Précédent Suivant

Textiles from Mons Claudianus, ‘Abu Sha’ar and other Roman Sites in the Eastern Desert.


Texte intégral

1Textiles are a major find group at many of the archaeological sites in the Eastern Desert. At the Imperial Roman quarry of Mons Claudianus,1 excavators filled no less than twelve large cantines or metal boxes with textiles –totalling about 1,25 cubic metres– containing tens of thousands of dirty rags, mostly fragments, but a few more or less complete items of clothing such as tunics, hats and socks.2 A selection of them has been investigated by the author, assisted by Ulla Mannering and handweaver Lena Hammarlund, and other specialists working from recorded data.3 At ‘Abu Sha’ar, a late Roman fort later turned Christian monastery on the Red Sea coast some 20 km N of Hurghada, Steven Sidebotham’s team found 1126 textiles during his excavations there 1987-1993.4 They were recorded by the author (1990-1991) and Marion van Waveren (1993).5 Further textiles have been logged from at least eight other sites in the Eastern Desert. At Berenike, Felicity and John Peter Wild have recorded more than 3,400 pieces.6 Dominique Cardon has investigated hundreds of textiles from the praesidia Krokodilô and Maximianon,7 more than 400 bags of textiles from Dios and Xeron Pelagos,8 and together with Hero Granger-Taylor recorded 838 pieces from the rich textile deposits at Didymoi.9 In addition, Gillian Vogelsang-Eastwood and Fiona Handley have examined many thousands of Roman and Medieval textiles from Myos Hormos;10 Handley also investigated the textile finds from The Porphyrites11 (Table 1).

Table 1

Site

Textile numbers

Investigator

'Abu Sha’ar

1126 pieces

L. Bender Jørgensen,

M. van Waveren

Berenike

>3,400 pieces

F.C. and J.P. Wild

Didymoi

838 pieces recorded

< a third of those found

D. Cardon and H. Granger-Taylor

Dios

346 bags of textiles

D. Cardon

Krokodilô

250 pieces

D. Cardon

Maximinanon

431

D. Cardon

Mons Claudianus

12 cantines

1,833 pieces recorded

L. Bender Jørgensen,

U. Mannering, L. Hammarlund

Porphyrites

891 pieces

F. Handley

Myos Hormos (1978-82)

>3,000 pieces (Roman + Medieval)

G.M. Vogelsang-Eastwood

Myos Hormos (1999-2003)

2,455 Roman,

7,000 Medieval pieces

F. Handley

Xeron Pelagos

72 bags of textiles

D. Cardon

Textile numbers. Sites in the Eastern Desert where textiles have been recorded.

Mons Claudianus

2What did we find among the thousands of rags from Mons Claudianus, and what did we do with them? First it was a matter of finding ways to deal with them, setting up a laboratory in the desert and later at the workrooms of the Egyptian Antiquities Service in Dendera so that they could be cleaned, examined with a microscope, described, photographed if possible, packed and stored.12 We had to do some hard choices, as it was impossible to investigate all of them. The solution was to create two bodies of records: a random sample of 1265 pieces to represent the relative proportion of the different types of fabrics; and a selected sample of 568 unusual pieces, to show the range of textiles that had been available to the people who lived and worked at the quarry. The great majority of the textiles derive from the South Sebakh (SS); over 50% of all the recorded textiles were recovered here (Table 2). For the random sample, FSE and FW1 are further major sources, while supplementary groups of the selected sample were drawn mainly from FN1 and FSE.13

Table 2

Site code

Location

Excavated

No of textiles, Random sample

No of textiles, Selected sample

Total number of textiles

AL1

Animal Lines 1

1992

10

0

10

AL3

Animal Lines 3

1991, 1992

24

2

26

AL4

Animal Lines 4

1991

0

5

5

FE1

Fort East 1

1991

0

2

2

FN1

Fort North 1

1991

0

194

194

FN2

Fort North 2

1991

0

4

4

FSE

Fort South East

1990, 1991

248

74

322

FW1

Fort West 1

1991, 1992

124

16

140

FW2

Fort West 2

1992

19

0

19

Gate

Fort Gate

1992

22

0

22

Gran

Granary

1990

14

0

14

NEB

North East Building

1991

0

3

3

SS

South Sebakh

1987, 1988, 1989, 90

709

231

940

SWS

South-West Sebakh

1991, 1992

15

37

52

Well

Well Sebakh

1992

80

0

80

Total

1265

568

1833

Locations, numbers and years of excavation for recorded textiles from Mons Claudianus.

3Deposition of the textiles happened throughout the existence of the quarry (Table 3); the great majority can however be dated to one of two phases, based on the combined evidence of stratigraphy and dated ostraka: c. 100-120 and 135-160 AD.14 It is however important to note that the remains represent the last stage of the textiles’ life; they show signs of long use, were threadbare, heavily patched and mended or had been torn up to be used to wrap or tie around something. Some appear to have been cut in neat rectangular pieces that may have been intended to use as patches for other textiles.15 Therefore most of the textiles will have been produced much earlier than the date of their deposition. It is difficult to assess their durability, but considering that even today well-made wool textiles can last for years if kept safe from moths and other destructive factors we may presume that they had a life-span of several decades. Those of the early group are therefore likely to have been made in the late 1st century AD, those of the later group in the beginning of the 2nd century AD.

Table 3

Dating of deposition

Number of textiles

Ca 110 AD

80

100-120 AD

481

100-140 AD

172

Ca 140 AD

293

135-145 AD

186

136-160 AD

473

150-235 AD

26

Undated

122

Total

1833

Dates of the deposition of recorded textiles from Mons Claudianus.

Fibres, yarns and weaves

4The first task was to do a general description of each recorded piece such as fibre, yarns, weave, density, colour, edges, decoration, repairs or other features.16 The great majority proved to be wool (90%), although other fibres such as goat hair (4%) could be identified; vegetal fibres were rare (5.5%) and could not be further identified, but are likely to include flax, hemp and possibly a few cottons.17

5Yarn can be twisted two ways, clockwise (z) and anti-clockwise (s), or they can be S- or Z-plied, i.e. composed of at least two single yarns (Fig. 1).

Fig. 1

Image 100000000000024F0000017081279E86.jpg

Yarn types. Single yarns are described as z or s, plied yarns as Z2s or S2z.

© L. Bender Jørgensen

6Most textiles from Mons Claudianus were made of s-twisted yarns in warp as well as weft. This, in particular, applies to fabrics made of bast fibres (Table 4). Goat hair fabrics were generally made of Z-plied yarns in one or both systems. Like those of bast fibres, wool textiles are mainly made of s-twisted yarns in both systems, but other variations appear.

Table 4

Fibre

Yarn twist

No

Bast (flax and hemp)

s-s

47

s-ss

2

ss-ss

11

s-zz

1

z-z

2

S2s-ssss

2

Goat hair

z-ss

2

Z2s-s

4

Z2s-ss

6

Z2s-Z2s

38

S2z-z

1

Yarn types in bast and goat hair textiles from Mons Claudianus (random sample).

7As we shall see below, the choice of yarn types varies significantly depending on the weave. This is of great interest, as the choice of twist direction often depends on local tradition. In most societies, one of these directions is the norm. The twist of warp yarns is usually according to this norm, but weft yarns twisted in the opposite direction are sometimes chosen to obtain a different texture. In some cases, groups of yarns in alternating twist direction have been used to obtain optical effects. These are known as spin patterns.18 Decorative stripes and bands may also be made from yarn in deviating twist. The basic twist may therefore serve as a diagnostic feature when looking for textiles made elsewhere.

8In ancient Egypt, s-twisted yarn was preferred. This preference can be traced back to the Early Dynastic Period and is not exclusive to linens, as almost all wool textiles from el-Amarna (Eighteenth Dynasty c.1550-1290 BC) were made from s-twisted yarn.19 This appears to have persisted into the Roman Period, and also applies to Nubia in the Meriotic, X-group and Early Christian periods (c. AD 100-850).20 In other parts of North Africa, z-twisted yarns appears to have been preferred, although the body of finds to back this up as yet is very small.21

9Textiles in wool and linen from Syria and Palestine also show a preference for s-twisted yarns, although at Palmyra cottons and the finer wool fabrics, especially twills, are made of z-twisted yarns.22 Textiles from Iran and India have long been considered as made from z-twisted yarns.23 Unfortunately, there is little archaeological evidence to support this.24 In the northern Roman provinces and much of Europe, z-twisted yarn was the norm since c.500 BC and throughout the first millennium AD.25

10Weaves at Mons Claudianus are mainly tabby, although a variety of twills and damasks make up almost 6% of the random sample and 10% of the total. Other weaves and techniques include basket and half-basket weaves (Fig. 2), taqueté, two items in double-layered tabby, five pieces of so-called ‘Coptic knitting’, a tablet-woven band and 76 pieces of felt.26

Fig. 2

Image 1000000000000376000002659C471A7B.jpg

Weaves. a: tabby; b: half-basket; c: basket weave; d: 2/1 diagonal twill; e: 2/1 ribbed twill damask; f: 2/1 chevron twill; g: 2/1 lozenge twill; h: 2/2 diagonal twill; i: 2/2 chevron twill; j: 2/2 broken twill; k: 2/2 lozenge twill; l: 2/2 diamond twill. For ribbed and block damasks see Ciszuk 2004; for taqueté see Ciszuk 2000.

© L. Bender Jørgensen

11The fabrics of bast and goat hair fibres are all made in tabby or tabby derivatives such as half-basket and basket weaves and the yarn types used are those in table 4. The wool tabbies are also to great extent (670 pieces) made of s-twisted yarn in both systems; 132 pieces has s-twisted warp and z-twisted weft, while 72 have z-twisted yarns in both systems. Other combinations are much rarer.

12A different picture emerges from the half-basket and basket weaves: here, z-twisted yarns appear to have been preferred, particularly for basket weaves (Table 5).

Table 5

Weave

Yarn types

Number of pieces

Half-basket

s-ss or ss-s

13

s-zz or ss-z

9

z-zz or zz-z

33

z-ss or zz-s

3

Basket weave

ss-ss

1

ss-zz

3

zz-zz

12

Wool fabrics in half-basket or basket weave from Mons Claudianus (random sample).

13As regards twills, z-twisted yarns in both systems were preferred for plain diagonal twills 2/1 and 2/2 but s-twisted yarns for 2/1 and 3/1 ribbed and block damasks (Table 6). This might suggest that the latter were produced in Egypt or the Eastern Roman provinces, while the diagonal twills may derive from the European part of the Roman world. For 2/2 broken and chevron twills no clear trend is discernible but for 2/2 diamond twills twisted warp and s-twisted weft is most common. Diamond twills z-s are common in northern Europe where they have been found at Roman military sites like Vindolanda in Britain, in burials all over the Germania libera, and wrapped around mainly Roman weapons given in sacrifice in bogs in Denmark and northern Germany.27

Table 6

Weave

Yarn types

Number of pieces

2/1 twill diagonal

s-s

1

s-z

1

z-z

11

S2z-z or zz

3

2/2 twill diagonal

s-s

5

z-z

15

z-s

1

S2z-z

4

Z2s-s or z

2

2/2 broken twill

s-s

2

s-z

1

z-z

1

2/2 chevron twill

zs-ss

1

2/2 diamond twill

s-s

1

z-z

1

z-s

7

2/1 ribbed damask

s-s

3

S2z-zz

1

2/1 block damask

s-s

3

3/1 ribbed damask

s-s

1

3/1 block damask

s-s

7

3/1 block damask

s-z

1

Wool twills and yarn types from Mons Claudianus (random sample).

14Sett and density of are further aspects of a textile that have important impact on its appearance as well as on properties such as drape, flexibility and whether it is heavy, medium-weight or light-weight. Accordingly, thread count, i.e. the number of threads per cm in warp and weft, was measured. This allows us to distinguish between fabrics that are balanced, i.e. have an approximately equal number of threads/cm in both systems, and weft- or warp-faced fabrics where one system is dominant. In some cases, the dominant system completely covers the other; in tabby, this is termed repp. For twills, no corresponding general term is available. The twills of Mons Claudianus fall in two categories: balanced twills and densely weft-faced twills. The latter are termed ‘tight’ or ‘flat’ twills.28

Textures, terms and reconstructions

15The dry, technical data of fibre, yarn types, weave and densities offer valuable information on preferences of spinners and other choices of the craftspeople that produced the textiles that ended their life in the rubbish heaps at Mons Claudianus. They do however not suffice to describe them fully. We needed a tool that could capture and describe surfaces and texture. In order to delve deeper into this, handweaver Lena Hammarlund established categories within the common weaves such as tabby and twill that makes it possible to distinguish between fabrics that were visibly distinct but had almost identical technical description.29 Hammarlund’s terms such as ‘movable tabby’, ‘flat tabby’ or ‘crows-feet tabby’ have proved transferable to other groups of archaeological textiles30 and have thus added to the terminological tools used by textile scholars. Terms for stripes, checks and edges were also developed to facilitate characterisation and discussions.31 Handweavers Martin Ciszuk and Lena Hammarlund reconstructed a number of the textiles, including various patterned weaves,32 and experimented with different looms; this has led to a better understanding of the textile technology that was used to produce the textiles that ended up at Mons Claudianus.33

Identification of items of clothing and soft furnishings

16Working our way through the textile finds a few items were immediately recognisable as garments. This included the tattered remains of three tunics, two hats and a cloth shoe that have been studied by Ulla Mannering,34 and socks in so-called ‘Coptic knitting. Other items of clothing and soft furnishings have been identified later, through comparisons with textiles from other sites as well as studies of iconography, textual evidence, and discussions with colleagues working on textiles from other sites in the Eastern Desert and from Masada in Israel.35

17Tunics and mantles: The three more or less complete tunics were found in the SS and datable to 100-120 AD.36 One, tunic A, had clavi, the purple bands that were typical for tunics in this period and can be seen in mummy portraits as well as Roman imagery.37 The tunic was heavily repaired and most of the original material replaced by patches, but the clavi had been meticulously preserved throughout all repairs, indicating that they conveyed important information on the owner’s social position (Fig. 3). 151 textile fragments from Mons Claudianus had bands of the type defined as clavi38 and indicate that clavied tunics were worn at the site, but hardly by everybody. Tunic B (Fig. 4) had remains of sleeves, and had been put together from pieces of a mantle, as revealed by three oddly placed woven-in signs of the type called notched gamma.39 These are characteristic for rectangular woollen mantles of the himation or pallium type.40

Fig. 3

Image 10000000000004490000049D37B83DA1.jpg

Tunic A, MC 1100. The neck opening is in the middle, the clavi on each side of it, running across the two sheets of the tunic. Although parts of one side of the tunic is missing and the preserved fabric heavily repaired the clavi have been kept. For further documentation, see Mannering 2000.

© L. Bender Jørgensen

Fig. 4

Image 100000000000049D000002A56454ADE7.jpg

Tunic B, MC 1101. Three gammae at odd places show that the tunic was made from a himation. For further documentation, see Mannering 2000.

© L. Bender Jørgensen

18Tunic and himation were the typical costume of civilians in Roman Egypt, as seen in many mummy portraits.41 The heavy patching of tunic A and the himation re-made into tunic B probably represent the work of centonarii, rag dealers and patchwork makers supplying clothing for low status people.42 Many of the workers at Mons Claudianus would fit into this description, in particular those belonging to the familia.43 A total of 44 gammas were found at Mons Claudianus. They were examined by Ulla Mannering along with 9 other woven-in signs that she termed Eta, Iota, Khi and Delta;44 Etas are elsewhere termed notched bands or H-shaped.45 Notched bands and gammas were also used on other items such as tunics and headscarves46 and as we shall see below notched bands and other types of decoration that easily may be mistaken for gammas also appear on cloaks.47 They cannot therefore be used unequivocally as evidence for mantles of the himation type, although we again may infer that some of the people living at Mons Claudianus owned such mantles. Mannering’s ‘Delta’ sign is by others termed ‘stepped pyramid’ and derives from the shoulder of a one-piece long-sleeved tunic of a type that is known from Dura Europos and Khirbet Qazone and turns up sat Didymoi towards the end of the 2nd century AD.48 The Mons Claudianus ‘Delta’ was found in the Fort SE in a layer dated between 140 and 200 AD which fits well with the Didymoi evidence.

19The baby tunic (Fig. 5) is constructed as a sleeveless tunic, but the armhole is shaped as if a sleeve was to be fitted in. Put together by several pieces of fabric it is folded along the neck-opening and shoulder-seam and undecorated.49 It indicates the presence of small children at the site.

Fig. 5

Image 100000000000049D0000031330055C87.jpg

Tunic C, MC 891. Baby tunic.

© L. Bender Jørgensen

20Socks: One virtually complete sock in so-called ‘Coptic knitting’ or crossed looping50 was recovered from the FSE (Fig. 6); another was found in FN1 was fragmented but still recognisable. Three small fragments in the same technique turned up at FN1, FW1 and the SS; all are likely to derive from similar socks and were deposited between 135 and 160 AD.51

Fig. 6

Image 100000000000049D0000036E2BBCD524.jpg

Sock, MC 331.

© L. Bender Jørgensen

21Hats: Two hats were found virtually complete.52 One is of green felt, shaped much like a fez or pillbox, and like the pileus pannonicus used by soldiers in Late Antiquity (Fig. 7). The second is a cap with ear mufflers and neck protector, made of triangular pieces in red, green and yellow, resembling a helmet in shape (Fig. 8). Fragments of two similar caps have been found, one at Mons Claudianus,53 another at Didymoi.54 All the Mons Claudianus hats derive from the SS; two are dated 100-120 AD, one 100-140 AD.

Fig. 7

Image 100000000000037600000345425406AB.jpg

Felt hat, MC 922 seen from the front. For further documentation, see Mannering 2006.

© L. Bender Jørgensen

Fig. 8

Image 100000000000049D00000315F1E65793.jpg

Cap with ear mufflers, MC 1110, made from triangular pieces of red, green and yellow block damask. For further documentation, see Mannering 2006.

© L. Bender Jørgensen

22Cloaks: The work of Hero Granger-Taylor55 has made it possible to identify remains of further garments, such as the semi-circular hooded cloaks that were used by soldiers and civilians for outdoor activities. Among their characteristics are a narrow purple band running horizontally along the top edge of the cloak, turning at right angles towards the hood; double row of twining are situated some 5cm above the band and a little less where it is horizontal; and a small notched bar (Mannering’s Eta) at the bottom. Original edges (straight as well as curved) were often cut away when they had become ragged and replaced by sewn edgings.56 At Didymoi, several textile fragments have been recognised as deriving from such cloaks, especially in the form of off-cuts.57 Some are dark blue ribbed and block damask twills. These are compared to the dark blue cloaks seen in a group of mummy portraits of soldiers dated largely to the 2nd century AD58 and interpreted as deriving from cloaks belonging to the military unit that built the fort.59 Other off-cuts are in undyed, orange or crimson fabrics in weft-faced tabby and 2/1 twills, 2/2 diamond twill and block damask.60

23Among the Mons Claudianus textiles, 46 fragments of ribbed and block damasks have been recovered and studied by handweaver Martin Ciszuk.61 Eight of them are dark blue ribbed damasks (Fig. 9-10); one is block damask. Most are dated to 100-120 AD, three to about 140 AD.62 The pieces appear quite similar to those from Didymoi and although they are smaller and less distinctive in shape they are likely to derive from dark blue cloaks of the same type. They are slightly later than the blue cloak fragments from Didymoi but this in fact makes them fit better to the dates of the mummy portraits. A dark blue 2/2 broken twill (Fig. 11) probably belongs to the same group.63 Other Mons Claudianus textiles that are likely to derive from cloaks is a weft-faced half-basket with a dark blue band and two rows of twining (Fig. 12),64 a weft-faced tabby with a corded starting border immediately above a green band (Fig. 13),65 a weft-faced twill 2/2 with a red band at right angles (Fig. 14)66 and a number of very densely weft-faced or ‘tight’ twills 2/1 and 2/2 that in several cases show evidence of redone edges (Fig. 15-16). Most of these are undyed or brown wool.67

Fig. 9

Image 1000000000000BB80000085CEF5C7528.jpg

Ribbed damask, MC 745.

© L. Bender Jørgensen

Fig. 10

Image 100000000000049D0000031C4CFF5670.jpg

Ribbed, dark blue damask, MC 815. Possibly cut-off from cloak.

© L. Bender Jørgensen

Fig. 11

Image 100000000000049D0000038BA14BBFB0.jpg

Dark blue 2/2 broken twill, MC 1098. Possibly cut-off from cloak.

© L. Bender Jørgensen

Fig. 12

Image 10000000000003130000049D2AC66E6F.jpg

Weft-faced half-basket weave with dark blue band and two rows of twining, MC 1357. Likely to derive from a cloak.

© L. Bender Jørgensen

Fig. 13

Image 10000000000005C400000282A4466E9A.jpg

Corded starting border immediately above green band, MC 1115. Likely to derive from cloak.

© L. Bender Jørgensen

Fig. 14

Image 100000000000049D000003248A722486.jpg

Weft-faced 2/2 twill with a red band forming a right angle, MC 1625. Likely to derive from cloak.

© L. Bender Jørgensen

Fig. 15

Image 1000000000000376000002A9CA60209D.jpg

Densely woven 2/2 twill with plied warp, MC 1247. The curved hem suggests that it derives from a cloak.

© L. Bender Jørgensen

Fig. 16

Image 1000000000000383000002578F9CBB76.jpg

3/1 block damask, MC 1133. The re-hemmed, partly curved shape suggests that it derives from a cloak.

© L. Bender Jørgensen

24Scarves: Items of garments identified at Didymoi include a number of scarves, several of which have composite checks that have close parallels in four pieces from Mons Claudianus (Fig. 17), all found in the SS and deposited between 100 and 140 AD.68

Fig. 17

Image 10000000000008CA0000069652864FC5.jpg

Wool tabby with composite checks, MC 975. Likely to derive from a scarf.

© L. Bender Jørgensen

25Soft furnishings: The study of soft furnishings in Antiquity is less advanced that that of clothing, although in recent years attempts have been made to identify such textiles in iconographical and textual sources as well as in textiles recovered from excavations or in museum collections.69 Much is still a matter of conjecture. Among the Mons Claudianus textiles are several pieces that do not derive from clothing. Studies of depictions of soft furnishings and other textiles in Roman art have revealed several examples that appeared to have counterparts among the rags from the rubbish heaps –items that are likely to have been carpets, wall hangings, curtains, and mattress or cushion covers. One piece with a thick pile and a meander pattern is likely to have been a rug (Fig. 18); another with an almost worn-away pile ornamented with a bunch of grapes may derive from another rug or perhaps a wall hanging or coverlet (Fig. 19).70 Both derive from the SS; the latter can be dated 100-120 AD. Several pieces show wear signs that indicate they had been used for sitting or reclining, and are likely to be cushion or mattress covers (Fig. 20). The designs of some of these are conspicuously like the mattresses depicted in the lupanars of Pompeii.71 The six taquetés72 are likely to be cushion or couch covers (Fig. 21). They, too were found in the SS, but appear mainly to derive from the later deposits as three are dated c. 140 AD, the others to the slot 100-140 AD. Eight resist-dyed fabrics are all light, ‘dry’ tabbies may well have served as wall-hangings or curtains73 (Fig. 22). Six of these were found in the SS and are mainly dated 100-120 AD;74 two derive from the FSE and are deposited 140-150 AD.75

Fig. 18

Image 1000000000000BB800000894A341A1F9.jpg

Rug with thick pile in a meander pattern, MC 300.

© L. Bender Jørgensen

Fig. 19

Image 10000000000007080000083A47FF8312.jpg

Red wool tabby with almost worn-away pile, MC 964. The pattern is a bunch of grapes.

© L. Bender Jørgensen

Fig. 20

Image 100000000000049D000002C2F25AF895.jpg

Green wool fabric with two red stripes, MC 945. The worn surface suggests that it derives from a cushion or mattress cover.

© L. Bender Jørgensen

Fig. 21

Image 100000000000049D00000414EBB0D911.jpg

Taqueté, MC 968. Taquetés were often used for cushions and mattress covers.

© L. Bender Jørgensen

Fig. 22

Image 100000000000049D0000034697F73537.jpg

Resist-dyed wool fabric, MC 631. It may derive from a curtain or wall hanging.

© L. Bender Jørgensen

26Utility textiles: At a workplace like Mons Claudianus, a number of textiles will have served for utility purposes, such as saddlebags for donkeys and dromedary, sacks, various types of straps, and sunshields. Many of the goat-hair fabrics must be considered to belong to this group; three of solid straps or girths are among the most easily recognisable76 (Fig. 23). Other goat hair fabrics may derive from sleeping mats like those still used by the men of the Egyptian Antiquities service that guarded the store then in Dendera (Fig. 24-25).

Fig. 23

Image 10000000000008CA0000039C1CBBCC55.jpg

Strap or girdle made of goat hair, MC 942.

© L. Bender Jørgensen

Fig. 24

Image 100000000000049D00000313F08307CC.jpg

A large, unrecorded piece of goat hair cloth from Mons Claudianus. Pattern and type is similar to sleeping mats still used (fig. 25) and may have had similar use.

© L. Bender Jørgensen

Fig. 25

Image 10000000000002FD0000049D19B71ABA.jpg

A night watchman’s bedding including a mat very similar to fig. 24, rolled up for the day, photographed 1994.

© L. Bender Jørgensen

Abu Sha’ar

27Abu Sha’ar was founded 309/310 AD as a military fort housing a mounted unit of cavalry or dromedaries, the Ala Nova Maximiana, consisting of 150-200 men (Fig. 26). This phase ended sometime towards the end of the 4th and beginning of the 5th century AD. After a period of abandonment, Christian monks or hermits moved in, staying until the 6th or the beginning of the 7th century AD.77 This means that the site has two phases of very different use. A well c. 1 km W of the fort was also excavated, as well as a hydreuma at Bir ‘Abu Sha’ar al-Qibly. The hydreuma was founded in the 2nd century AD but remained in use and served the late Roman fort.78

Fig. 26

Image 10000000000004B5000005C44FEEBC71.jpg

Plan of the fort at ‘Abu Sha’ar with list of trenches.

© Reproduced by permission from S.E. Sidebotham

28The textiles from ‘Abu Sha’ar, the well and the Bir ‘Abu Sha’ar al-Qibly hydreuma number 1126 pieces, most of which have proved possible to sort into one of the two chronological groups (Tables 7-8). 673 are attributed the 4th century military phase, 272 to the later Christian settlement.79

Table 7

Area

Trench

Key in plan

No of Textiles

% of Phase

Administrative area/Commandants quarter

S, X, X EXT

27, 32

27

4%

Barracks

EE

38

23

3%

Bath

CC, DD

37

37

5%

Mill/oil

T, T WEX

28

8

1%

N of Fort

BB

36

5

0,5%

NE Sebakh

C, F, H, I, J, L, M

17

482

71%

Principia

O, V

23, 30

44

6,5%

Stores

R/Z

26, 34

1

0%

West gate

P

24

18

2,5%

‘Abu Sha’ar Well

ASW A, S

30

4%

Bir ‘Abu Sha’ar Hydreuma

BAS A, B, E

18

2,5%

Total phase 1

673

100%

Textiles from ‘Abu Sha’ar, the ‘Abu Sha’ar Well and Bir ‘Abu Sha’ar al-Qibly dated to phase 1.

Table 8

Area

Trench

Key in plan

No Textiles

% of Phase

Kitchen

N

22

42

15,4%

Kitchen/street

R, R/N, RS Balk

26

26

9,5%

Mill/oil

T, T WEX, TS BEX

28

38

14%

N gate

W

31

6

2,2%

Principia/church

V

30

1

0,4%

Stores

R, Z, AA

26, 34, 35

121

44,5%

Street/stores

Y

33

38

14%

Total phase 2

272

100%

Textiles from ‘Abu Sha’ar dated to phase 2.

Fibres, yarns and weaves

29As at Mons Claudianus, it was not possible to carry out formal fibre identification of the ‘Abu Sha’ar textiles. In addition, the proximity to the Red Sea caused the textiles to be heavily impregnated with saline substances that made it particularly difficult to recognize fibres. The following statistics should therefore be taken with a pinch of salt.

30The textiles appear to fall in two major and two smaller groups: wool, bast (i.e. flax, hemp), goat hair and cotton. 355 fragments from phase 1 or 53% have thus been classified as wool, 242 pieces or 36% as bast (or specifically as flax), 28 pieces or 4% as goat hair and 14 pieces or 2% as cotton or possibly cotton. From phase 2, 75 fragments or 27% are categorised as wool, 143 pieces or 52% as bast, 4 pieces or 1% as goat hair and 21 pieces or 7% as cotton (Fig. 27).

Fig. 27

Image 100000000000079E000005DD5EDE081A.jpg

Cotton basket weave, AS 877.

© M. van Waveren

31Fabrics from phase 1 that are categorised as bast fibres are generally made from of s-twisted yarn in both warp and weft. Goat hair fabrics have Z2s-plied yarn in one or both systems, while ten of fourteen cottons are made of s-twisted, four of z-twisted yarns (Table 9).

Table 9

Fibre

Yarn twist

Numbers

Bast (flax and hemp)

s-s

207

s-ss

2

ss-ss

13

s-z

3

z-z

7

z-zz

2

Goat hair

Z2s-s

2

Z2s-ss

2

Z2s-Z2s

18

Cotton

s-s

8

ss-ss

2

z-z

4

Yarn types in bast fibres, goat hair and cotton recorded in textiles from ‘Abu Sha’ar phase 1.

32Wool fabrics from phase 1 show a greater variety. The largest group of 173 pieces are made of s-twisted yarns in both systems; 50 pieces have z-twisted yarn in both systems, 43 have s-twisted warp and z-twisted weft. Paired or plied yarns in one or both systems also occur. While fabrics in bast fibres, goat hair and cotton all are in tabby, half-basket or basket weave, a significant number of those in wool are twills. The relation between weave and twist of the wool textiles is scrutinised below (Tables 10 and 11).

33In phase 2, fabrics made entirely from s-twisted yarns are even more prominent; among those made of bast fibres, this applies to 138 pieces while two are s-z and five z-z. The woollens show more variation; 60 are made of s-s, s-ss, ss-s or sss-ss twisted yarns, four are z-z or z-zz, six s-z or z-s and two Z2s-s. Two goat hair fabrics are s-s, one z-z, and all cottons but one is s-s or ss-ss.

34Weaves found in textile fragments from phase 1 are mainly tabby or tabby derivate such as half-basket and basket weave; 95 pieces are some kind of twill with 2/2 diagonal twills and 2/2 diamond twills as the largest groups. 15 pieces are felt. Those from phase 2 also have a majority of tabby and tabby derivates. Two pieces are felt, one piece taqueté (Fig. 28) and three tapestry (Fig. 29). Only three are twills, all 2/2 diamond twill.80

Fig. 28

Image 100000000000049D000002F98D038CD0.jpg

Taqueté, AS 642.

© L. Bender Jørgensen

Fig. 29

Image 10000000000003B10000030D89E9275E.jpg

Tapestry in the shape of a cross, AS 649.

© L. Bender Jørgensen

35When combining weave and yarn twist of the wool textiles we find the great majority of the wool tabbies to be made of s-twisted yarns in both systems. They make up 21% of all textiles from phase 1 while wool tabbies s-z and z-z each are about 5%.Wool half-basket and basket weaves from phase 1are also predominantly made of s-twisted yarns in both systems; only one basket weave is made solely from z-twisted yarn, one half-basket weave from mixed yarns (Table 10).

Table 10

Weave

Yarn type

Number

Half-basket

s-ss

12

s-zz,z

1

Z2s-ss

3

S2z-ss

1

Basket weave

ss-ss

3

zz-zz

1

Wool fabrics in half-basket or basket weave from ‘Abu Sha’ar phase 1.

36The twills from ‘Abu Sha’ar phase 1 are a mixed lot (Table 11). Compared to those from Mons Claudianus, emphasis has changed from z-z to s-s in diagonal twills 2/2. Among the twill variants further changes are indicated by a range of combinations of single, paired and plied yarns.

Table 11

Weave

Spin

Number of textiles

2/1 twill diagonal

s-z

1

2/2 twill diagonal

s-s

27

s-z, s-zz, z-s

4

z-z

8

2/1 diamond twill

s-zz

1

2/1 broken or diamond twill

Z2s-s or Z2s-ss

2

2/2 broken, chevron or diamond twill

s-s or s-ss

3

s-z or s-zz

18

z-z

8

S2z-z or S2z-zz

10

Z2s-s or Z2s-ss

6

Z2s-z

2

Yarn twist in wool twills from ‘Abu Sha’ar phase 1.

37When sett and density in the form of thread counts are added to the data on weaves and yarns it becomes obvious that they consist of several groups. Diagonal twills 2/1 and 2/2 are balanced and have thread counts between 5 and 15 threads/cm, i.e. coarse to medium qualities roughly corresponding to modern tweeds. Broken, chevron or diamond twills appear in at least four varieties. Those with s-twisted warp and paired weft have slightly higher thread counts compared to the plain diagonal twills (paired weft threads count as 1 weft) but remain balanced or slightly weft-faced, except for a single piece that is densely packed. Those made of single z-twisted yarns in both systems show a wide variety, from balanced fabrics of medium density to some with very densely packed weft. Broken/diamond twills with plied warp and single or paired weft also show great diversity, ranging from balanced, coarse to medium fabrics to some with high weft density. Those with Z-plied warp tend to have lower warp counts than those with S-plied warp, while the latter are more balanced (see Fig. 39).

38Among the textiles from phase 2 almost all bast tabbies, half-basket and basket weaves are made of s-twisted yarns in warp as well as weft; only five tabbies have z-z twisted yarns; among the woollen fabrics, 54 tabbies, all the half-basket and basket weaves and the taqueté are s-s; one tabby is s-z and another z-z while two of the twills are z-zz and one z-s. The goat hair tabbies consist as mentioned above of two made of s-s yarns and one z-z and the cottons are all s-s except for the knotted net that is made from z-twisted yarn. Two tapestries are s-s, one Z2s-s.

39The ‘Abu Sha’ar textiles did not include any complete or immediately recognisable garments, although several pieces undoubtedly are remains of clothing (Fig. 30-33). Two tapestries are likely to derive from tunics; one belongs to the large piece depicted in Fig 31, the one in Fig. 33 probably derives from the neckline of a tunic. Among other interesting pieces three embroideries can be mentioned, two of them on small pieces of red felt (Figs. 34-35). A group of resist dyed cotton fragments dated to phase 1 have close parallels at Berenike and Myos Hormos and are likely to be of Indian origin81 (Fig. 36).

Fig. 30

Image 10000000000003000000049D0CCEFCAF.jpg

Densely woven half-basket wool fabric, AS 073. The shape, cut and stitching suggest that it derives from a garment.

© L. Bender Jørgensen

Fig. 31

Image 100000000000049D00000313A70EF47F.jpg

Large pieces of linen tabby with bands in weft-float pattern and one small tapestry motif, AS 400. It probably derives from a tunic or other garment.

© L. Bender Jørgensen

Fig. 32

Image 100000000000049D000002E3F825F55F.jpg

Three pieces of weft-faced wool tabby sewn together, in part by joining the selvedges of two pieces, AS 018. Likely to derive from garment.

© L. Bender Jørgensen

Fig. 33

Image 100000000000049D000002E80907C267.jpg

Fragments of tapestry, AS 699. Likely to derive from neck line of under-tunic.

© L. Bender Jørgensen

Fig. 34

Image 10000000000008CA00000601781F2ADB.jpg

Embroidery on a piece of red felt, AS 006.

© L. Bender Jørgensen

Fig. 35

Image 10000000000008CA00000696905E9D36.jpg

Embroidery, red, green, blue and yellow wool on a tabby in bast fibres, AS 036.

© L. Bender Jørgensen

Fig. 36

Image 100000000000049D000003E9E9B3D9ED.jpg

Resist-dyed cotton, AS 708.

© L. Bender Jørgensen

Comparing ‘Abu Sha’ar and Mons Claudianus

40To some extent the ‘Abu Sha’ar textiles resemble those from Mons Claudianus, but they also differ in many ways. One of the major differences is in fibres. At Mons Claudianus, about 90% of the textiles are of wool while flax and other vegetable fibres only are found in about 5%. At ‘Abu Sha’ar, vegetable fibres are much more common, and include several pieces of cotton. In the early, military phase of ‘Abu Sha’ar, wool is the most common fibre, but in the later, Christian settlement, this has been reversed.

41Another difference is in the twills. At both sites, these fall into several groups. Apart from a single piece of ribbed damask found in a top layer at the NE Sebakh at ‘Abu Sha’ar (Fig. 37), damasks are only found at Mons Claudianus. Both sites have groups of largely balanced twills, with approximately the same amount of threads/cm in warp and weft, and of distinctly weft-faced twills, many with very densely packed weft (Fig. 38-39). At Mons Claudianus, the weft-faced ‘tight’ twills are always plain diagonal twills 2/1 or 2/2, while those from ‘Abu Sha’ar are 2/2 chevron, herringbone or diamond twills (Fig. 40). Almost all the Mons Claudianus twills with plied warp can be termed ‘tight’. When twills from the selected sample are included, ‘tight twills’ make up more than 20% of all Mons Claudianus twills, and seem to increase towards the middle and late part of the 2nd century.82

Fig. 37

Image 10000000000007D000000A8C7D0E1BA8.jpg

Ribbed 2/1 twill, AS 309.

© L. Bender Jørgensen

Fig. 38

Image 10000000000003130000035B15FB3360.jpg

Diagram of twills from Mons Claudianus.

© L. Bender Jørgensen

Fig. 39

Image 10000000000003130000033A1E671734.jpg

Diagram of twills from ‘Abu Sha’ar.

© L. Bender Jørgensen

Fig. 40

Image 100000000000049D0000034CFCB992C4.jpg

Three pieces of very densely, ‘tight’ diamond twills, AS 050, AS 099 and AS 200.

© L. Bender Jørgensen

42At both sites, the most common selvedges are either simple ones, where the weft just turns into the next shed without further ado, or reinforced selvedges created by wrapping the weft around groups or bundles of the outermost warp threads. At ‘Abu Sha’ar, this is usually done over two bundles, at Mons Claudianus three. Details like this are likely to reflect workshop preferences. End borders are usually made by twisting the remaining warp threads into a cord (Fig. 41-42).

Fig. 41

Image 100000000000049D00000313E893FF3C.jpg

Strip torn off the corner of a fabric consisting of a corded end border as well as a reinforced selvedge, AS 214.

© L. Bender Jørgensen

Fig. 42

Image 1000000000000313000001E2E748B5F1.jpg

Selvedges and transverse borders at Mons Claudianus (random sample only) and ‘Abu Sha’ar phases 1 and 2. Simple selvedges, reinforced selvedges over 2 and 3 bundles of warp threads resåectively, and corded transverse borders.

© L. Bender Jørgensen

Comparing sites from the Eastern Desert

43Comparing the textiles from these two sites with those from the other recently excavated sites in the Eastern Desert offer some intriguing possibilities. The sites fall in five different categories: quarries, ports, praesidia, a fort and a monastic settlement (Table 12).

Table 12

Site

Quarry

Port

Praesidia

Fort

Monastic settlement

‘Abu Sha’ar 1

x

‘Abu Sha’ar 2

X

Berenike

X

Didymoi

X

Dios

X

Krokodilô

X

Maximianon

X

Mons Claudianus

X

The Porphyrites

X

Myos Hormos

X

Xeron Pelagos

X

Types of Roman sites with textile remains in the Eastern Desert.

44They, of course, had different types of populations.83 The quarries were worked by stone masons and other artisans along with imperial slaves and freedmen labelled familia; in the ports, sailors and merchants mingled with a variety of service trades. Soldiers were stationed in quarries and ports as well as in the praesidia and the fort. Women –wives and prostitutes– and children were also part of the population at most of the sites although presumably in limited numbers. Only in the late phase at ‘Abu Sha’ar we encounter a different group of inhabitants: Christian settlers and monks.

45Most of the sites are dated within the 1st and 2nd centuries; the 4th century, and the 5th-7th centuries are represented at ‘Abu Sha’ar. The Porphyrites and Berenike also cover several periods. The variety of sites makes it possible to ask several questions. One of them deals with fibres. Wool is significantly more common at Mons Claudianus, Maximianon and Krokodilo than at ’Abu Sha’ar, Myos Hormos and the Porphyrites (Fig. 43). Several alternative explanations have been investigated. A chronological development seems excluded as Myos Hormos is contemporary with Mons Claudianus and the two praesidia. The Porphyrites cover a longer period, and both phases of ’Abu Sha’ar are later. Differential survival, favouring the preservation of wool fabrics at the desert sites is a possibility, but as argued in an earlier paper I prefer to explain the difference as due to climate.84 ’Abu Sha’ar and Myos Hormos are both situated on the Red Sea coast. Evaporation from the sea is strong, causing the air to be humid, contrary to the dry air of the desert. That makes you sweat. Clothes made from vegetable fibres absorb sweat. In the desert, absorbing clothing is unnecessary, as sweat evaporates before it reaches the surface of the skin. Instead, insulation against dehydration is vital. Temperature changes are violent in the desert, and warm clothes are certainly needed for cold nights and mornings, particularly in the winter. Wool is a fibre that is eminently suited for both these purposes. The textiles from the Porphyrites do however not fit this explanation, as the percentage of wool here is on the same level as the coastal sites. Fiona Handley has argued that the fabrics of bast fibres that make up 26% of the textiles from the Porphyrites were little suited for clothing and instead represent textiles used in industrial contexts in the quarry.85

Fig. 43

Image 100000000000031300000228E0C024DD.jpg

Fibre preferences at the praesidia Maximianon and Krokodilô, the quarries Mons Claudianus and the Porphyrites, the port Myos Hormos, the fort ‘Abu Sha’ar 1 and the Christian settlement ‘Abu Sha’ar 2.

© L. Bender Jørgensen

46Another interesting aspect is the occurrence of twills. It is noteworthy that twills are much more frequent at military installations, regardless their date (Fig. 44). At ‘Abu Sha’ar 1, and the praesidia, military personnel were the predominant population group. With the Monastic phase, twills virtually disappear from the site –the three pieces allocated to phase 2 may be re-deposited and originally derive from phase 1. Soldiers administered Mons Claudianus and Myos Hormos, and formed an important although relatively small section of its population. This indicates that twills were used for military garbs rather than civilian. Again, the Porphyrites does not quite fit; according to Fiona Handley, only 3% of the textiles are twills.86

Fig. 44

Image 1000000000000313000001DBFF85C7A7.jpg

The proportions of the main weaves (bast and wool tabby, and twills) at Krokodilô, Maximianon, Mons Claudianus, Myos Hormos, and ‘Abu Sha’ar phases 1 and 2.

© L. Bender Jørgensen

Rags and riches

47How do the rags from Mons Claudianus and ‘Abu Sha’ar compare with the textile riches in the form of elaborated decorated garments and soft furnishings from the Roman-Egyptian town Antinoë87 and the so-called Coptic textiles that can be seen in museums around the world?88 As we have seen, we have few recognizable garments, and merely small fragments that offer glimpses of the rich and colourful decoration that appears commonplace in burials at Antinoë and other sites. We have no silks; we have a few fragments of tapestry at ’Abu Sha’ar, all from phase 2, which means that they are dated to after AD 400. None were found at Mons Claudianus. We do, on the other hand, have scraps of resist dyed textiles, piled fabrics, taquetés and other patterned textiles –fabrics that are known in greater splendour, and much larger quantities from Antinoë and other sites. This shows us that officers and other dignitaries at Mons Claudianus and ‘Abu Sha’ar are likely to have had taqueté cushions and mattresses, resist-dyed curtains, piled rugs and other vestiges of civilization.

48So what, then, can we learn from the textiles from Mons Claudianus and ‘Abu Sha’ar, and indeed from the recent textile bonanza of the Eastern Desert of Egypt? Their archaeological context gives us a chronological framework, enabling us to sort them into three phases –the 1-2nd centuries, the 4th century and the 5th-early 7th century. We can identify a growing number of items of clothing and soft furnishings. And we may begin to divide them between different groups of owners: Roman military, various civilian groups, and Christian monks. This supplies us a hitherto unknown foundation for building theories and substantiate arguments –in short, a wonderful new toolbox for further studies of the textiles from Roman Egypt and indeed the Roman world.

Acknowledgements

49Thanks are due to the organisers and the Collège de France for inviting me to the conference on the Eastern Desert and publishing the proceedings. Further thanks are owed to the directors and team members of the Mons Claudianus and ‘Abu Sha’ar projects, in particular Adam Bülow-Jacobsen and Steven Sidebotham who invited me to join the projects; to members of the Mons Claudianus Textile Project Martin Ciszuk, Lena Hammarlund and Ulla Mannering; and to fellow scholars of Roman textiles in Egypt Dominique Cardon, Hero Granger-Taylor, Fiona Handley, Gillian Vogelsang-Eastwood and Felicity and John Peter Wild who all have contributed by sharing their knowledge.

50Work on the Mons Claudianus textiles has been generously funded over the years by the British Academy, the Carlsberg Foundation, G.E.C. Gad’s Fond, Agnes Geijer’s Foundation for Nordic Textile Research, the Joint Committee of the Nordic Research Councils for the Humanities (NOS-H), Novo’s Fond and VKR’s Familiefond. A debt of gratitude are owed to all.

Bibliographie

  

Bender Jørgensen L. 1990. “Mons Claudianus”. Archaeological Textiles Newletter, No. 10, pp. 10-11.

Bender Jørgensen L. 1991a. “Textiles from Mons Claudianus. A Preliminary Report”. Acta Hyperborea 3, Copenhagen, pp. 83-95.

Bender Jørgensen L. 1991b. The Textiles from Mons Claudianus, Recorded in 1991. Archaeological Textiles Newletter, No. 12, pp. 8-10

Bender Jørgensen L. 1992. North European Textiles until AD 1000. Aarhus, Aarhus University Press.

Bender Jørgensen L. 2000. “The Mons Claudianus Textile Project”. In Archéologie des textiles des origines au ve siècle. Actes du colloque de Lattes, oct. 1999. D. Cardon, M. Feugère (eds), Montagnac, Éditions Monique Mergoil, pp. 253-264.

Bender Jørgensen L. 2004a. “Team work on Roman textiles. The Mons Claudianus Textile Project”. In PURPUREAE VESTES. Textiles y tintes del Mediterráneo en época romana. C. Alfaro, J.P. Wild, B. Costa (eds), Valencia, PUV, pp. 69-76.

Bender Jørgensen L. 2004b. “A matter of material. changes in textiles from Roman sites in Egypt’s Eastern Desert”. Antiquité Tardive 12, pp. 87-99.

Bender Jørgensen L. 2006a. “The Late Roman Fort at ‘Abu Sha’ar, Egypt. textiles in their archaeological context”. Riggisberger Berichte 13, pp. 161-74.

Bender Jørgensen L. 2006b. “A Terminology of Roman Stripes, Bands and Checks”. Archaeological Textiles Newsletter No. 43, pp. 36-37.

Bender Jørgensen L. 2007. Dated textiles from Mons Claudianus and 'Abu Sha'ar. In Methods of dating ancient textiles of the 1st millennium AD from Egypt and neighbouring countries. A. De Moor and C. Fluck (eds), Tielt, Lannoo publishers, pp. 26-35.

Bender Jørgensen L. 2008. Self-bands and other subtle patterns in Roman textiles. In PURPUREAE VESTES II. Vestidos, textiles y tintes. Estudios sobre la producción de bienes de consume en la Antigüedad. C. Alfaro, L. Karali (eds), Valencia, PUV, pp. 135-142.

Bender Jørgensen L. 2011. Clavi and non-clavi. definitions of various bands on Roman textiles. In Purpureae Vestes III. Textiles y tintes en la ciudad antigua. C. Alfaro, J.-P. Brun, Ph. Borgard and R. Pierobon Benoit (eds), Valencia/Naples. PUV/CJB, pp. 75-82.

Bender Jørgensen L. 2015. “Representing Colours”, SAITIBA revista de la facultat de geografia i historia 64-65, 2014-2015, pp. 53-62.

Bender Jørgensen L. 2017. “Textiles and Textile Trade in the First Millennium AD - Evidence from Egypt”. In Trade in the Ancient Sahara and Beyond. D.J. Mattingly, V. Leitch, C.N. Duckworth, A. Cuenod, M. Sterry, F. Cole (eds), Trans-Saharan Archaeology Volume I. Series editor D.J. Mattingly. Cambridge. Cambridge University Press and The Society for Libyan Studies.

Bender Jørgensen L., Mannering U. 2001. “Mons Claudianus. Investigating Roman textiles in the Desert. In The Roman Textile Industry and its Influence. A Birthday Tribute to John Peter Wild. P. Walton Rogers, L. Bender Jørgensen, A. Rast-Eicher (Ed), Oxford. Oxbow Books, pp. 1-11.

Bergman I. 1975. Late Nubian Textiles. Lund, The Scandinavian Joint Expedition to Sudanese Nubia.

Bingen J., Bülow-Jacobsen A., Cockle W.E.H., Cuvigny H., Rubinstein L., Van Rengen W. 1992. Mons Claudianus. Ostraca Graeca et Latina I (O. Claud. 1 à 190). Cairo. Institut français d’archéologie orientale du Caire.

Bingen J., Bülow-Jacobsen A., Cockle W.E.H., Cuvigny H., Kayser F., and Van Rengen W. 1997. Mons Claudianus. Ostraca Graeca et Latina II (O. Claud. 191 à 416). Cairo. Institut français d’archéologie orientale du Caire.

Bülow-Jacobsen A. 2009. Mons Claudianus. Ostraca Graeca et Latina IV. The Quarry Texts O. Claud. 632 à 896). Cairo. Institut français d’archéologie orientale du Caire.

Burnham D.K. 1992. “Coptic Knitting. An Ancient Technique”. Textile History 3, pp. 116-124.

Calament F., Durand M. (eds). 2013. Antinoé à la vie, à la mode. Visions d’élégance dans les solitudes. Lyon, Musée des Tissus.

Cardon D. 2003. “Chiffons dans le désert. textiles des dépotoirs de Maximianon et Krokodilô”. In La route de Myos Hormos. L’armée romaine dans le désert Oriental d’Égypte. H. Cuvigny (ed.), Volume 2. Cairo, Institut français d’archéologie orientale, pp. 619-659.

Cardon D., Cuvigny H., Bülow-Jacobsen A. 2010. “Recent textile finds from Dios and Xeron”. Archaeological Textiles Newsletter 50, pp. 2-13.

Cardon D., Granger-Taylor H., Nowik W. 2011. What did they look like? Fragments of Clothing Found at Didymoi. case studies”. In Didymoi. Une garnison romaine dans le désert Oriental d’Égypte. I – Les fouilles et le matériel. H. Cuvigny (ed), Fouilles de l’Ifao 64. Cairo, Institut français d’archéologie orientale, pp. 273-362.

Ciszuk M. 2000. “Taquetés from Mons Claudianus – Analyses and Reconstruction”. In Archéologie des textiles des origines au ve siècle. Actes du colloque de Lattes, oct. 1999. D. Cardon, M. Feugère (eds), Montagnac, Éditions Monique Mergoil, pp. 265-282.

Ciszuk M. 2004 “Taqueté and damask from Mons Claudianus. a discussion of Roman looms for patterned textiles”. In PURPUREAE VESTES. Textiles y tintes del Mediterráneo en época romana. C. Alfaro, Wild, J.P., Costa, B. (eds), Valencia, PUV, pp. 107-114.

Ciszuk M., Hammarlund L. 2008. “Roman looms – a study of craftsmanship and technology in the Mons Claudianus Textile Project”. In PURPUREAE VESTES II. Vestidos, textiles y tintes. Estudios sobre la producción de bienes de consume en la Antigüedad. C. Alfaro and L. Karali (eds),Valencia, PUV, pp. 119-133.

Croom A. 2000. Roman Clothing and Fashion. Stroud, Gloucestershire, TempusPublishing Ltd.

Cuvigny H. 2000. Mons Claudianus. Ostraca Graeca et Latina III (O. Claud. 417 à 631). Cairo. Institut français d’archéologie orientale du Caire.

Cuvigny H. 2005. “ L'organigramme du personnel d'une carrière impériale d'après un ostracon du Mons Claudianus”. Chiron, 35, pp. 309-353.

Cuvigny H. 2016. “Travailler pour l'empereur”. Les nouvelles de l'archéologie [En ligne], 143 | 2016, mis en ligne le 12 mai 2016, consulté le 07 juin 2016. URL: http://nda.revues.org/3305. DOI. 10.4000/nda.3305.

De Moor A., Verhecken-Lammens C., Verhecken A. 2008. 3500 years of textile art, Tielt. Lannoo.

De Moor A. and Fluck C. (eds). 2009. Clothing the House. Furnishing textiles of the 1st millennium AD from Egypt and neighbouring countries, Tielt. Lannoo.

Eastwood G.M., 1982. Textiles. In Quseir al-Qadim 1980. Preliminary Report. D. Whitcomb & J. Johnson (eds), American Research Center in Egypt, pp. 285-326.

Erikson M. 1997. Textiles in Egypt 200-1500 A.D. in Swedish Museum Collections, Göteborg, Röhsska Museet.

Fluck C., Linscheid P., Merz S. (eds). 2000. Textilien aus Ägypten. Teil 1. Textilien aus dem Vorbesitz von Theodor Graf, Carl Schmidt und dem Ägyptischen Museum Berlin, Wiesbaden, Reichert Verlag.

Granger-Taylor H. 1982. “Weaving Clothes to Shape in the Ancient World. The Tunic and Toga of the Arringatore”. Textile History 13, pp. 3-25.

Granger-Taylor H. 1987. “The Emperor’s Clothes. The Fold Lines”. Bulletin of the Cleveland Museum of Art 74, No. 3, pp. 114-123.

Granger-Taylor H. 2000. “The Textiles from Khirbet Qazone (Jordan)”. In Archéologie des textiles des origines au ve siècle. Actes du colloque de Lattes, oct. 1999. D. Cardon and M. Feugère (eds), Montagnac, Éditions Monique Mergoil, pp. 149-162.

Granger-Taylor H. 2007. “‘Weaving Clothes to Shape in the Ancient World’ 25 years on. Corrections and Further Details with Particular Reference to the Cloaks from Lahun”. Archaeological Textiles Newsletter No. 45, pp. 26-35.

Granger-Taylor H. 2008. “A fragmentary Roman cloak probably of the 1st c. CE and cut-offs from other semicircular cloaks”. Archaeological Textiles Newsletter No 46, pp. 6-16.

Hadian M., Good I., Pollard A.M., Zhang X., Laursen R. 2012. Textiles from Douzlakh salt mine at Chehr Abad, Iran. A technical and contextual study of late pre-Islamic Iranian textiles”. International Journal of the Humanities 19.3, pp. 152-73.

Hammarlund L. 2005. Handicraft knowledge applied to archaeological textiles”. The Nordic Textile Journal, pp. 87-119.

Hammarlund L. 2013. “Visual analysis and grouping of the Hallstatt textiles”. In Textiles from Hallstatt. Weaving Culture in Bronze Age and Iron Age Salt Mines. K. Grömer, A.  Kern, H. Reschreiter and H. Rösel-Mautendorfer (eds), Budapest, Archaeolingua, pp. 179-182.

Hammarlund L., Kirjavainen H., Petersen K.V., M. Vedeler. 2008. “Visual textiles. A Study of Appearance and Visual Impression in Archaeological Textiles”. In Medieval Clothing and Textiles 4. R. Netherton and G.R. Owen-Crocker (eds), Suffolk, The Boydell Press, pp. 69-98.

Handley F. 2000. The Roman textiles from Myos Hormos. Archaeological Textiles Newsletter 31, pp. 12-17.

Handley F. 2007. Textiles”. In The Roman Imperial Quarries. Survey and Excavation at Mons Porphyrites 1994-1998. D. Peacock and V. Maxfield (eds), Volume 2. The Excavations, London, Egypt Exploration Society, pp. 355-378.

Handley F. 2011. The textiles. A preliminary report. In Myos Hormos – Quseir al-Qadim Roman and Islamic Ports on the Red Sea. Volume 2. Finds from the excavations 1999-2003, D. Peacock, and L. Blue (eds), Oxford, Archaeopress, pp. 321-334.

Henshall A.S. 1984. A fragment of tufted cloth”. In Ghirza. A Libyan Settlement in the Roman Period. O. Brogan, D.  Smith (eds), Libyan Antiquities Series I, Tripoli, Department of Antiquities, 93, Pl. 33a.

Kemp B. J., Vogelsang-Eastwood G. 2001. The Ancient Textile Industry at Amarna. London, Egypt Exploration Society.

Mannering U. 2000a. “Roman garments from Mons Claudianus”. In Archéologie des textiles des origines au ve siècle. Actes du colloque de Lattes, oct. 1999. D. Cardon, M. Feugère (eds), Montagnac, Éditions Monique Mergoil, pp. 283-290.

Mannering U. 2000b. “The Roman Tradition of Weaving and Sewing. A guide to function”. Archaeological Textiles Newsletter 30, pp. 10-16.

Mannering 2006. Mannering, U., Questions and answers on textiles and their find spots. The Mons Claudianus Textile Project”. Riggisberger Berichte 13, pp. 149-160.

Mannering U. Forthcoming. Garments from Mons Claudianus”. In Survey and Excavation Mons Claudianus 1987-1993. Volume IV. The Textiles. L. Bender Jørgensen (ed), Cairo. Institut français d’archéologie orientale du Caire.

Maspero A., Bruni S., Cattaneo C., Lovisolo A. 2002. “Textiles and leather. Raw materials and manufacture. In Sand, Stones, and Bones. The Archaeology of Death in the Wâdi Tanezzuft Valley (5000-2000 BP). S. di Lernia, G. Manzi (eds). The Archaeology of Libyan Sahara Vol. I, Arid Zone Archaeology Monographs 3, Florence, All’Insegna del Giglio, pp. 157-168.

Maxfield V.A., Peacock D.P.S. 2001. Survey and Excavation Mons Claudianus 1987-1993. Volume II. Excavations . Part 1. Cairo. Institut français d’archéologie orientale du Caire.

Maxfield V.A., Peacock D.P.S. 2006. Survey and Excavation Mons Claudianus 1987-1993. Volume III. Ceramic Vessels & Related Objects, Cairo, Institut français d’archéologie orientale du Caire.

Möller-Wiering S. 2011. War and Worship. Textiles from 3rd to 4th-century AD Weapon Deposits in Denmark and Northern Germany, Oxford and Oakville, Oxbow Books, 2011.

Palme B., Zdiarsky A., Palme B., Zdiarsky A. (eds). 2012. Gewebte Geschichte. Stoffe und Papyri aus dem spätantike Ägypten, Wien, Phoibos Verlag.

Peacock D. and Blue L. (eds). 2011. Myos Hormos – Quseir al-Qadim Roman and Islamic Ports on the Red Sea. Volume 2. Finds from the excavations 1999-2003, Oxford, Archaeopress.

Peacock D., Maxfield V. 1997. Survey and Excavation Mons Claudianus 1987-1993. Volume I. Topography & Quarries, Cairo, Institut français d’archéologie orientale du Caire.

Peacock D., Maxfield V. (eds). 2007. The Roman Imperial Quarries. Survey and Excavation at Mons Porphyrites 1994-1998, Volume 2. The Excavations, London, Egypt Exploration Society.

Pfister R. 1937. Nouveaux textiles de Palmyre, Paris, Les Editions d’Art et d’Histoire.

Pritchard F. 2006. Dress in Egypt in the First Millennium AD. Clothing from Egypt in the collection of The Whitworth Art Gallery, The University of Manchester, Manchester, The Whitworth Art Gallery.

Ripoll G. 2004. “Los tejidos en la arquitectura de la antigüedad tardía. Una primera aproxmación a su uso y función”. Antiquité Tardive 12, pp. 169-182.

Rutschowscaya M.H. 1990. Tissus coptes, Paris, Éditions Adam Biro.

Schmidt-Colinet A., Stauffer A., Al-Asad K. 2000. Die Textilien aus Palmyra, Mainz am Rhein, Verlag Philipp von Zabern.

Sidebotham S.E. 1993. “University of Delaware Archaeological Project at ‘Abu Sha’ar. The 1992 Season”. Newsletter of the American Research Center in Egypt 161/162 (Spring/Summer 1993), pp. 1-9.

Sidebotham S.E. 1994a. “Preliminary Report on thr 1990-1991 Seasons of Fieldwork at ‘Abu Sha’ar (Red Sea Coast)”. Journal of the American Research Center in Egypt 31, pp. 133-158.

Sidebotham S.E. 1994b. “University of Delaware Fieldwork in the Eastern Desert of Egypt, 1993”. Dumbarton Oaks Papers 48, pp. 263-275.

Sidebotham S.E., Hense M., Nouwens H.M. 2008. The Red Land. The Illustrated Archaeology of Egypt’s Eastern Desert, Cairo and New York, The American University in Cairo Press.

Szymaszek M. 2015. “On the interpretation of textile finds with right-angled or H-shaped tapestry bands”. In Textiles, tools and techniques of the 1st millennium AD from Egypt and neighbouring countries. A. de Moor, C. Fluck and P. Linscheid (eds) Tielt, Lannoo Publishers, pp. 169-175.

Thomas T.K. 2001. Textiles from Karanis, Egypt in the Kelsey Museum of Archaeology. Artifacts of Everyday Life, Ann Arbor, The Kelsey Museum of Archaeology.

Thurman C.C.M. and Williams B. 1979. Ancient Textiles from Nubia. Meroitic, X-group, and Christian Fabrics from Ballana and Qustul, Chicago, The Art Institute of Chicago and The University of Chicago.

Van‘t Hooft Ph.P.M., Raven M.J., van Rooij E.H.C., Vogelsang-Eastwood G.M. 1994. Pharaonic and Early Medieval Egyptian Textiles, Collections of the National Museum of Antiquities at Leiden VIII, Leiden, Rijksmuseum van Oudheden.

Viklund U. 2007. Towards a Cosmology of Colours – Roman Textiles from Mons Claudianus. A methodology of approaching meaning of colours in ancient textiles. Unpublished MA thesis, Department of Archaeology and Religious Studies, Norwegian University of Science and Technology, Trondheim.

Vogelsang-Eastwood G.M. 2006. Quseir al-Qadim Textile Report 1978-1982, CD version. Leiden, Textile Research Center.

Walton P. and Eastwood G. 1988. A brief guide to the cataloguing of archaeological textiles, London, Institute of Archaeology Publications.

Wild J.P. 1970. Textile Manufacture in the Northern Roman Provinces, Cambridge, Cambridge University Press.

Wild J.P. 1977. The Textiles. Vindolanda III. Hexham, The Vindolanda Trust.

Wild J.P. 1984. The textiles from Building 32”. In Ghirza. A Libyan Settlement in the Roman Period. O. Brogan, D. Smith (eds), Libyan Antiquities Series I, Tripoli, Department of Antiquities, pp. 291-308, Pl. 151-72.

Wild J.P. 2010. Textiles from Zinkekra”. In The Archaeology of Fazzān. Volume 3; Excavations of C.M. Daniels. D.J. Mattingly (ed.), London and Tripoli, Society of Libyan Studies and Department of Antiquities, Tripoli, pp. 486-488.

Wild J.P. and Wild F.C. 2014. Through Roman eyes. cotton textiles from Early Historic India. In A Stitch in Time. Essays in Honour of Lise Bender Jørgensen. S. Bergerbrant and S.H. Fossøy (eds), Gothenburg, Gothenburg University, pp. 209-235.

Wild J.-P. and Wild F.C. 2018. “Textile Contrasts at Berenike. In The Eastern Desert of Egypt during the Greco-Roman Period: Archaeological Reports. J.-P. Brun, T. Faucher, B. Redon and S. Sidebotham (ed.). On line: https://books.openedition.org/cdf/5254.

Yadin Y. 1963. The Finds from the Bar-Kokhba Period in the Cave of Letters, Jerusalem, The Israel Exploration Society.

Notes de fin

1 Bingen et al. 1992, 1997, Cuvigny 2000, Bülow-Jacobsen 2009; Maxfield, Peacock 2001, 2006; Peacock, Maxfield 1997.

2 Bender Jørgensen, Mannering 2001. Metal boxes (cantines) MC 1987/3; 1988/4; 1989/17, 18, 19, 20, and Varia; 1990/4, 5; 1991/5; 1992/20 and 1993/1.

3 See Bender Jørgensen 2004a, p. 69 for the full composition of the Mons Claudianus textile team, and Bender Jørgensen 2004b for an overview of results. Survey and Excavation Mons Claudianus 1987-1993 volume IV: Textiles and clothing is in preparation and will contain contributions by all members of the textile team. Preliminary reports and studies can be found in Bender Jørgensen 1990, 1991a, 1991b, 2000, 2004a, 2004b, 2006b, 2007, 2008, 2011, 2015; Bender Jørgensen, Mannering 2001; Ciszuk 2000, 2004; Ciszuk, Hammarlund 2008; Mannering 2000a, 2000b, 2006, in prep; Viklund 2007 for preliminary studies. See Bender Jørgensen 2004a, p. 69 for the full composition of the Mons Claudianus textile team, and Bender Jørgensen 2004b for an overview of results. Survey and Excavation Mons Claudianus 1987-1993 volume IV: Textiles and clothing is in preparation and will contain contributions by all members of the textile team.

4 Sidebotham 1993, 1994a, 1994b.

5 Bender Jørgensen 2006a, 2007.

6 Wild, Wild 2018.

7 Cardon 2003.

8 Cardon, Cuvigny, Bülow-Jacobsen 2010.

9 Personal communication, D. Cardon; it is less than a third of the body of textiles found. See Cardon, Granger-Taylor, Nowik 2011 for detailed description and discussion of c. 100 pieces.

10 Eastwood 1982; Handley 2000, 2011; Vogelsang-Eastwood 2006.

11 Handley 2007. The site is commonly known as Mons Porphyrites. As shown in H. Cuvigny’s paper (this volume) the ancient name was the Porphyrites.

12 See Bender Jørgensen, Mannering 2001 for a more detailed description of our work process including selection and criteria for the two groups. MC textiles Nos. 1-468 are stored in Box 1990/4; this box also contains larger pieces of the series 1116-1460. MC Nos. 469-904 in Box 1991/V. MC Nos. 905-115 in Box 1989/20; this box also contains garments, gammas, and other unusual pieces from all series. MC Nos. 1116-1460 are kept in Box 1989/18. MC Nos. 1461-1722 and a collection from Barud (Nos. 1723-51) are kept in Box 1992/20. MC Nos. 1752-63 are re-labelled pieces, mainly kept in Box 1989/20.

13 For maps of the site and location of the excavated area see Maxfield, Peacock 2001, fig. 1.4, and Maxfield (see http://www.college-de-france.fr/site/jean-pierre-brun/symposium-2016-03-30-14h30.htm).

14 For phasing and dating of Mons Claudianus see Maxfield, Peacock 2001, pp. 423-428.

15 Mannering 2000a, p. 283; see also Cardon, Granger-Taylor, Nowik 2011, pp. 276-278.

16 Cf. Walton, Eastwood 1988.

17 The presence of a few cottons is inferred after having seen (and felt) cotton textiles from ‘Abu Sha’ar. Unfortunately it was not possible to carry out fibre analysis at either site.

18 Bender Jørgensen 1992, pp. 120-152.

19 Kemp, Vogelsang-Eastwood 2001, pp. 58-60; Van ‘t Hooft et al. 1994, p. 14.

20 Bergman 1975; Thurman, Williams 1979.

21 Bender Jørgensen 2017; Henshall 1984; Maspero et al. 2002; Wild 1984, 2010, pp. 486-488.

22 Schmidt-Colinet, Stauffer, Al-Asad 2000; Yadin 1963.

23 Pfister 1937; Wild 1970, p. 44.

24 Wild, Wild 2014; see also Hadian et al. 2012 for recent Iranian examples.

25 Bender Jørgensen 1992, pp. 126-150.

26 For definitions of weaves see Bender Jørgensen 2004b, fig. 2 and Ciszuk 2004.

27 Bender Jørgensen 1992; Möller-Wiering 2011; Wild 1977.

28 Bender Jørgensen 2004b, p. 95; Hammarlund 2005, p. 110.

29 Hammarlund 2005, accessible at http://www.hb.se/PageFiles/3251/Filer/ctfjournal_2005.pdf#page=88

30 Hammarlund et al. 2008; Hammarlund 2013.

31 Bender Jørgensen 2006b, 2007, 2008, 2011.

32 Ciszuk 2000, 2004

33 Ciszuk, Hammarlund 2008.

34 Mannering 2000a, 2006.

35 In this, we owe particular thanks to Hero Granger-Taylor, who generously shared her knowledge and understanding of Roman garments and the way they were made.

36 MC 891, 1100 and 1101. See Mannering 2000a for details, patterns and discussion.

37 Mannering 2000a; Bender Jørgensen 2004b, pp. 92-93; Cardon, Granger-Taylor, Nowik 2011, fig. 306, Pl. 10.

38 Bender Jørgensen 2011.

39 Mannering 2000a, pp. 286-88.

40 Granger-Taylor 1982, 1987, 2000, p. 157; Cardon, Granger-Taylor, Nowik 2011, pp. 308-310, Fig. 318 and Pl. 18b-d.

41 Cardon, Granger-Taylor, Nowik 2011, Pl. 18b-d.

42 Mannering 2000a, p. 289.

43 Cf Cuvigny 2000; see also Cuvigny 2005, 2016.

44 Mannering 2000a, p. 287.

45 Cardon, Granger-Taylor, Nowik 2011, pp. 309-310.

46 Szymaszek 2015.

47 Following Croom 2000, p. 35 and Cardon, Granger-Taylor, Nowik 2011, pp. 308 and 319-320, the term ‘cloak’ designates semicircular, hooded outer garments such as the paenula, in contrast to ‘mantle’ that is used of the rectangular, draped himation.

48 Granger-Taylor 2000; Cardon, Granger-Taylor, Nowik 2011, pp. 293-294.

49 Mannering in prep.

50 Burnham 1972; Cardon, Granger-Taylor, Nowik 2011, p. 349.

51 MC 331, MC 861, MC 1144, MC 1234, MC 1628. The complete sock is MC 331, the fragmented one MC 1234.

52 Mannering 2006. MC 922, p. 1110.

53 Mannering 2006, p. 158, fig. 8. MC 548.

54 Cardon, Granger-Taylor, Nowik 2011, Pl. 30c-d.

55 Granger-Taylor 2007, 2008.

56 Granger-Taylor 2008.

57 Cardon, Granger-Taylor, Nowik 2011, pp. 319-341.

58 Cardon, Granger-Taylor, Nowik 2011, p. 356 Table 5.

59 Cardon, Granger-Taylor, Nowik 2011, p. 325.

60 Cardon, Granger-Taylor, Nowik 2011, pp. 330-335.

61 Ciszuk 2004. In Ciszuk’s work, ribbed damasks are termed ‘barred 2/1 or 3/1 twills’.

62 Ribbed damasks MC 745, 815, 818, 973, 1066, 1084, 1089, 1485; block damasks MC 505.

63 MC 1098. It also derives from the SS and is dated 100-120 AD.

64 MC 1357.

65 MC 1115.

66 MC 1625.

67 E.g. MC 974 (blue), 998, 1123, 1139, 1346. MC 974 is dated 100-120 AD, the others are from about 140 AD and later.

68 MC 975, 1040, 1081, 1114.

69 Ripoll 2004; De Moor, Fluck 2009.

70 Bender Jørgensen 2000, Fig. 6, the design then interpreted as a ‘tree of life’.

71 Bender Jørgensen 2011, p. 77.

72 Ciszuk 2000.

73 Bender Jørgensen 2000, Fig. 10.

74 MC 631, 940, 1016, 1036, 1056, 1752.

75 MC 396, 399.

76 MC 923, 942, 1028.

77 Sidebotham 1993, 1994a, 1994b; Sidebotham, Hense, Nouwens 2008, p. 57.

78 Sidebotham 1994b; Sidebotham, Hense, Nouwens 2008, p. 317.

79 Bender Jørgensen 2006a, 2007.

80 The twills derive trenches R and R/N, two of them from the upper layers. They are all quite small, and may have been re-deposited during the monastic settlement.

81 Wild, Wild 2014, Fig. 8.

82 See Bender Jørgensen 2004b, note 58, for find list.

83 Bender Jørgensen 2004b, pp. 88-91; for recent updates see also Cuvigny 2005, p. 334; Peacock, Maxfield 2007, pp. 427-429; Peacock, Blue 2011, pp. 346-347; Sidebotham, Hense, Nouwens 2008, pp. 53-60, and 189-192.

84 Bender Jørgensen 2004b, pp. 91-92.

85 Handley 2007, p. 370.

86 Handley 2007, pp. 370-371.

87 Calament, Durand 2013.

88 See e.g. De Moor, Verhecken-Lammens, Verhecken 2008; Erikson 1997; Fluck, Linscheid, Merz 2000; Palme, Zdiarsky 2012; Pritchard 2006; Rutschowscaya 1990; Thomas 2001; Van‘t Hooft et al. 1994.

Précédent Suivant

Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.