This document is part of the online version of the book Amheida II: A Late Romano-Egyptian House in the Dakhla Oasis / Amheida House B2 by Anna Lucille Boozer, which is available at http://dlib.nyu.edu/awdl/isaw/amheida-ii-house-b2/. It is published as part of the NYU Library's Ancient World Digital Library and in partnership with the Institute for the Study of the Ancient World (ISAW). Further information about ISAW's publication program is available on the ISAW website. Please note that while the base URI of this publication is stable, the exact content available at that address is likely to change over time.
Woven materials may be defined as objects that have been formed by interlacing or intertwining material. Both textiles and basketry fall under the general rubric of woven materials, and both of these forms are represented in House B2 at Amheida. Woven material was a much-neglected artifact category in the earliest days of Egyptology. In the past one hundred years, and particularly since the 1940s, a number of scholars have greatly furthered research on these topics.1↩#p3145
The preservation of organic material is exceptional in Egyptian desert sites, and it is not uncommon to find well preserved textiles and baskets in Egypt. Unfortunately, soft organics from Amheida do not derive from sufficiently dry contexts to have the survival level we have come to expect from oasis sites such as Kellis.↩#p3146
The author examined all of the woven material from House B2 on site during the 2008 field season.2 Data for this study derive primarily from three fragments. There is one fragment from a sewn plaits basket (Inv. 11554), one fragment of tabby-weave reddish-brown linen (Inv. 11526), and one fragment of light blue cotton (Inv. 11527). All of these fragments derive from rooms inside House B2.↩#p3147
Textiles
A textile is a woven fabric or any sort of cloth. We have two textile fragments from House B2, one fragment of tabby-weave reddish-brown linen (Inv. 11526) and one fragment of light blue cotton (Inv. 11527). The low density of textiles in House B2 compared to Kellis can be expected due to both the damper conditions at Amheida as well as the type of context that we have for House B2. Most textiles excavated at Kellis, as well as other Egyptian sites, were found in cemeteries.3↩#p3148
Ancient Egypt is known for linen cloth, but the Egyptians used other textile fibers as well, including sheep’s wool,goat hair, and palm fiber.4 The majority of ancient Egyptian textiles are made of linen, which is made from the bast fiber, flax.5 Linen fabric in Egypt was made from cultivated flax plant fibers (linum usitatissimum) and produces a light fabric that is appropriate for hot climates.6 Linen, usually in its unbleached form, was the most common yarn used at Kellis. It ranges in color from off-white to light brown. Such plants were probably grown locally, as flax capsules have been found at Kellis as well as at Amheida, such as in room 3 of House B2.7↩#p3149
In the Ptolemaic and Roman Periods (from 323 BCE onward), production of linen was strictly controlled by the state, both for the quality and quantity produced.8 The Kellis Agricultural Account Book lists a weaving workshop among its tenants but there is no evidence for large-scale, controlled linen production in Dakhla itself.9 Even so, fine linen appears to have been produced in the Dakhla Oasis and transported to the Nile Valley.10 It seems that Tarsian linens were the only linen goods that were exported from Egypt in the late third and early fourth centuries CE.11 The local linens, such as those found at Amheida, were probably of inferior quality.12 This evidence suggests that linen could have been among the luxury agricultural goods Dakhla inhabitants produced for sale to other oases or to the Nile Valley, but that linen from Dakhla would not travel further in trade.↩#p3150
Egyptians did not generally use cotton until the first century CE.13 By at least the second century CE, cotton was grown in Egypt.14 Local cotton cultivation is suggested because it has been recorded in the Kellis Agricultural Account Book and material evidence for cotton has been found at Kellis in the presence of numerous cotton bolls and seeds.15 Evidence of cotton has also emerged from the Kharga Oasis, in the form of both documentary sources and botanicals.16 Cotton was certainly an established crop in both oases by the fourth century.17 As many scholars have observed, the Oases provided a favorable agricultural venue for cotton production over the Nile Valley.18 It is probable that this product contributed significantly to the Roman investment in the region.↩#p3151
Our fragment of light blue cotton (Inv. 11527) is one of a few finds of cotton cloth in both oases.19 There is a general lack of cotton fabric found in metropolitan Egypt, despite documentary evidence that cotton cloth was neither expensive nor rare in Roman Egypt.20 The House B2 cotton fragment adds to this growing corpus of finds from the oases, justifying previous arguments that the concentration of finds in this locus is significant.21↩#p3152
Weaving
The loom weights found in and around House B2 suggest that the type of loom used was the warp-weighted loom.22 The warp-weighted loom had a single major beam to which the warp was attached. The warp could be prepared by weaving a narrow strip leaving a long fringe of weft loops at one edge; the woven strip was then attached to the horizontal beam. The evidence for a warp-weighted loom at Amheida is consistent with the numerous mud loom weights found in the domestic structures of Kellis.23 The use of a warp-weighted loom for producing linen in House B2 is of particular interest, as Wild previously suggested that this method of linen weaving may have been what distinguished Tarsian linen from local linen.24 Evidence from B2 suggests that local linen may also have been woven using a warp-weighted loom, although a larger corpus of evidence would be required to promote this idea more securely.↩#p3153
There is strong archaeological evidence for domestic weaving in Roman Egypt. Spindle whorls and loom weights were found in every domestic structure excavated at Kellis, which suggests that spinning and weaving was carried out within the household, probably to meet needs within the family.25 The Kellis corpus also attests to a small textile and tailoring business organized by one of the female occupants of an excavated house, which provides evidence that such goods could be sold locally.26 Weaving in Roman Egypt was not necessarily confined to women, as an archive spanning three generations of male weavers from Oxyrhynchus attests.27 Goods found in Kellis houses include private letters, deeds of sale, contracts, accounts, fragments from implements used in the weaving trade, as well as scraps of decorated fabrics and unwoven warp threads.28 Unfortunately these finds have not been published fully yet so it is not possible to test the assemblage of a business-oriented house against a house that contains an assemblage of weaving implements for domestic usage. Such comparanda would prove useful for future inquiries into textile production.↩#p3154
The identifiable weaving patterns used for Amheidan textiles are in the basic plain or tabby weave that is ubiquitous in Egypt and pre-Meroitic Lower Nubia. The quality of this weave can vary between fine gauze and heavy canvas. Our scraps are of a medium density and unremarkable in quality, suggesting that they were produced for local consumption.↩#p3155
Dying
Both of our textile fragments were dyed during the course of their fabrication, although the poor quality of the fragments renders it difficult to determine the stage at which these fragments were dyed.29 The linen fragment (Inv. 11526) was reddish-brown and the cotton fragment (Inv. 11527) was light blue.↩#p3156
One of the most common sources of the blue color in Egyptian textiles is indigitin, a substance found in both indigo and woad.30 These plants are not native to Egypt and were probably first imported from the Levant at some point during the Eighteenth Dynasty.31 Dying was accomplished by dipping fibers into a colorless vat of indigotin. The blue color came with exposure of the fibers to the air, which catalyzes the necessary oxidization. Dipping the fibers more times into the woad or indigo dye vat created deeper blues.↩#p3157
The rust-colored red fabric (Inv. 11526) is potentially the result of the degeneration of red or purple hued fabric. Purple was the most common fabric color at Kellis, and several texts, in both Greek and Coptic, suggest that purple dye was once manufactured at Kellis.32↩#p3158
Care of Textiles
Textiles were carefully cared for during their life history of usage. Individuals frequently mended textiles in antiquity—sometimes as many as three or four times—before they decided to use them for a different purpose. We do not have direct evidence for textile menders, but it is likely that women took charge of repairing worn clothing and cloth goods. Sewing and repairing textiles may have been a social activity, like spinning, that involved groups of women together.33 Most repairs consisted of darning, mending, and patching, which only became common during Ptolemaic and Roman Egypt.34 Once an item of clothing became too worn, it was sometimes made into a smaller garment, cutting away the most worn portions and refashioning the stronger portion into a new item of clothing.↩#p3159
Wooden storage chests were expensive and rare, and household cupboards unknown. The few possessions an ancient Egyptian family had were kept mostly in baskets. Cloth was frequently stored in lidded baskets, or, in more elaborate circumstances, in boxes.35 Petrie found various goods stored in baskets at Kahun, which attests to this practice of storing cloth goods.36 Within House B2, it seems likely that individuals stored extra clothing and cloth in baskets.↩#p3160
Uses of Textiles
Throughout Egyptian history, textiles were widely used in households for cushions, towels, bags, lamp wicks, curtains, bedding, and wall-hangings. Outside of the home, cloth could be used as sacks for transportation, as strainers for the production of oils, and equipping animals and boats. Cloth even takes on a significant role in medical, religious, and funerary uses. For example, cloth was used on a daily basis in the washing, feeding and clothing of divine statues in temples.↩#p3161
When we have only small fragments of textiles it is difficult to tell whether scraps are from items of clothing, tablecloths, cushions, blankets, or wall hangings. All of these types of cloth objects are attested at Kellis in documentary sources and suggest that all of these options should be considered viable for the fragments recovered from House B2.37↩#p3162
Basketry
Scholars often describe basketry and matting as “textile techniques.” A basket and a textile are clearly distinguishable, but a gray area can exist between these two categories. The object described here (Inv. 11554) falls unambiguously into the category of basket, as is clear by the tools and raw materials that individuals used to create it.38 Wendrich has defined basketry as “objects made of plant parts of limited length often with a shape to that specific plant part.”39 The ways in which a basket maker must compensate for the raw material’s irregular shape makes basketry techniques fundamentally different from textile techniques. With textile manufacture, the maker controls the long yarns used for their craft.↩#p3163
Our basket fragment (Inv. 11554) is a sewn plaits basket and, as a result, unambiguously dates to the Greco-Roman period. There was clear continuity in basket forms and techniques from the Neolithic through Pharaonic cultures in Egypt. During the Greco-Roman period there is a shift towards sewn plates and stake-and-strand baskets, which rely on techniques that rarely occurred in Pharaonic Egypt. Although these innovations took place during the Greco-Roman Period, traditional Egyptian techniques remained alongside these innovations.40 Today sewn-plaits basketry is still the most common type of basketry that one can find throughout Egypt.41↩#p3164
Sewn-plaits techniques involve two stages of manufacture: first the basket maker plaits long strips. Then the craftsman sews them into an ongoing fabric and they sew baskets from one spiraling strip.42 Although the sewn-plaits technique does not occur until the Greco-Roman Period, it is ubiquitous during that time, so further nuance for the dating of our fragment is not possible. Craftsmen used this technique to fashion floor mats and flexible baskets.43 From the Ptolemaic Period onward sewn-plaits basketry were used mainly outside the house for carrying seeds, grain, and other commodities or for moving pots, harvested plants, earth, or dung on donkey-back.44↩#p3165
In order to create plaited basketry, craftsmen required no tools other than a pot with water to soak the palm leaf and a large needle. The needles used for these baskets range from 10-30 cm in modern Egypt and Nubia. They can be flat or round.45 Unfortunately basket production areas and tools are often difficult to locate archaeologically since the materials used for production do not leave a distinctive residue, if there is a residue at all.46 It is not possible to determine where our fragment was produced in B2 or if it was produced outside of the house.↩#p3166
Plaiting is a technique in which both systems (a strand or a number of parallel strands of the same material, orientation, and function) are active. In the first stage, long strips are plaited out of strands of palm leaf or bundles of grass. In the second stage the plaited strip becomes the passive system, which is sewn with a string (the active system) into an ongoing fabric.47 Size, shape, flexibility and the space between the strands are the most important properties involved in the crafting of baskets and mats.48↩#p3167
There is no simple correlation between function and size, shape, flexibility, and spacing attributes. Ethnoarchaeological research shows that the function of baskets can be highly specific and that tradition plays an important role in determining the function of specific forms. Moreover, a basket might be used for something other than its original function, and there are many examples of baskets and mats being reused for purposes other than their original function.49↩#p3168
The production of basketry involved several stages, and the individual who made the basket probably executed all of these stages. For most baskets made from grass and palm leaf this task can be accomplished year-round. Other basket types, such as those made from rushes and sedges, are seasonal products. The harvesting and storing of all of these raw materials required some level of specialized knowledge.↩#p3169
The producers of sewn-plaits basketry—the type found in House B2—were probably part- or full-time professional males and there is evidence of this specialization among the early Christian hermits.50 Indeed, we have a number of ancient texts about Christian monks throughout Egypt who made sewn-plaits basketry.51 That there were (semi)-professional basket- and mat-makers suggests that basketry was bought and sold. It seems likely that the prices of the baskets were sometimes included in the price of the materials that they contained.52 Individuals naturally reused the baskets after the original goods that they contained were used.↩#p3170
Baskets were useful in a variety of contexts, including trade, domestic, transportation, and storage. Less wealthy individuals, in particular, had fewer options for storage or matting, and baskets must have played an important role in household furnishings. Among more wealthy individuals, baskets may not have been used as ubiquitously, although they were certainly part of the household assemblage.↩#p3171
The trajectory of basket reuse depended, in part, upon the type of basket, but most baskets were reused in some manner, since they were labor-intensive goods. Evidence suggests that old baskets were used to line storage pits. Families passed down decorated and particularly prized basketry as heirlooms. The baskets from Kharga, for example, were thought of as fine works, at least during the New Kingdom.53 Once the baskets had transcended all practical use, they were often employed as fuel for ovens, kilns, and other fires, thereby ceasing their life-cycle of usage.↩#p3172
Description: Fragment of textile. The string is twisted in S-direction. This textile is not found after the very early fourth century and is a signature of the third century in Dakhla. This color of textile is found only in House B2 thus far in Dakhla. Another cotton textile fragment was found at Amheida (Inv. 47) from windblown sand in House B1 (DSU 2) and is probably modern.↩#p3183
Parallels: Two blue, tabby weave cotton curtains were found in fragments from a mud brick temple found at Qasr Ibrim, see Adams 1987: Qasr Ibrim Numbers 86T/025 and 86T/023. On blue yarn hank-dyed of spun cotton yarn from Qasr Ibrim, see Wild and Wild 2009:4. Plain cotton tabby weaves have been described but not published with inventory numbers for Berenike, although many of these may be considered to be imports (Wild and Wild 1996; Wild and Wild 1998; Wild and Wild 2000). Kellis has produced several fragments of simple open weave and tight-weave textiles.54 All of them are cotton and brown.↩#p3184
Description: Dimensions refer to the biggest fragment. Four fragments of tight weave textile. The string is twisted in S-direction. Tabby weave.↩#p3195
Parallels: The closest parallels are from Kellis. The plain Kellis fragments are described rather than inventoried and photographed in preliminary reports, although each fragment was catalogued and analyzed (Bowen 1999:9; Bowen 2001:18, 20; Bowen 2002b:91).↩#p3196
Figure 19.2: Three fragments of sewn plaits basket 11554 (M. Hense).
Description: Three fragments probably of a sewn-plaits basket. The sewn-plaits technique does not occur until the Greco-Roman Period, and it is ubiquitous during that time, so further nuance for the dating is not possible. Weaving pattern: twills. The bigger fragment is made of six plaits, fiber width: 1 cm. The weft is passed over more than one warp strand. This sort of weave based construction was typically used for matting and flexible, bag-like baskets, which generally had handles for carrying. See Inv. 11519, a basket handle, from the wooden objects catalogue.↩#p3207
Parallels: Kellis has well preserved examples of plaited baskets from domestic contexts (e.g. A/6/73 from House 4), see Bowen 2002b:91, 101, pl. 11. For the wooden basket handle, see Petrie 1889:11, pl. XIII, 18.↩#p3208
2 There were no woven materials from Street S2 or Courtyard C2 recovered during the 2007 field season. The author would like to acknowledge the assistance of Gillian Bowen and Colin Hope in her examination of these objects during the 2008 field season. Their expertise and willingness to provide unpublished comparanda and suggestions was greatly appreciated.
9 On the KAB evidence, see Bagnall 1997:line 1266.
10 (P. Kell I. Gr. 51.3–6). On this evidence of weaving, see Bowen 2001.
11 The term tarsikarios seems to have been introduced to describe immigrant Roman weavers in Egypt, who differed in some way from Egyptian weavers: Wild 1969:811.
12 Ibid.:818-819. See also Edictum Diocletiani, XXVI, P.Cair. Masp., I 67006.66–67.
14 Wild 1997:8, Wild and Wild 2007, Wild, Wild and Clapham 2008:145-149, Thomas 2007:156 and note 57.
15 On the KAB mention of cotton, see Bagnall 1997:lines 547, 556, 558-59, 720 and 1484. On the material evidence of cotton at Kellis, see Thanheiser 1999. See also P. Kell. I, Gr. 61 (Worp 1995).
16 On the documentary sources, see P.Iand. VII, 142, II, 8 (Kalbfleisch 1912-1938), O.Douch IV, 381, 489; V, 596, 600, 634 (Wagner 2001).
18 On the production of cotton in the Small Oasis, see Bagnall 2008b. Excavations in the Eastern Desert at Berenike and Myos Hormos recovered a substantial number of fragments of cotton textiles from all Roman Period phases, although these were most likely imported: Wild 2006:179, 180 (fig. 4), 181, 183, 183 (fig. 7), 184, Wild and Wild 1996, Wild and Wild 1998, Wild and Wild 2000, Wild and Wild 2005, Wild and Wild 2008, Thomas 2007:155-156. On Myos Hormos (Quseir al-Qadim), see Vogelsang-Eastwood 1989.
19 On Kellis, see Coombs, Woodnead and Church 2002, Bowen 2002b.
