Mr Katsuhiko Abe
Ueno Gakuen University, Faculty of International Cultural Studies, Tokyo, Japan

Safavid Textiles in Kire-Tekagami, Japanese Textile Albums

Iranian textiles began arriving in Japan more than one thousand years ago and they were appreciated as prestigious “objets d’art” among the elite. During the Edo Period, many Safavid textiles were brought by the Dutch, and were collected by Daimyo (feudal lords) as well as by many connoisseurs connected with the Tea Ceremony. Textile fragments of various origins were assembled in the form of Kire-Tekagami (Textile Albums), which were in effect books of samples. Safavid textiles were called “MOURU”, a word sometimes alleged to have been derived from the word “Moghol” (Mughal Empire). However, the term has been used indifferently for Iranian or Indian, and as well as for European textiles. In addition to the difficulty of identifying whether particular samples are Safavid or Mughal, the question of collector’s tastes and the aesthetic and historical values on which their systems of classification were based has yet to be explored, especially where Safavid textiles are concerned. This paper will examine the significance of Safavid samples in the compilation of these albums of exotic textiles as well as the aesthetic sensibilities reflected in their selection and appreciation amongst the connoisseurs of 17^th  and 18^th  century Japan.

Dr Mary Anderson McWilliams
Arthur M. Sackler Museum, University of Harvard, Cambridge MA, USA

A Persian Velvet for a Boston Brahmin

In 1895, early in her collecting career, Isabella Stewart Gardner purchased a fragment of a superb Safavid velvet from an art dealer in Rome.  Its current location, set in a faux-Renaissance frame, gracing the Titian Room of her Venetian palazzo in Fenway Park exemplifies the eclecticism that inspired Mrs. Gardner's purchase and exhibition of a sprinkling of Islamic works of art.  

The Gardner velvet, an excellent example of Sir John Chardin's "gold velvets", appears to be the unique surviving fragment of this pattern.  This lecture will begin with an historiographic examination of Safavid textiles at the turn of the 19^th  century, and conclude by placing the Gardner velvet in the context of our understanding of the field a century later.

Ms Julia Bailey
Aga Khan Program, University of Harvard, Cambridge MA, USA

Carpets and "Kufesque"

I will consider a border element that appears on carpets depicted in the 14^th century Great Mongol Shahnama, and that is also characteristic of later carpets, depicted or real, from Anatolia and Timurid and Safavid Iran. Richard Ettinghausen maintained that this so-called "kufesque" element was pseudo-epigraphic and derived from a stylization of "Allah." Ettinghausen’s argument has often drawn the objection that, however abstract, allusion to the deity on rugs trodden underfoot is unlikely. Herwig Bartels has suggested that the element in question is not even epigraphic in origin. I will argue that the kufesque border design is indeed script-derived, and that in illustrations from the Great Mongol Shahnama, where the appearance and use of textiles and objects at the Ilkhanid court are represented with documentary accuracy, a particular Arabic formula serves as the legible equivalent of kufesque in images of enthronement. Not restricted to the Great Mongol Shahnama, this formula is depicted in later manuscript painting as a "kufesque variant" carpet  border design. A proclamation of royal dominion, it is a more plausible source than is "Allah" for kufesque itself.  

Dr Patricia L. Baker
Independent scholar, London, UK

"Wrought of Gold or Silver": Honorific Garments in 17th Century Iran

The principles regarding the presentation of khalat have been summarised by various modern scholars utilising court chronicles and reports of non-Iranian visitors to the Safavid court. This paper examines the pictorial evidence alongside this data, of such portraits as the well-known Van Dyck painting of Sir Robert Sherley (1581-1628) now at Petworth, Sussex but there are a number of other, lesser known works all painted in the first three-quarters of the 17th century, depicting both foreigners and Iranian nationals who were recognised for their services by the shah. While these can only add to our knowledge of the preferred textiles employed for such khalat garments, the opportunity will be taken to relate the fabrics so depicted to surviving examples of contemporaneous Safavid woven stuffs and the evidence, recorded in various sources, diplomatic, commercial and administrative, concerning the differing qualities of silk filaments, the fabrics, the dye colours and the metallic thread. In doing this, it will seek to persuade students of Islamic textiles of the usefulness of even the basic structural analysis rather than relying on the established pattern of identifying and determining production centres solely on aesthetic criteria.  

Dr René Bekius
Institute of Social Studies, The Hague, The Netherlands

Competition or Cooperation: New Archival Sources about Armenian Textile Traders and Dutch Trading Companies in the Safavid and Ottoman Empires

This article traces the relations between Armenians and the Dutch East Indies Company in the Safavid empire and Dutch Levant Company in the Ottoman empire. In addition, an attempt will be made to quantify how many Amsterdam Armenian merchants were active in the import and export of Iranian, Ottoman and Dutch textiles. The argument is based on 1200 pieces of Armeniaca in the Amsterdam notarial archives. The main research question is to what extent Armenian merchants competed against or cooperated with the Dutch companies, with reference to raw and processed silk, tapestries, wool, cloth and carpets? Through which routes did the textile trade take place?  To what extent did the Dutch get their knowledge of local supply from Armenians?

Dr Beata Biedronska-Slota
The National Museum of Krakow, Poland

Persian Sashes preserved in Polish Collections

The majority of the Persian sashes (12 sashes and one liturgical cape made from a Persian silk sash) preserved in Poland are attributed to Kashan.  Three of them are slightly different in technique and design to the others. For reasons which will be explained, these were probably made in India at the end of the 17th century.

An archival document preserved in Poland makes frequent mention of the Persian sashes, but there is no evidence for Persian weavers in Poland. Whereas, the existence of Armenian weavers in Istanbul and of Armenian merchants in Poland is well attested. I will discuss how these sash workshops were established in Poland towards the end of the 18th century and how these sashes were produced by Armenian weavers in Poland, after Persian designs.

Dr Steven Cohen
Independent scholar, London, UK

Parallels Between the Manufacture of Luxury Textiles from Kork, Fine Kirmani Goat Hair, and Pashmina, Indian 'Kashmir' Goat Hair

Both the Iranian Safavids and the Indian Mughals highly valued the short, fine underhair of certain domesticated goats. However, when one considers the three main uses of this material: knotted pile carpets, twill tapestry shawls, and felt rugs, it is obvious that the craftsmen of the two Empires used fine goat hair (kork in Iran and pashm in India) in different ways at different times. In the 16th century Iranians in Khorasan were probably the first to use fine goat hair rather than silk, for the pile of their finest, most luxurious knotted-pile carpets. But what was most likely a strictly regional practice ended long before Indian carpet weavers adopted the same habit in the middle of the 17th century to produce the world's most densely knotted classical carpets. Therefore, it is India, rather than Iran, which has become so closely associated with the production of pashmina piled carpets.

Kashmir in the north-west of India has probably been producing twill tapestry woven shawls of imported Tibetan pashm since at least the 11th century, but the manufacture of Kashmiri pashmina shawls, possibly in more simple structures, had undoubtedly existed there long before that date. From at least the 17th century, Iran also manufactured shawls using local kork from Kirman, but since no early examples have survived and we have absolutely no idea of their structures or designs, it remains impossible to compare any pre-19th century Kirmani kork shawls with their Kashmiri contemporaries. However, to merely read about the continuing massive import of pashmina shawls from Kashmir to Iran throughout the 18th and 19th centuries (with no reciprocal export from Iran to India), and the occasional attempts by various Iranian governments to reduce these Indian imports, and to learn about the efforts of Iranian weavers to imitate Indian shawls, one must conclude that Kirmani kork shawls were probably always inferior to the Kashmir originals in both technique, material and design.

Finally, we come to the most profligate use of this valuable material: takiyeh namads, felt carpets. This was possibly a strictly Iranian practice because although we have a great deal of documentation proving that often the greatest use of Kirmani kork was for the manufacture of soft felted namads, no such use of pashm has yet been discovered in any Mughal records. We know that Shah Tahmasp ordered goat hair felts to be presented to the exiled Mughal Emperor Humayun in 1544 and that extremely luxurious, inscribed felts, which one assumes are the famous takiyeh namads, quite often appear in both Safavid and early Mughal miniature paintings so they must have been known and highly valued in both Iran and India. The Mughal Emperor Akbar's biographer also wrote in around 1590 that takiyeh namads were both imported into India from Kabul and Wilayat (probably Herat) as well as being manufactured in India itself, but it does not state that they were made of kork or pashmina, so that was probably never an Indian habit.

Dr Yolande Crowe
Independent scholar, Geneva, Switzerland

The Importance of a 1243 Chinese Tomb and Other Dated Documents During the Pax Mongolica

Recent exhibition and auction catalogues have shown marvellous fragments of Chinese textiles which have been attributed to the Song and Yuan periods without precise dating. With the garments found in the Chinese tomb of 1243 it is possible to relate some of the embroideries  to  motifs reproduced for instance on the tiles of the twin octagonal halls of the Takht-i Sulayman. These buildings have been dated by lustre tiles to 1271-3 and 1275-6. Furthermore manuscripts produced in the Armenian kingdom of Cilicia  contain illustrations showing garments with Chinese dragons and tables with spandrels decorated with Chinese phoenixes. The manuscripts are dated to the 1280s. Other manuscripts such as those of dated World histories of Rashid al-Din and Shahnamas also help to give a more precise picture of early copies of late Song Chinese motifs. These motifs will continue to appear in often simplified forms on textiles and carpets up to this day.

Professor Walter B. Denny
University of Massachusetts at Amherst, MA, USA

Anatolia, Tabriz, and the Carpet Design Revolution

The tendency of carpet scholarship to deal with Anatolian and Iranian carpet traditions as separate art-historical phenomena, while both convenient and often practical as well, may occasionally tend to obscure vital connections between the two.  The prevailing conception of what Kurt Erdmann termed the Carpet Design Revolution of the later 15th century, which views Iranian culture as a unified continuum from Türkmen Tabriz to Timurid Herat, is a case in point.

In this paper we will examine a number of such assumptions made about carpets of the late 15th and the early 16th century.  These include not only the identification of the Carpet Design Revolution with Herat, but the attribution of many of the great early Persian medallion carpets either to north-west Persia or to Tabriz itself.  It will be argued that the preponderance of evidence available today points to a western (Tabriz-Ushak) origin of the so-called Design Revolution.  It will be further argued that Tabriz under the Kara Koyunlu and the Ak Koyunlu, which for a brief period enjoyed great prosperity from the silk trade, may have been the site of a limited court production of carpets today termed "para-Mamluk." It will be further argued that many of the "north-west Persian" medallion carpets were probably not made in Tabriz, and that the two "Ardebil" carpets, while exhibiting a design that undoubtedly originated in Tabriz, were almost certainly woven elsewhere, probably in Kashan.

Finally, the paper will explore the Wright-Wertime hypothesis that the so-called "dragon" carpets were woven in Tabriz -- as opposed to Heriz, whose production  of a hundred years ago far more closely recalls the dragon carpet tradition than that of Tabriz -- and will examine the historical identification of the symmetrical knot with Tabriz weaving traditions,

Dr Willem Floor
Independent scholar, Bethesda MD, USA

Import of Indian Textiles into 17th Century Persia

Many of the textiles used in Safavid Persia were not woven in-country, but were imported. Most of the imports of textiles into Safavid Persia came from India, which had been a supplier of textiles to the Persian Gulf markets before the establishment of the Safavid kingdom, and it would continue to do so well into the 20th century. However, India is a big country, it is even a subcontinent, and therefore many production centres supplied the Persian Gulf markets. Also, the relative importance of each production centre with regards to exports to Persia changed over time. Furthermore, there was a seemingly endless variety of textiles that were produced, of which we sometimes do not know more about than their name. It is impossible to give an answer to all these questions, but an attempt will be made to analyze the nature and composition of the import trade of textiles from India into Safavid Persia in particular and the Persian Gulf in general. Also, evidence will be offered to show that textiles that look as if they are Safavid Persian are not necessarily Persian, but are in fact of Indian origin.

Dr Jessica Hallett
Independent scholar, Lisbon, Portugal

"From The Looms of Yazd and Isfahan":  The Persian Carpet In Portugal

Knotted oriental carpets had long been a luxury in western Europe, accessible only to the wealthiest few, but in the 16th century this situation changed with the Portuguese discovery of the direct sea route to the East. Persian and Indian carpets began entering Atlantic ports, and especially Portugal, in ever increasing numbers until changes in interior decoration in the 18^th century led to a decline in their popularity.

The esteemed status of the oriental carpet as an object of value and admiration is still visible in Portugal today, and a surprisingly large number of carpets (over 100) survive in national collections. Perhaps not surprisingly more than 80% of these carpets are of "Persian" or "Indo-Persian" type. This material evidence is further corroborated by some forty representations of Persian carpets in Portuguese painting and by archival data concerning their provenance, chronology, patronage, status and function.

The existence, size and importance of these collections has received only minor attention, especially outside of Portugal, and the aim of this talk is to provide an introduction to the range, diversity and potential of this material for enhancing our understanding of the Persian and Indo-Persian carpet.

Mrs Christine Klose
Independent scholar, Karlsruhe, Germany

Traces of Timurid Carpets in Contemporary and Later Carpets from the Near East

Carpets from Timurid times are preserved in a few surviving pieces and are documented in some 1500 Persian miniature paintings, which were painstakingly analyzed by Amy Briggs in the 1940s.  The two most important groups of such carpets, called 'Lattice Carpets' and 'Cartouche Carpets', will be presented in the discussion.

1. 'Lattice Carpets' appear in miniatures as early as the beginning of the 15th century. These carpets are characterised by a rectangular lattice design, occasionally also by a lozenge lattice. Only one fragmentary Persian example is preserved, in the Benaki Museum in Athens. The fragment has asymmetric Persian knots and a lozenge pattern. Early pieces from Anatolia with symmetric knots are more frequent. Among these the famous so-called 'Compartment Carpet' in the Vakıflar Museum in Istanbul, which shows some East-Anatolian and Mamluk characteristics, is of special interest. I shall present two further 'Lattice Carpets', one from Sion, in Switzerland, and one from the London art-market, which are clearly West-Anatolian pieces.  I will discuss the relationships between these carpets and the early Ottoman and 'Damascus' carpet groups.

2. 'Cartouche Carpets' appear in miniatures towards the second half of the 15th century. They are characterized by lobed cartouches which overlap in a complex manner. There is no common ground colour. I will present a fragment from a German private collection with overlapping cartouches in the field and border and discuss fragments dating from about 1500 with separated cartouches. We know several pieces from the early Safavid period with separated cartouches in the field pattern, among them the famous pair of 'Cartouche Carpets' split between Lyon and New York, and the 'Medallion Cartouche Carpets'.

As regards border designs, in some early Safavid pieces we see patterns with overlapping cartouches. I will trace their development to the later 'Three-Cartouche' borders with separated cartouches. From this it becomes evident that the special form of shield cartouches comes from overlapping patterns.

In the 17th century overlapping cartouches were again used in field patterns. Three examples are shown from the Metropolitan Museum in New York. Also from the 17th century, we have one carpet in Vienna (MAK) and four silk carpets from different collections, all with separated cartouches.  Even in the 18th century traces of the Timurid cartouche patterns may be found in the 'Shield and Tree' carpets from Khorasan.

In my opinion the relationship between Timurid cartouche patterns and the 'Star Ushak' family is only superficial. I would like to group these Anatolian carpets together with other 'Medallion' carpets.

Ms Jennifer Scarce
University of Dundee, UK

The Textiles of 16th Century Safavid Dress – Their Interpretation in the Illustrations of Contemporary Manuscripts

Three groups of sources provide information about this theme – the textiles themselves, the elegant contemporary fashions of the characters of illustrated manuscripts and references in Persian texts to production centres and in European descriptions of  clothing.

These sources do present problems.  Few securely dated 16th century textiles and   garments have survived.   It is often difficult to collate fabrics shown in the paintings with actual textiles; for example textiles decorated with figure motifs do not seem to feature in paintings. Persian  sources may list textiles but do not describe them, while European accounts depend on access and the writer’s skills of observation and description.

It is possible, however, to attempt an identification of the textiles by applying certain criteria.

1. The conventions of manuscript illustration which include a repertoire of motifs allocated to specific textiles i.e. silks and velvets.

2. Changes in Safavid fashion which affect representation in painting.  It is easier to paint large-scale designs, including those with figural motifs, which appear on the heavy coats worn as cloaks during the 17th century than on the layers of close-fitting garments of the 16th century.

3. A practical knowledge of the qualities of fabrics and methods of garment-construction.  16th century paintings include a lot of information about folds, creases and seams which is essential to textile identification.

Using these criteria I aim to analyse the sources to produce an introductory classification of the principal textiles of 16th century dress including a suggested link between figural designs and painting.

Dr Jon Thompson
Ashmolean Museum, University of Oxford, UK

Timurid Carpets, a Reappraisal

The principal source of information on Timurid carpets is miniature painting. The general picture has been well documented. In the 14^th and early 15^th century we see carpets with small-scale endless repeat patterns and interlaced kufesque borders with ‘mitred’ corners. In mid-15^th century paintings this style gives way to centralised designs, arabesques and floral motifs both in the field and the borders, though kufesque borders persist. Another design with overlapping compartments appears in the last third of the 15^th  century, mainly in the paintings of Bihzad. In Persian carpets this style persists into the 16^th century and later, mainly as a border design, but is unknown in Turkish carpets.

The carpets that correspond most closely in appearance to those featured in miniature paintings of the 15th century are Turkish carpets, which have survived in fair numbers as a result of trade with the west and gifts to mosques. These 15^th century survivors generally do not have the character of court carpets, so in view of the virtual absence of Persian carpets of this type their relationship to the carpets depicted in the courtly art of Herat and Shiraz is a puzzle. It is hard to imagine Turkish village rugs being used at the Timurid court.