscript of the teachings of the Church fathers (Vat. Gr. 2200), believed
to have been copied in Damascus ca. 800. Apart from
a manuscript in the Alexandria public library recently
discovered by the Israeli scholar Malachi Beit-Arie, 7
the oldest surviving book on paper in Arabic (in
Europe) is a work in Leiden on unusual terms in the
prophetic traditions, which is dated Dhu'l-Qa'da 252
(November-December 867). It bears no indication
of where it was copied. 8
Over the course of the ninth and tenth centuries
the use of paper became increasingly common as the
early Islamic traditions of oral culture were transformed into, although not entirely replaced by, a textbased culture of books. 9 As in many cases, the lead
seems to have been taken in Iraq and Iran, where
paper had been known longest and used in various
contexts and by bureaucrats, who were the first to
use paper in large quantities, although few, if any,
examples of paper documents have survived from the
early period.'" Several dated manuscripts of the Koran copied on paper, presumably in Iran and Iraq,
survive from the tenth century, the most famous of
which is, of course, that copied by the noted calligrapher Ibn al-Bawwab at Baghdad exactly one thousand years ago." In Egypt, over the course of the
tenth century, the manufacture of paper completely
supplanted the 4,000-year-old papyrus industry, and
archaeology confirms what the medieval geographers
say, that papyrus was no longer used in Egypt.12 George
Scanlon's excavations at Fustat showed an overwhelming preponderance of paper over papyrus from the
eleventh-century levels.' 3 Only in North Africa and
Spain, which was known for its production of leather-and hides, did parchment remain the preferred
material for copying manuscripts, particularly the
Koran, but by the year 1000 even in this region paper was being made in significant quantities. Several
Christian manuscripts in the library of the monastery of Burgos, for example, were partly copied on
paper, presumably of Muslim manufacture, as early
as the tenth century.' 4
The tradition of the Islamic illustrated book only
developed in the period after ca. 1000 as a product
of the increased availability of paper in the Islamic
lands. The increased familiarity with and use of paper in the two centuries before the Mongol conquests
18
JONATHAN M. BLOOM
led not only to the burgeoning production of illustrated books but also to important changes in the
ways in which artists and architects in the Islamic lands
thought and worked.
Despite the survival of hundreds of Islamic manuscripts copied before the mid-thirteenth century, only
about three dozen illustrated manuscripts survive from
this period, of which only three date from before the
twelfth century.15 In contrast, for example, to Byzantine illustrated manuscripts, the vast majority of which
are on parchment, all surviving Islamic illustrated
manuscripts were copied on paper. The oldest is 'Abd
al-Rahman al-Sufi's Treatise on the Fixed Stars, dated
1009-10 and now in the Bodleian Library at Oxford.16
The illustrations show the configurations of the stars
in the forty-eight constellations recognized by Ptolemy, but the figures are dressed in Oriental garb rather
than with classical attributes. Al-Sufi wrote that he
knew of another illustrated astronomical treatise, but
he copied his original illustrations directly from images on a celestial globe, and the Oxford manuscript
was copied from the author's original by his son.
The second oldest illustrated manuscript, hitherto ignored by art historians, probably because there
are no people in the pictures, is a 1037 copy of the
geography by the noted mathematician and geographer, Muhammad ibn Musa al-Khwarizmi (or al-Khwarazmi, d. ca. 846). Al-Khwarizmi, a native of the Khwarizm region of Central Asia south of the Aral Sea,
had been attached to the caliph al-Ma'mun's House
of Knowledge in Baghdad and is best known for his
mathematical works. The English word "algorithm"
derives from his epithet, for the first Latin translation of his work on algebra began with the words
"Dixit Algorismi." While al-Khwarizmi's mathematical work has long been known in the West (indeed,
the lost Arabic text has been reconstructed from Latin
translations), his geographical work was also thought
to be entirely lost until a unique manuscript was discovered in Cairo at the end of the nineteenth century. This manuscript, now in Strasbourg, contains four
sketch maps showing the Island of theJewel, the World
Ocean, the Nile, and the Sea of Azov. Scholars are
undecided as to what extent these images reflect alKhwarizmi's original maps.l 7
The third illustrated manuscript from the eleventh
century is an Arabic translation of Dioscorides' De
Materia Medica, dated 1083, in Leiden.' 8 Most of the
pictures in the Dioscorides manuscript depict just
plants; the sole figural miniature demonstrates the
_____1___^______11______·
_
____l______eC______
pharmaceutical use of a plant and seems to have been
modeled on an earlier Greek prototype.' 9 As geographical, scientific and technical works were routinely illustrated in antiquity, it seems logical to assume that
writers and translators in early Islamic times simply
continued the tradition of producing illustrated works.
It is therefore possible that illustrated texts were made
on parchment or papyrus, although none has survived. Considering, however, that Islamic interest in
the ancient sciences only burgeoned in the ninth
century along with the translation movement in Baghdad, itself contemporary with the introduction and
spread of paper, it seems more likely that the development of scientific illustration in the Islamic lands
was itself a function of the increased production and
use of paper.
The scattered references in textual accounts to a
few early illustrated manuscripts, which scholars have
so assiduously collected over the last century, do not
indicate that illustrated manuscripts were common.2 0
Indeed, another way of interpreting these same tantalizing references is that authors noted illustrated
manuscripts precisely because they were so unusual.
When the historian Mas'udi, visited the house of a
notable in the Iranian city of Istakhr in 915, he reported that he saw an illustrated manuscript about
the Sasanian kings and their achievements. The manuscript, supposedly compiled two centuries earlier, in
731, from sources found in the old Sasanian library,
seems to have been copied on leaves of parchment
tinted purple and embellished with twenty-seven "lifelike" portraits of rulers, twenty-five kings and two
queens, painted with gold, silver, and copper. It was
said that a copy of the manuscript had been sent to
the Umayyad Caliph Hisham. 21 Even assuming that
this report is something more than a medieval tour
guide's flight of fancy, there is no reason to believe
that there were many such manuscripts, but there is
also no question that some illustrated manuscripts
were produced in the early centuries of Islam; the
important questions are how many and what kind?
Ibn al-Nadim, the tenth-century author of the Fihrist, a guide to writers and their works, intended his
book to be illustrated, but one could hardly imagine
that these illustrations were "art." In several places
the text is followed by a space for an illustration to
clarify the author's meaning. In one place, for example, he writes, "It is said that this is the form of
the idol that is at Multan [in India]," and in several
late manuscripts this phrase was followed by a blank
___
19
DEVELOPMENT OF THE ILLUSTRATED MANUSCRIPT
space intended for an illustration, although one was
never supplied. 2 2
Quite unusually, a series of illustrations appear to
have been specifically associated with manuscripts of
the animal fables known as Kalila and Dimna, which
were translated into Arabic by Ibn al-Muqaffa' (d. ca.
758-60), although no illustrated Islamic example
survives from before the early thirteenth century.2 3
According to Ibn al-Muqaffa"s introduction,
he who peruses this book should know that its intention is fourfold. Firstly it was put into the mouths of
dumb animals so that light-hearted youths might flock
to read it and that their hearts be captivated by the rare
ruses of the animals. Secondly, it was intended to show
the images of the animals in varieties of paints and
colors, so as to delight the hearts of princes, increase
their pleasure, and also the degree of care which they
would bestow on the work. Thirdly, it was intended that
the book be such that both kings and common folk
should not cease to acquire it; that it might be repeatedly copied and recreated in the course of time, thus
giving work to the painter and copyist. The fourth
purpose of the work concerns the philosophers in particular. 24
Ibn al-Muqaffa', who was of Iranian origin, made his
Arabic translation from a Middle Persian collection
of the tales, itself derived from the popular Indian
fables of the Panchatantra.As Ibn al-Muqaffa"s text
was repeatedly embellished in later times, scholars
are undecided about whether the passage ascribed
to him really belongs to his eighth-century edition
or to some later edition to justify the inclusion of
illustrations.
Leaving that unanswerable question aside, however,
these tales were extraordinarily popular in the Middle Ages, for medieval Hebrew, Latin, and New Persian translations are also known, including a fragment
of an illustrated Greek version on parchment, attributed to southern Italy between 980 and 1050. The
similarities between the illustrations in the Greek fragment to the illustrations in later Arabic and Persian
manuscripts of the text have led Julian Raby to suggest that they all derive from a common, but lost,
manuscript source of the tenth century.2 5 Moreover,
the similarity between the supposed images in the
presumed manuscript source to Central Asian wall
paintings made at the time of the Islamic conquest
of the region, as well as to images molded on ceramics dated to second- or third-century Sri Lanka, suggests that illustrations may have been an integral part
_____
1___
of the Kalila and Dimna manuscript tradition from
pre-Islamic times. Even if some manuscripts of the
Kalila and Dimna fables were illustrated, however, there
is still no reason to assume that manuscripts illustrated with superfluous pictures were common.
The earliest known fragment of an Islamic illustrated book has been dated on paleographic grounds
to the late ninth or early tenth century, but the illustration is only a simple and schematic drawing used
to fill up the empty space at the end of the text page.
Like the unillustrated Thousand and One Nights fragment in Chicago, it was discovered in Egypt, and
consists of a worn fragment of paper, measuring only
16 x 14.5 cm (about 6 inches square) when open
and representing, when folded, the first and last leaves
of a quire.2 6 The text begins on the verso of folio 1
with the standard invocation at the start of a written
text and concludes on the recto of the last folio with
"until death did them part. This is their tomb, may
God have mercy upon them," followed by a crude
painting of two stepped tombs separated by a tree.
There is no way to tell how much text is missing between the first and last pages, although it would have
had to be relatively short to fit within the leaves in
the quire, or whether there were any other illustrations in the booklet. D. S. Rice, who first identified
the text and image on this fragment, suggested that
it belonged to a well-known literary genre of stories
concerning unhappy lovers united only in death. 2 7
In the Fihrist, Ibn al-Nadim classifies this type of story under the genre of night entertainments and fanciful tales, and explains that such works enjoyed great
popularity under the Abbasids in the early tenth century. The schematic simplicity of the image and its
technique, combined with the small size of the text,
confirms that this was a popular work of no great
artistic importance and suggests that the copyist inserted the illustration only to fill up the blank space
on his page.
Many other fragments of paper have been found
in the rubbish heaps of Fustat, largely in the course
of rogue excavations, and some of these are inscribed
with drawings from medieval times which have been
interpreted as evidence for a lost Islamic art of the
book. In general, however, these drawings provide
little or no evidence for any art of book illustration,
because few of them have both text and pictures. For
example, one of the most impressive is a somewhat
large (280 x 180 mm) drawing in the Israel Museum
of a nude and tattooed woman carrying a lute.2 8 The
___11_1_)_1___1______I---
--------
20
JONATHAN M. BLOOM
drawing, on yellowish paper, was first done in red
ink and then gone over in black, with touches of white
and crimson, a technique that can be traced back to
classical times. Some scholars have suggested that this
image of a dancing girl represents a famous courtesan in Fatimid Egypt, while others have suggested that
it represents Venus playing the lute. As there is no
text around or on the back of the image, it seems
unlikely that the page was taken from a book, and its
purpose remains a matter of speculation.
Another well-known fragment from Egypt bears a
calligraphic drawing of a lion on one side accompanied by a few lines of text which have been identified as a discourse on wild animals by the earlyJewish convert to Islam, Ka'b al-Ahbar (d. 652-53); this
would suggest that the drawing came from an early
Islamic book about animals. The drawing, however,
is attributed to the twelfth century. A drawing of a
hare on the reverse, however, is accompanied by a
text irrelevant to the text on the front of the sheet,
so it is difficult to envision how this page could once
have formed part of an illustrated book.2 9 Another
painting claims to be the frontispiece to a collection
of poems by a well-known Umayyad poet, but similar
historical and technical problems raise doubts about
the authenticity of this painting and several other
drawings.3 0
Taken together, this evidence does not amount to
very much, let alone sufficient evidence for the existence of a coherent tradition of manuscript illustration in the first four centuries of Islam. Faced with
this paucity of evidence, some scholars have attempted
to use the images preserved on other papers and in
other media. For example, Ernst Grube interpreted
many of the drawings and paintings found at Fustat
that are not from books as preparatory studies for
painters working on manuscripts or even pottery, ivory
boxes, glass vessels, wooden panels, and the like. Other
drawings are thought to have served the textile weaver
or embroiderer, the bookbinder and metalworker for
the creation of his designs.31 Yet there is no evidence
that wall painters, potters, metalworkers, glassblowers, or weavers used preparatory drawings or sketches on paper at this date. 3 2 The drawings have little
or nothing to do with contemporary representations
on pottery. A potter in ninth-century Iraq or eleventh-century Egypt, for example, would have learned
to draw on ceramics by practicing, not with ink on
paper, but with a brush on unfired ceramics or whatever other surface was readily available. His artistic
_1____1
11
__
1____11____1_11___1_l___il__
repertoire would have been stored in his memory and
maintained by his muscles. The vast majority of his
sketches would therefore have been painted over or
thrown out, leaving only the finished product with
little or no indication of how the artist had worked.
A few ceramics, such as tiles in the British and Metropolitan museums, retain traces of totally unrelated
drawings on their backs, which show that potters
normally practiced their drawing on disposable or
concealed surfaces. 3 3
The freshness of representation and execution in
most early Islamic art, whether on ceramics, metalwork, or textiles, is further evidence for artists' use
of a direct technique, in which they drew or worked
directly on the medium itself. Had potters or metalworkers copied images from another medium, such
as paper, their drawing would have lacked the intensity and individuality that characterize them. No two
pots are exactly-or even nearly-alike, and drawings on early Islamic ceramics are not studied and
repeatedly rehashed but fresh and quite independent
of representations in any other medium. In short,
the men who painted ceramics did not make-or
probably even look at-designs for metalwares or
carved wood, let alone those on paper.
One group of potters, however, probably did use
paper designs from an early date, for the use of paper best explains the group of ceramics inscribed with
Arabic aphorisms in a studied, elegant script and attributed to tenth-century northeastern Iran and Transoxiana, particularly the cities of Nishapur and Samarqand.3 4 Unlike the other pottery attributed to the
region in this period, which was painted directly on
the surface of the dish, this group of inscribed plates
and bowls has clearly been planned out in advance
so that the calligraphy fits the surface exactly. The
inscribed texts include pithy but non-religious proverbs, such as "Learning is at first bitter to the taste
but in the end sweeter than honey" or "Deliberation
from action protects you from regret."
The inscriptions on these ceramics are written in
a studied and deliberate style. The script is closest to
that used on parchment manuscripts of the Koran
attributed to the ninth century. It is entirely different in style from that used on contemporary manuscripts, whether of secular or religious texts.3 5 Considering the high status accorded calligraphers and
the relatively low status of potters, it is unlikely that
calligraphers would have deigned to decorate pots
or known how to use a brush; it is equally improba-
_
DEVELOPMENT OF THE ILLUSTRATED MANUSCRIPT
ble that potters were accomplished in the Koranic
scripts or knew how to use a calligrapher's pen. Rather,
one must imagine that calligraphers prepared designs
which potters then copied onto plates and bowls. Although the script in which these aphorisms are written is closest to that of Koran manuscripts on parchment, it is unlikely that calligraphers prepared their
designs on expensive parchment. Rather, they probably used paper. The several centuries of paper production in Samarqand and neighboring Khurasan
would have made it likely that artisans serving the
wealthy urban population able to afford such fine
wares would have been among the first to explore
the artistic possibilities of this new medium.
The other artistic realm in which paper probably
played an early and important role is textile production, but it is difficult to be certain since our knowledge about how weavers worked in medieval Islamic
times is based largely on extrapolation from surviving fragments, which may not be representative. To
produce the plain textiles or simple geometric patterns such as stripes and checks that formed the great
majority of textiles used in the period, weavers had
no need of cartoons or graphic instructions. Tapestry weavers may have used drawings under or behind
their looms to guide them when inserting colored
threads to make a pattern, but the largely repetitive
nature of many tapestry border designs suggests that
all but novices would have been able to produce many
of the designs from memory. 3 6
Among the most familiar of medieval Islamic textiles are those inscribed with texts naming the ruler
and indicating that they were made on his order for
presentation to courtiers. These so-called tiraz textiles were made from early Islamic times in Central
Asia, Iraq, and Egypt and were an important sign of
sovereignty. The inscribed texts were either embroidered onto the finished fabric or woven into it using
the tapestry technique. The texts, while formulaic,
were prepared by court officials and then copied onto
or into the textile by the embroiderers or weavers,
who were themselves presumably illiterate, to judge
from the mistakes they often made. While one can
imagine that weavers were able to produce the geometric, vegetal and animal designs in the borders from
memory, one must imagine that they needed to copy
inscriptions. The court officials must have provided
the weavers with correct texts on some portable medium. Whether it was papyrus, parchment, or paper
must have depended on cost and availability.
__II______________·____ll___l____
21
The techniques of tapestry weaving and embroidery reveal the artisan's activity as he works; he sees
the results of his decisions and can change them if
he likes. In contrast, in drawloom weaving, which
became increasingly popular for luxury silk textiles
in the eastern Islamic lands around the year 1000, a
pattern is prepared in advance and entered on the
loom prior to weaving, much as a program is installed
on a modern computer. 3 7 Entering the pattern consists of tying the warp threads in bundles so that a
complicated pattern can be woven. The bundles in
themselves, however, do not reveal how the finished
textile will look, just as program files do not show
what a program will do. Once the pattern has been
entered on the loom, the weaver and his assistant,
known as the drawboy, start the process of weaving,
the drawboy selecting bundles of threads to raise and
lower and the weaver inserting the colored wefts to
make the cloth. As long as the weaver and his assistant follow the instructions encoded in the bundles
of cords, the designer's pattern will be reproduced
in the weave. The weaver cannot decide to weave
another pattern on the spur of the moment.
We have absolutely no knowledge about how medieval Islamic weavers encoded their patterns, but
analogies can be drawn from later times, where the
same technique was used. A careful drawing of the
pattern had to be transferred to graph paper, and
the graph then transferred to the simple, a series of
cords, each of which controlled a single warp thread.
The cords were then tied into the bundles. The process of entering the pattern into the loom could take
a month or more, and progress in weaving might vary
from a few centimeters to a meter per day, depending on the fineness of the pattern and number of
colors used. This complexity explains why this elaborate technique was only used for the finest silks, such
as the so-called Shroud of Saint-Josse, inscribed in
Arabic with the name and titles of a Turkish commander active in northeastern Iran who was executed on the orders of his Samanid sovereign in 961.38
It is inconceivable that this splendid textile was prepared without using a preliminary drawing to set up
the loom. Although we cannot know on what material this drawing was made, it is quite probable, considering when and where the textile was made, that
it was paper.
The increased, but still rather limited, role of paper in the production of some of the arts in the eastern Islamic lands in the period before the year 1000
___
22
JONATHAN M. BLOOM
suggests that the development of the illustrated book
as an art form was very much a product of the period after ca. 1150. Paper's smooth surface allowed for
new types of scripts; paper's absorbency demanded
new types of ink, and probably new ways of preparing pigments and binders. While there can be no
question that some illustrated books were produced
in the earlier period, they seem to have been relatively uncommon. The increased number of illustrated
manuscripts that have survived from the twelfth century suggests that by then they were being produced
in greater quantities. This supposition is confirmed
not only by textual references to the production of
manuscripts but by the increased use of paper in the
production of ceramics and the other portable arts,
as well as architecture and its decoration.
Richmond, New Hampshire
NOTES
1.
2.
3.
4.
5.
6.
7.
1_1
I am using the word "illustration" to refer only to pictures
added to the text. "Illumination" is non-representational
decoration added to the text. Illuminated-as opposed to
illustrated-frontispieces and headings are frequently found
on parchment manuscripts of the Koran that may date from
as early as the ninth century c.E.
The literature on this subject is vast. See, for example,
Richard Ettinghausen, "Painting in the Fatimid Period: A
Reconstruction," Ars Islamica 9 (1942): 112-24; Ernst J.
Grube, "Fustat Fragments," in Islamic Paintingand the Arts
of the Book, The Keir Collection, ed. B. W. Robinson (London:
Faber and Faber, 1976), pp. 23-66.
Kurt Weitzmann, "The Greek Sources of Islamic Scientific
Illustrations," in Studies in Classicaland Byzantine Manuscript
Illumination, ed. Herbert L. Kessler (Chicago: University of
Chicago Press, 1971), pp. 20-44.
Oleg Grabar, "The Illustrated Maqamat of the Thirteenth
Century: The Bourgeoisie and the Arts," in The Islamic City,
ed. A. H. Hourani and S. M. Stern (Oxford: Bruno Cassirer,
1970), pp. 207-22; idem, The Illustrations of the Maqamat
(Chicago: University of Chicago Press, 1984).
For a summary of the question, see Julian Raby, "Between
Sogdia and the Mamluks: A Note on the Earliest Illustrations to Kalila wa Dimna," OrientalArt 33, 4 (Winter 198788): 381-98. The introduction of paper was mentioned by
Eva R. Hoffman, "The Emergence of Illustration in Arabic
Manuscripts: Classical Legacy and Islamic Transformation,"
Ph.D. diss., Harvard University, 1982.
In anticipation of the publication of my book, PaperBefore
Print: The Introduction ofPaperin the Islamic Lands (New Haven:
Yale University Press, 2001), seeJonathan M. Bloom, "Revolution by the Ream," Aramco World 50,3 (May-June 1999):
26-39.
Malachi Beit-Arih, "The Oriental Arabic Paper," Gazette du
_
_I
_I
livre mediival 28 (Spring 1996): 9-12.
8. About two-fifths of the text is missing. See P. M. Voorhoeve,
Handlist of Arabic Manuscripts (Leiden, 1957), Codices
Manuscripti VII, 2nd ed. (The Hague, 1980), before p. 1. It
is dated in the colophon on fol. 241b as Dhu'l-Qa'da 252
(Nov.-Dec. 866). Voorhoeve is very cautious: "Apparently
the earliest dated paper codex in Europe"; see also D. A.
Felix, "What Is the Oldest Dated Paper in Europe?" Papiergeschichte 2, 6 (December 1952): 73-75; and idem, Levinus
Warnerand His Legacy: Three Centuries of Legatum Warnerianum
in the Leiden University Library (Leiden: E.J. Brill, 1970), pp.
75-76. The script of the colophon is not representative, however, of the script in the codex. A more representative text
page was reproduced in the exhibition catalogue, Levinus
Warner and His Legacy, last illustration. I am most grateful
to J. J. Wiktam for supplying this information.
9. The shift in Islamic civilization from an oral to a text-based
culture has not been fully explored. In the meantime, see
William A. Graham, Beyond the Written Word: OralAspects of
Scripture in the History of Religion ( 1987; rpt. New York: Cambridge University Press, 1993), and for a parallel case in
medieval England, M. T. Clanchy, From Memory to Written
Record: England 1066-1307, 2d ed. (1979; rpt. Oxford: B. H.
Blackwell, 1993).
10. Apart from those papyrus (and paper) documents discovered at such Egyptian sites as Akhmim, Ashmunayn, and
Aphrodito, the oldest actual documents on paper to have
survived are the Fatimid decrees preserved at Mt. Sinai, for
which see S. M. Stern, Fgtimid Decrees: OriginalDocuments from
the Fatimid Chancery, All Souls Studies (London: Faber and
Faber, 1964).
11. D. S. Rice, The Unique Ibn al-Bawwab Manuscript in the Chester
Beatty Library (Dublin: Chester Beatty Library, 1955).
12. Geoffrey Khan, Bills, Letters and Deeds: Arabic Papyri of the 7th
to ilth Centuries, The Nasser D. Khalili Collection of Islamic
Art, ed. Julian Raby, vol. 6 (London: The Nour Foundation
in association with Azimuth Editions and Oxford University
Press, 1993).
13. Wladyslaw Kubiak and George T. Scanlon, FustatExpedition
FinalReport, vol. 2: Fustet-C (Winona Lake, Ind.: Eisenbrauns
for.the American Research Center in Egypt, Inc., 1989).
14. Oriol Valls i Subirk, Paper and Watermarks in Catalonia,
Monumenta Chartae Papyracea Historiam Illustrantia
(Amsterdam: Paper Publications Society [Labarre Foundation], 1970), pp. 5, 9.
15. Kurt Holter, "Die islamischen Miniaturhandschriften vor
1350," ZentralblattflrBibliothekwesen 54 (1937): 1-34; Hugo
Buchthal, Otto Kurz, and Richard Ettinghausen, "Supplementary Notes to K. Holter's Check List of Illuminated Islamic Manuscripts before A.D. 1350," Ars Islamica 7 (1940):
147-64. I am deliberately neglecting the fragmentary double
frontispiece to a magnificent parchment manuscript of the
Koran discovered in the Great Mosque of San'a (for which
see Hans-Caspar Graf von Bothmer, "Friihislamische KoranIlluminationen: Meisterwerke aus den Handschriften der
GroBen Moschee in Sanaa/Yemen," Kunst & Antiquititen,
no. 1 [1986]: 22-23; idem, "Architekturbilder im Koran: Eine
Prachthandschrift der Umayyadenzeit aus dem Yemen,"
Bruckmanns Pantheon 45 [1987]: 4-20; idem, "Meisterwerke
Islamischer Buchkunst: Koranische Kalligraphie und Illumi-
1
_11_1_____
DEVELOPMENT OF THE ILLUSTRATED MANUSCRIPT
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
nation in Handschriften aus der GroBen Moschee von
Sanaa," inJemen, ed. Werner Daum [Innsbruck-Frankfurt/
Main: Pinguin-Verlag, Umshau-Verlag, 1987], pp. 177-80),
because there is no scientific proof for von Bothmer's claim
that the manuscript has been carbon dated to the Umayyad
period, and a ninth-century date seems more likely on the
basis of the script.
Emmy Wellesz, "An Early al-Sufi Manuscript in the Bodleian
Library in Oxford: A Study in Islamic Constellation Images,"
Ars Orientalis3 (1959): 1-27.
For this manuscript, see Gerald R. Tibbetts, "The Beginnings
of a Cartographic Tradition," in Cartography in the Traditional
Islamic and South Asian Societies, ed. J. B. Harley and David
Woodward, The History of Cartography, vol. 2 (Chicago: University of Chicago Press, 1992), pp. 90-107.
Leiden, University Library, Cod. Ar. 289 Warn. See
Weitzmann, "Greek Sources," p. 252, pl XXXIV, fig. 8; Ernst
J. Grube, "Materiellen zum Dioskorides Arabicus," in Aus
der Welt der Islamischen Kunst: FestschriftfiirErnst Kfihnel zum
75. Geburtstag, ed. R. Ettinghausen (Berlin: Mann, 1959), pp.
169, 175, fig. 5.
Weitzmann, "Greek Sources," fig. 11.
See, for example, T. W. Arnold, Paintingin Islam (Oxford:
Clarendon Press, 1928).
Mas'udi, Kitdb al-tanbth wa'l-ishraf (Cairo, 1938), pp. 92 ff.,
quoted in D. S. Rice, "The Oldest Illustrated Arabic Manuscript," Bulletin of the School of Oriental and African Studies 22
(1959): 207-20.
Al-Nadim, The Fihrist of al-Nadim: A Tenth-Century Survey of
Muslim Culture, ed. and trans. Bayard Dodge (New York and
London: Columbia University Press, 1970), p. 832.
ErnstJ. Grube, "Prolegomena for a Corpus Publication of
Illustrated Kalila wa Dimna Manuscripts," Islamic Art 4 (199091): 301-482.
Quoted in Rice, "The Oldest Illustrated Arabic Manuscript,"
p. 209. There is some doubt as to whether this passage was
written by Ibn al-Muqaffa' or by a later editor.
Raby, "Between Sogdia and the Mamluks."
Rice, "The Oldest Illustrated Arabic Manuscript."
Ibid.
Jerusalem, Israel Museum, M 165-4-65. Rachel Milstein and
N. Brosch, Islamic Paintingin the Israel Museum Jerusalem,
1984), p. 23.
See Stefano Carboni in Trisorsfatimides du Caire, exhibition
catalogue, 28 April-30 August 1998 (Paris: Institut du Monde
Arabe, 1998), p. 99. As dealers have been known to "embellish" or "improve" some genuine medieval pages and
30.
31.
32.
33.
34.
35.
36.
37.
38.
23
drawings to increase their worth on the art market by providing them with identifying texts or illustrations, it would
be hazardous to use this leaf as evidence for early book illustration.
See Grube, "Fustat Fragments," p. 33; and Gaston Wiet, "Une
peinture du XIIe siecle," Bulletin de l'Institut d'Egypte 26
(1944): 109-18. Another well-known drawing of two warriors
in Cairo (Museum of Islamic Art, Inv. 13703; see Trisors
fatimides du Caire, cat. no. 22) is equally suspicious, although
it is repeatedly reproduced in books on Islamic art. The
truncated inscription, which reads "Power and prosperity
to the commander Abi Mans . . " was never meant to be
complete-an extraordinary and unexplained anomaly.
Virtually all of these works appeared on the art market at
the same time as the "Buyid" silks, now definitively shown
to be modern forgeries. See Sheila S. Blair, Jonathan M.
Bloom, and Anne E. Wardwell, "Reevaluating the Date of
the 'Buyid' Silks by Epigraphic and Radiocarbon Analysis,"
Ars Orientalis 23 (1993): 1-42. The drawings, like the textiles, were "authenticated" by the same unsuspecting scholars in the same way using the same types of evidence. Technical reexamination of these drawings is essential.
Grube, "Fustat Fragments," pp. 26-27.
A convenient selection of irhages can be found in Richard
Ettinghausen and Oleg Grabar, The Art and Architecture of
Islam: 650-1250 (Harmondsworth: Penguin, 1987).
Venetia Porter, Islamic Tiles (New York: Interlink Books,
1995), pp. 46-47.
Lisa Volov (Golombek), "Plaited Kufic on Samanid Epigraphic Pottery," Ars Orientalis 6 (1966): 107-34.
See, for example, manuscripts illustrated in Francois
D6roche, The Abbasid Tradition:Qur'ans of the 8th to the 10th
Centuries A.D., The Nasser D. Khalili Collection of Islamic Art,
vol. 1, ed. Julian Raby (London: The Nour Foundation in
association with Azimuth Editions and Oxford University
Press, 1992).
For a convenient introduction, see Patricia L. Baker, Islamic
Textiles (London: British Museum Press, 1995).
On the technique of drawloom weaving, see The Dictionary
of Art, s.v. "Textile," §II, (ii) (a) Draw loom; and Hans E.
Wulff, The TraditionalCrafts of Persia: TheirDevelopment, Technology, and Influence on Eastern and Western Civilizations (Cambridge, Mass.: M.I.T. Press, 1966), pp. 205-9.
M. Bernus, H. Marchal; and G. Vial, "Le suaire de SaintJosse," Bulletin de Liaison du Centre Internationald'Etudes des
Textiles anciens 33 (1971): 1-57.