Ashmolean Eastern Art Online, Yousef Jameel Centre for Islamic and Asian Art

Textiles are produced to fulfil two primary functions: to clothe the body and to provide furnishing material for the human habitat. As far as we can determine from the fragmentary state of the material in the Newberry Collection, it was imported into Egypt for both purposes. This is confirmed by the numerous remains of seams and stitches on the fragments. A total of 548 textiles have stitching; of these a relatively high percentage have sewing done with flax: 322 are definitely flax, and in addition 32 are almost certainly flax fibre [1]. This fibre was rarely used for textile purposes in India, but was common in Egypt. We can therefore assume that the fabrics in general were exported from India as long, uncut cloths, the so-called piece-goods. Complete piece-goods survive from the Gujarati trade with South-East Asia, and to some degree these can give us an impression of the likely original size of the Egyptian goods as well, as often the cloths were of a similar type. This is borne out not only by many related patterns, but by a similar quality of cloth. Even the current production of block-printed fabrics, in particular of ajrakh textiles in Gujarat and Sindh (Pakistan), still use a cloth size and general layout that is similar to the one employed in the seventeenth century. Hypothetically we may imagine this standard of production to go back further, by at least several hundred years [2].

The quality of the cotton generally made it suitable for both dress material and furnishings, as it was not too fine to be effective as a curtain or hanging, and not too heavy to provide good fabric for clothing. The remains of stitching often indicate the specific use. For example, [EA1990.718] has two seams that move at an angle, as is used when sewing the gusset of a garment, and at right angles to one of the seams a third seam could be the yoke [EA1990.718]. Similarly, [EA1990.404] clearly shows that the fragment was used as a child’s tunic. Interestingly, the same textile also has evidence that its original purpose had been different, and that it probably was first used as a larger garment. This can be seen in the presence of seams that do not make structural sense for the shirt as we now see it, as well as in the different types of stitching. The original seams are very finely sewn, while the cut-down version is much less carefully executed. It shows that the textiles were commonly used and re-used, as is still done in many families and communities nowadays in most parts of the world, although it may have become uncommon in highly industrialized societies. Many of the textiles also have evidence of mending, sometimes with different patches sewn on. Considering the extensive use the fabrics had, it is quite remarkable how well preserved many of the designs are. There is of course evidence of much fading and discoloration, but the fact that we can still see as much subtlety and distinction of colour is to the credit of the craftsmen’s skill.

In all collections of Indo-Egyptian textiles there are fragments or—occasionally—complete cloth panels that have tabs sewn to their borders. These textiles were probably once used as part of domestic furnishings, as room-dividers or door hangings, as one still finds commonly in use in Asian and Near Eastern houses. Alternatively they could also have been part of large canopies used outdoors at ceremonial functions, such as weddings [3]. Sometimes the tabs were sewn from parts of the fabric that made up the main body, but most frequently they were actually printed separately as tab shapes on to the fabric and could then be sewn up. Several examples that have not yet been sewn indicate that the tabs were printed, but not completed in their final shape when they were sold [4]. The same fragments also show that the tabs were block-printed as the borders of large textiles, rather than produced separately.

The tabs are examples where we can already tell from the design of the fabric that the purpose of use was not to be dress, but furnishing for indoor or outdoor hangings. Similarly, if less obviously, certain designs would be more suitable for one purpose than another. The very large medallions, such as [EA1990.219] or [EA1990.243], would hardly be used for tailored dress, but for domestic decoration. If worn, however, they were more likely to be used as wraps to cover the whole body, as similar cloths are still used by women in the Yemen. On the other hand, the small-scale, repetitive design of [EA1990.9] and following, or [EA1990.277] and following, can easily be imagined sewn into a garment. The seams that are frequently found in these fragments were probably stitched to shape the fabric into a shirt or dress.

Fatimid and Mamluk dress

As the historical evidence suggests that the Indo-Egyptian textile trade was well established by the twelfth century, it may be helpful to investigate the appearance and social function of both Fatimid and Mamluk dress. This gives us the context in which to visualize the use of the Indian cloth. At the outset, though, a particular problem has to be recognized: we are addressing a topic that has only recently become the focus of historical scholarship, and in many respects the study of Islamic dress in medieval times is still in its infancy [5]. In 1845 Reinhart Dozy published a monumental dictionary of terms referring to garments found throughout the Arab world, but for nearly a century afterwards it remained the primary source, as the topic of Islamic dress was not further investigated (Dozy 1845). Serjeant’s pioneering study on Islamic textile production also inevitably touched on aspects of dress, but detailed considerations of the topic were first undertaken by Mayer in his publication on Mamluk costume (1952). Goitein stressed the importance of the textile industry, and hence the resulting significance of dress, but it was really his student Yedida Stillman who took the historical study of medieval Islamic dress into a new realm of scholarship.

For the Fatimid period we find numerous references in the Genizah documents. For the eleventh and twelfth centuries, the following brief discussion and description depends mostly on the translations and interpretations given by Goitein, in particular in his volume on daily life (Goitein 1983), as well as on the excellent analysis provided by Stillman (1972, 1976). Regarding specifically the dress of women, a large number—a total of 750—of trousseau lists are of outstanding value. These lists not only allow us to visualize the dress apparel of a middle-class woman living in Egypt in the eleventh and twelfth centuries, but also to comprehend the domestic environment and furnishings that adorned the interiors of houses at the time prior to the thirteenth century. The documents are primarily, or even exclusively, concerned with Jewish wedding con­tracts and trousseau lists, but these families at the time seemed to live in a generally integrated manner with other Cairene or Fustat urban families. For that reason, their lifestyle may be informative for our purposes [6]. The lists include all durable items that a woman brought into the marriage, and many textile items were considered durable rather than fragile [7]. The lists do not include, for that same reason, pottery, mats, and glassware, although these items were in common use: they could easily be broken and would be discarded. The trousseau, however, represented the wife’s wealth, which would be inheritable but also could move away with her, should the marriage be terminated. Without the sources provided by the female side of society, we would not be able to have as clear a picture of the domestic setting at the time.

For the Mamluk period the evidence is more dispersed; L.A. Mayer has compiled some of the historical sources with a specific reference to dress (Mayer 1952). Considerably more work needs to be done in this field, though; it would be most useful, for example, if a more ambitious inte­gration of historical documents with surviving textiles were attempted. Mayer begins this study, but far more could now be done. Fortunately we also have some manuscript illustrations for the Mamluk period that show people, and the garments worn by them. These representations can tell us the general appearance of dress, although they are too generalized to provide a detailed picture. The Mamluk manuscripts showing people are in contrast, in this respect, to the Jain manuscripts referred to in the previous chapter. It certainly is not possible to say that the manuscripts actually illustrate Indian textiles, even where they do show designs and patterns. In fact beyond the evidence we can piece together from the surviving seams and hems, we do not know specifically for what type of garment the Indian textiles were used. The surviving child’s shirt [EA1990.404] is a singular piece, and we cannot generalize from it alone.

First a few general comments on Islamic dress may be in place, as certain aspects developed early and continued into and past the Fatimid and Mamluk period. Early Islamic laws and customs were quite austere about the appropriate manner of dress, but by the time of the Umayyad period and the rise of an Islamic political empire, luxury cloths had become important and were used by the caliph and his retinue. It was already during this time that two specifically Islamic dress-related customs developed: sumptuary laws that required distinctive dress for the non-Muslim population, and the production of royal embroidered fabrics for clothing (Stillman 1986: 736) [8]. Neither practice is directly connected with the Indian trade cloth, but each is nevertheless mentioned here because it played a role in the society that eventually received the printed textiles as imports.

The requirements for non-believers to wear specific clothing, or the restrictions from using dress items reserved for Muslims, were worked out in detailed sumptuary laws at a later time (Tritton 1930: 115-26), although there also were periods when these rules were either relaxed, or did not apply at all. One such time, for example, was the Fatimid era in Egypt, which was generally distinguished by its religious tolerance. During the Mamluk reign in Egypt, however, these sumptuary laws were again applied (Mayer 1952: 65-8). They focused on the wearing of particular colours: male Christians and Jews had to wear blue and yellow turbans, and their women had to wear dresses of a particular colour. When it became fashionable to use ever bigger turbans, and the sleeves of robes became wider, Christians and Jews were restricted in the amount of cloth they were allowed to put into their garments. However, the major visual sign of differentiation between different religious groups seems to have been the wearing of specific colours (Mayer 1952: 67). By visually separating the believers from the non-Muslim, divisions within the society were made apparent. The dress of a person identified him or her with a particular section, possibly with a religious or ethnic group.

The development of royal workshops that produced embroidered textiles with a controlled distribution also began during Umayyad times. The fabrics were known as tiraz, literally ‘embroidered’, more especially with bands of inscriptions. By extension, the term was applied to the whole garment as well. Tiraz textiles were given to officials and courtiers, as tokens of the ruler’s esteem, but also to identify the wearer as a member of his retinue, and hence his political dependant. As the inscriptions referred specifically to the reign and the workshop, the surviving examples of royal tiraz can give information about date and location of production. The custom of using a tiraz system probably had its origin in both Byzantine and Sasanian practices (Kühnel and Bellinger 1952: 1; Serjeant 1972: 8-9; Stillman 1986: 736).

Both aspects of dress control—the sumptuary laws and the handing out of textiles as a royal favour—touch on the important role textiles had in medieval Islamic society. Far beyond the pro­tective function of dress, textiles had a social and political function as well. Although this is by no means unique, as it is clearly a fundamental aspect of human behaviour to use dress for the purpose of defining ethnic and social identity, it was particularly developed and refined in the medieval Islamic world [9].

The tiraz system elaborates on the potential political purpose of dress, as it identifies the wearer as a person with allegiance to the ruler. The gift of cloth in this case was a political statement that implied ties of loyalty and dependence. At the same time, it also represented material value of a considerable kind, as the cloth produced in the royal workshop was of the best quality. Textiles were not mere consumables, but represented economic wealth as well [10]. Clothing was part of a family’s financial investments, to be passed on from parents to their descendants, or—if need be—to be sold.

These different aspects of the ownership of cloth—social status, political alliance, and material wealth—are essential to an understanding of the role of textiles in the medieval Islamic world. To some degree surprising for us is that there was often no gender distinction between types of cloths. Many textiles could be worn by men or women; this was particularly so for wraps and tunics:

male and female fashions did not differ very much in their make. If needed or desired, husband and wife could use the same outerwear. Examples of such occurrences are found in Talmudic and Arabic literatures and are mentioned in Geniza letters. (Goitein 1983: 153)

However, there certainly were specific gender-related ways of wearing the cloths.

One can say generally that Islamic dress as worn throughout the Middle East has proved to be conservative in its basic structure. Over the centuries, it has consisted primarily of tunics and wraps, with relatively little tailoring. The basic undergarment of early times was the izar, a loincloth already mentioned by Herodotus (1890: Histories VII. 66) as being worn by the pre-Islamic Arabs. Persian influence introduced the sirwal, tailored underdrawers worn by women, in particular. Both men and women alike wore a shirt called the qamis [11]. Over this garment a variety of robes could be worn, usually tunics or wraps. When one went out in public, a mantle was put on over everything else. Always of exceptional importance was the head-dress: the turban or wrapped hat for men, and the veil or wimple for women (Goitein 1983: 158-9). Therefore, when leaving the house, the entire body was covered with wraps; as Goitein says:

Medieval Islamic illustrations... usually show us the human body concealed by several layers of garments, mostly wide and flowing, so that often the contours and limbs of the body are hardly recognizable. (Only persons of low manual occupations, such as peasants, butchers, or bakers, possess a body and are also occasionally permitted to go around bare-headed). (1983: 160)

Dress, in other words, was assembled from layers. Only very poor people had a single wrap or tunic; to be moderately well dressed meant several cloths to cover the body. To quote Gotein again in a reference from a Genizah document:

A man on his deathbed wishes to have a simple burial: ‘No wailing women, please; and of garments in which I shall be buried I wish to have no more than these: two cloaks, three robes, a washed turban of fine linen—it is already wound up—new underpants of mine, and a new waistband of mine. ’ Clearly, five covers of the body were regarded by persons of the class to which our dying man belonged as something very austere. (1983: 160)

A specifically Persian influence was the sleeved mantle. This was already present in Arabian pre- Islamic times, wherever people were influenced by western Asian cultures, and in affinity to the Parthian style of dress. The sleeved coat was a typical Central Asian garment, which ultimately connects with the long-standing tradition of Chinese tailored dress. In Egypt during the Ayyubid and Mamluk reign, Central Asian dress became the appropriate military and ceremonial wear. This marked a real change from the loose covers and sewn tunics of the Fatimid period, where belts were the primary way of pulling together loosely hung clothes. The Ayyubids preferred Turkish coats, which had hems that crossed the chest diagonally from right to left, while the Mamluks wore the Tartar coats that crossed from left to right (Stillman 1986: 739).

The general form of dress may not have changed much over the centuries, but fashion was cer­tainly expressed in the type of fabric, in the manner of wrapping, and in accessories. At different times and locations, different products were sought after, and imports from other production centres influenced the appearance of dress. This goes back to early Islamic times. When discussing the nature of clothing during the Abbasid period (750 to 945 CE), Stillman says:

Fine garments were brought to Baghdad from all over the Muslim world, as well as being imported from abroad. From India came the futa, a long piece of sari-like cloth which served a variety of functions: as a loincloth, apron, and a variety of headgear... From China during this period there came oilcloth rain-cloaks... (1986: 737)

The textile trade with India was no doubt of great importance in the creation of fashion, although we cannot claim that the cloth mentioned was of the block-printed variety.

In Egypt the Fatimid period in particular was extremely fashion conscious. Fine linen and silk were the most precious textiles, and these were often decorated with embroidery or brocade weav­ing. We hear, via al-Maqrizi and al-Qalqashandi, of the pomp and ceremony of the caliphal court, and of the significance given to ceremonial dress on these occasions. The inventory lists of the Fatimid treasury, made in connection with its dispersal in 1061-9 CE, provide ample evidence for the type of garments used (Romberg 1985). The Genizah documents, on the other hand, give evi­dence of the court’s preoccupation with dress in at least one section of contemporary middle-class society. In the collections of documents, as mentioned above, approximately 750 trousseau lists have survived (Stillman 1972).

These give unique evidence for the appearance of female attire at the time, not only for women of the Jewish community, but for the society in general. As is clear from the bulk of documents, there was very little or no segregation between different religious groups living in Egypt, in par­ticular in cosmopolitan Cairo and Fustat. The documents also show that the middle class of the time tried to imitate the ruling class in many ways. For example, tiraz textiles were given to rela­tives and friends on particular occasions: not of course to be confused with the royal tiraz gifts, but certainly inspired by them. These textiles could be very similar in appearance to the royal gift textiles, but their inscriptions were general rather than specific. They might even only be ‘fantasy’ inscriptions, inspired by Arabic script, but without meaning. Precious textiles became generally available, and they were used by all who could afford them. Over 60 types of fabrics are mentioned in the Genizah: 46 correspond to textiles known from other literary sources, as collected by Serjeant (1972: 135-62). These fabrics were used for numerous different garments. The trousseaux mention almost 70 types of clothing, referring, of course, only to women’s dress. It must be remembered though that the trousseau lists cannot be representative of the cloth generally available, as they will only mention durable and valuable textiles. Purely utilitarian cloths are absent, although these will have made up the bulk of material in all households. Nevertheless, the lists give an impression of the wide range of dress articles available, as well as providing an appar­ently reliable account of what was considered essential for the comfortable interior of a home.

Textiles as furnishings

As was noted above, the trousseau lists give not only the clothing a woman brought with her, but also furnishing articles. From them it is evident that the use of textiles in the furnishing of houses was as important as their use for clothing. Again, here also we find that there was no fundamental difference in form between the courtly and the well-to-do middle-class habitat. Of course the level of luxury at court was extreme, while the professional classes, the merchants, and well-established artisans used far more modest materials. But in all levels of society textiles were a primary part of furnishing rooms, as mobile furniture was kept at a minimum [12]. Seats may have been part of the architectural building structure, as is still the case in Mediterranean and Middle Eastern houses. They could also be separately constructed out of wood or bricks. In any case they would be covered with cushions and pillows, and it is these that were referred to as part of the trousseaux. Divans and sofas were composed of mattresses and bolsters, covered with decorative textiles. The wealth of a household was reflected in the quality and sumptuousness of the cloth used as upholstery, as well as in the number of divan mattresses available.

Bedsteads were in use in Byzantine society, as a continuation of the classical Roman legged bed and sofa. In Fatimid Egypt, however, they had gone out of fashion by the eleventh century, and sleeping mattresses called firash were used instead (Goitein 1983: 108-9) [13]. In less well-off house­holds these could be used as day-beds, for sleeping on at night and sitting on during the daytime.

The trousseau lists indicate, however, that, whenever possible, there should be separate mattresses available. The mattress for daytime repose was called martaba. It took first place in the lists of furnishings; it was usually made up from several pieces of seating sections. In addition there were bolsters and cushions to give comfort and support.

Rugs were spread on the floors, and large hangings may have been used to partition off parts of a room. Curtains were not placed over the windows, however, as these had window screens carved of wood, the mashrabiya still found in traditional Cairene architecture. Bedding could be rolled up during the day. Chests and trunks were used, but storage and display shelves could be built in as well. In effect, there was little movable furniture, and the appearance of domestic interiors would have been dominated by textiles and household items, which could be ornamented metal ware, ceramics, and glass.

Indian trade cloth

How did our Indian printed textiles fit into this setting? Regarding the furnishing of interiors, the presence of borders and tabs show that they certainly had a role in domestic decoration, in addition to their use as dress items. They were almost certainly used as curtains, room-dividers, or door hangings, as well as outdoor canopies on ceremonial occasions. They were also used as tailored garments, as is obvious from the presence of seams in many of the fragments. As piece-goods, uncut and not tailored, they were possibly also used as body wraps, as loincloths, much as the futa mentioned above from Baghdad, or as turbans. They still function like this in their countries of origin (India and Pakistan), as well as in parts of the Yemen.

In the period that coincides with the Genizah documentation, it is quite clear though that the Indian cotton cloths were a far cry from the refined luxury items of the Fatimid era, either as furnishing or dress material. Therefore, their function in Egyptian society was distinctly different from the role similar cloths took on when they were carried from India to South-East Asia. The printed textiles became prestige goods in several South-East Asian societies, and they were used as ceremonial cloths and funeral shrouds. This was not generally their function in Egypt. While the majority of tiraz textiles have come from burials, we have only scanty evidence for the use of the Indian printed cloths as funerary dress [14]. They are also not mentioned in the trousseau lists, nor is any cotton fabric, for that matter. Goitein (1983: 170) speculates that the absence of cotton goods may be due to the fibre’s relative fragility, when compared to flax. Certainly the trousseaux had to contain articles that were durable, as they represented a woman’s major portable wealth. Care was taken that clothing of value could be passed on, rather than used as utilitarian, disposable items. Linen garments are generally longer-wearing than cotton ones, but the trousseaux also list silk clothing, which is often more fragile than well-spun and closely woven cotton. The relative fragility of cotton, therefore, can be no reason for its absence from the marriage chests. It is more likely to be for reasons of prestige and economic value.

To determine the social and economic standing of cotton, let us look at the Genizah reports on cotton goods in general. In his discussion of the main industries in the communities with relation to the Genizah, Goitein says:

Cotton, today the staple crop of Egypt, was of no great consequence in our period. In its raw state, it was imported into Egypt both from Syria and from the West (Sicily, Tunisia), as we read repeatedly in the Geniza records. Cotton goods, of course, were one of the main imports, both from Tunisia and from India. Therefore, we hear much about the trade in cotton, but next to nothing about its manufacture. (1967: 105)

Discussing clothing in his volume on daily life, we see the following quotation of the famous Jewish religious scholar Abraham Maimonides (1186-1237), from a Genizah document:

The ancient sages did not make it obligatory that the clothes of a scholar be silk in the winter and thin fine linen in the summer, but that his garments should be spotlessly clean, even though they consist of coarse cotton in the winter and second-hand linen in the summer. (Goitein 1983: 165)

Goitein himself says:

In Egypt, cotton seems to have been used mostly for working clothes, pants, and bedding, in addition to serving as lining or filling for cloaks and wraps made from other fibers. (1983: 170)

Then however he continues with a reference to textiles traded from India, which he believes to be cotton; we have already heard about these fabrics in Chapter 4 on dating, but I want to use the extract again, in a slightly expanded context:

A textile constantly mentioned as a main export from India to the West, and especially from the north-Indian province of Nahrawara-Gujerat, requires some attention. It was named mihbas (more commonly occurring in the plural mahabis), which might be translated as wrapper. It was traded in scores, consisting of twenty complete or half thawbs, a word usually denoting the robe worn by everyone, male or female, but also the piece of material needed for a robe. The mahabis were also used for the manufacture of pillows; red and black are mentioned as their colors, and their average price, approximately 1 Adenese, or 1/3 Egyptian dinar, makes it more than likely that they were made of cotton, the most important of Indian textiles. The term mahabis was [otherwise] not used in Egypt [when speaking of cotton]; when a Jewish merchant on the India route sends ‘a fine cotton robe’ to a Muslim business friend in Qus, Upper Egypt, he uses the regular Arabic term qutn (from which the English word is derived). (1983: 171)

In other words, although cotton cloth itself is spoken of as an inferior fabric, only suitable for the indigent, he also thinks it likely that a frequently referred to import cloth from north-western India is cotton. From the context of the quotations, it is evident that the poor quality of cotton cloth must refer to fabric that is not Indian, but probably comes from Tunisia or Syria, or even is a poor quality local product. The Indian textiles, however, were a common type of utilitarian fabric. They were not expensive and certainly not of luxury quality, but they were not for the poorest either. Important is the colour reference of red and black, i.e. mordant- and indigo-dyed. It is revealing that in addition to their use as dress items, he has also found references to the cloth for the making of pillows, i.e. domestic furnishings. This coincides with the surviving textile evidence. Goitein’s description of the cloth, as being red and black, is not in contradiction with our material. Often no clear distinction is made between black and blue, especially when a deeply dyed indigo cloth is referred to.

To sum up, I think it very possible that the mahabis mentioned by Goitein are the printed textiles of our collection. The same term of mahabis appears in a twelfth-century tariff record from Aden, at the time of the Zurai'id dynasty that ruled from 1080 to 1173 CE; the dues mentioned are ‘on a score of cloaks (kawradja al-Mahabis), four dinars’ (Serjeant 1972: 130). Of course this identification has to remain hypothetical, but the descriptions of their appearance, quality, and purpose are closely comparable to what our fragments suggest.

So far our investigation has kept closely to the internal evidence that the Indo-Egyptian textiles themselves provide. To round off the picture drawn here, the following two chapters will place the textiles in a much wider field.