Examination of two wheel-made pottery traditions: R30 and W26

January 17th, 2018

The ceramic evidence in context

The two types of pottery examined in this paper were recorded during the 1980 excavation at two points along the western fortification walls of Qasr Ibrim (for detailed map see Adams, 1996, 13). The context consisted of a dense mass of refuse that had accumulated on the inner side of the fortification wall for a period of almost two centuries, which had begun during the late Ptolemaic period, around c. 100 BC and continued until the end of the first century AD, reaching a height of 4m. Dating was looked on archaeological evidence such as ostraka, fragmentary inscriptions on papyrus and about two dozen of coins. Almost 35,000 sherds of pottery were also recovered on the site (Adams, 1983, 95-96).

Stratified contexts of excavation consisted of five main levels. Level 5 was the lowest one, about 0.8m in depth, which represented the late Ptolemaic period of constructions in the fort. Pottery in this level was present but not abundant. Level 4, about 0.5m deep, contained wares which were mostly the same as in level 5 and a large number of amphorae believed to be trans-Mediterranean. Organic materials (e.g. Roman military boots) indicated the presence of the Roman army. Level 3 was the softest context, roughly 2m in depth, and containing similar pottery as in level 4, quantities of straw and dung. Level 2, about 0.5m deep, contained mostly river cobbles that had been deliberately piled up to serve as ballista ammunition. Finally, Level 1, again roughly 0.5m in depth, contained pottery material recognised as Nubian, identical to the one found in Meroitic graves from the local area; however, Romano-Egyptian types of pottery still constituted more than 50% of the total pottery in Level 1 (Adams, 1983, 96-97; Adams, 1985, 10-12).

Sampling criteria and description of sherds

For the purposes of this case study, 30 samples of pottery were randomly selected for examination, drawn from two specific major ware groups: the R30 Aswani ware of the Romano-Egyptian Family A and the W26 ware of the Meroitic Family M. The samples that were selected for this project are relatively small in number in order to allow a greater depth of analysis to be conducted within the limitations of a project of this scale.

According to Adams (1986b, 525) the Aswan wares of Family A are not of indigenous origin and they comprise 10% to 30% of the total ceramic record found between the First and the Second Cataract. Even though no actual kilns have been discovered, the large distribution of pottery in the vicinity of Aswan makes it almost certain that these wares have been produced in the local area. However, the possibility of multiple production centres cannot be entirely ruled out. The most distinctive characteristic of this family is the fine, hard, dense and pink paste. The technological properties of these wares are “typically Egyptian” (Adams, 1986b, 525). The use of imported Aswani wares of Family A continued for a long period at Qasr Ibrim, lasting from Roman (first century BC) to Terminal Christian times (c. 1500 AD). Chronologically, group A.I (between first and fifth centuries AD) is the most appropriate for this study because it represents the Roman period in Egypt and Qasr Ibrim (parallel to the Late Meroitic period in Kush). Group A.I is comprised of six wares: the R30 ordinary red wares, the R31 flaky pink wares, the R37 polished red wares, the W24 ordinary cream wares and the W32 fine cream wares (Adams, 1986b, 526-538). Ware R30 is both a utility ware and a fineware, described as Graeco-Roman Aswan ware of the Meroitic and X-Group periods, dated to between AD 100 and AD 475 (Adams, 1986b, 534-535). The R30 wares have been selected for analysis here, based on the fact that group A.I is “overwhelmingly dominated” (Adams, 1986b, 526) by this specific ware. The 15 sherd samples from ware R30 (Plate 1) are described in Table 1.

Plate 1: The 15 sherd samples of ware R30

Table 1: Description of the R30 samples

Adams describes Family M wares as indigenous Meroitic finewares made from fine residual clays rather than Nile mud (Adams, 1986b, 435). They are usually found in funerary contexts and they also appear in small quantities in all Meroitic occupation deposits (usually less than 2% of the total ceramic assemblage). It has been postulated that Family M wares were being produced in different centres, both in Northern Nubia and in the Meroe area (1) (Adams, 1986b, 435). The Family M wares are made out of fine clay and their colour varies from light cream to light grey or pink. The fabric usually contains very fine black, white and red particles, rare sand and no mica (Adams, 1986b, 436). According to Adams (1986b, 435-440) the Meroitic wheel-made wares of Family M are the R35 fine red wares, the W26 ‘eggshell’ fine white wares and the W27 pale pink wares. “Eggshell’ ware W26 is the main representative of this family. It is dated to between the Late Meroitic Period (c. AD 100) and the Terminal Meroitic-Early X-Group period (c. AD 350) (Adams, 1986b, 438). The W26 ‘eggshell’ wares were selected here for further examination based on their suitability for stylistic analysis because of their elaborate decoration. The 15 sherd samples from ware W26 (Plate 2) are described in Table 2.

Plate 2: The 15 sherd samples of ware W26

Table 2: Description of the W26 samples

Analysis of external characteristics (form and decoration)

6.3.1 Aswani R30

According to Adams (1986b, 534) the most commonly found forms for ware R30 are a wide variety of goblets, footed bowls, small jugs, small and normal size amphorae, lekythoi and jars. Less commonly found are cups, plain bowls, beakers, vases and oil bottles. The analysis of the forms of the 15 selected samples of ware R30 indicates the presence of mainly footed bowls, goblets (or small cups) and amphorae of both sizes. More specifically, sherds R30/1, 12 and 14 (Figures 1 & 2 and Plate 1) come from small to medium-sized bowls, possibly footed. Similar forms have been recorded as typical Aswani R30 designs (Adams, 1986b, 531, figure 298). Only one sherd (R30/3) (Plate 1 and Figure 2) comes from a larger bowl, which resembles an Aswani ware recorded by Rose (1996, figure 4.34, P261b) during the Qasr Ibrim hinterland survey. This bowl resembles other Roman traditions and appears to be a coarse copy of a Samian bowl type Dragendorff 18 (Tyers, 1996, 108-9). However, some other similar bowls from the area have been recorded as Meroitic (for more details see Rose, 1996, figure 4.16, illustration P68a). Sherds R30/2, 5, 6, 9 and 15 (Plate 1 and Figure 1) come from goblets or small cups, mainly designed for drinking. A series of similar R30 vessels has been recorded by Adams (1986b, 531, figure 298). It is interesting that R30 goblets come in rather complicated forms, exhibiting a variety of shapes, curves and incisions along the vessel’s body. The techniques which must have been necessary for their manufacture indicate that these vessels were thrown on the wheel by experienced potters who were able to master complicated forms. Body sherds that exhibit thickness above 5mm (R30/7, 8, 10, 11 and 13) (Plate 1 and Figure 1) possibly come from small table amphorae used for serving, or larger amphorae used for transportation. Sherds R30/10 and 11 are internally resinated and this is an indication that they had been used to transport or store wine (also see Adams, 1996b, 533, figure 300).

Out of the total of 15 samples, eight sherds carry painted motifs belonging to four main decorative styles:

  1. Dots in continuous rows (sherd R30/1) (Plate 1 and Figure 1). This decorative pattern has also been recorded in other larger Aswani footed bowls by Adams (1986b, 531, figure 298, D5) and belongs to his decorative element EG30-12 of style A.IA (1986a, 359, figure 211). Adams also records an almost identical form of bowl with exactly the same decorative pattern on a Nubian R1 ware of Group N.II (1986b, 461, Figure 265, illustration D6); therefore, similarities in decorative motifs can be postulated between Egyptian (Family A) and Nubian (Family N) traditions.

  2. Vine wreath motifs (Plate 3 and Figures 1 & 2), frequently combined with linear patterns, mainly found on decorated goblets, plates and amphorae (R30/4, 6, and 7) (see also Adams, 1986b, 531-3, figure 298-300). This element is classified as EG30-10 of style A.IA (Adams, 1986a, 359, figure 211).

Plate 3: Decorative motifs of samples R30/5, 6 and 7

  1. Linear motifs, mainly stripes and thicker bands, usually found in pairs, mainly on the external surfaces of amphorae (R30/ 8, 10, 13) (Plates 1 & 4 and Figure 2). These patterns have also been recorded on decorated amphorae and large footed jars (Adams, 1986b, 533, figure 300) and belong to the decorative motif type CD of style A.IA (Adams, 1986a, 359, figure 211).

Plate 4: Decorative motifs of samples R30/11 and 13

  1. Random splashes on the body of thick-walled vessels (R30/11) (Plate 4). This motif has no previously recorded parallels in Lower Nubia but part of it resembles motif HK8-1 of style A.IB (Adams, 1986a, 360, figure 212).

The basic similarity among all decorated Aswani samples is that the forms of the vine wreaths are rather abstracted and the linear patterns consist of relatively thick bands instead of delicate thin lines. All the motifs are painted in the same dark brown colour, which shows less colour variation than the decoration of Meroitic pottery. This preference for one colour may indicate that the decorated Aswani R30 wares were being mass produced along with the undecorated vessels, intended mainly for utilitarian purposes; therefore, elaborate decoration may not have been required.

Meroitic W26

According to Adams (1986b, 436) the most commonly found forms of ware W26 are cups, plain bowls, goblets and various small pots and bottles including lekythoi. Less common appear to be footed bowls, beakers, lamps, lids and jars. Analysis of the forms of the 15 selected samples of W26 indicates that the majority of the sherds come from small bowls, cups and goblets, while only a few sherds may indicate the presence of small amphorae. The basic problem in the examination of the W26 samples is that the material is highly fragmented and the absence of rim sherds makes it difficult to identify exact forms; however, an examination of wall thickness may be useful (2).

Even though ware W26 is characterised as an ‘eggshell’ ware (Adams, 1986b, 438), only five sherds or a third of the total sample (W26/3, 8, 10, 14 and 15) exhibit a wall thickness between 2mm and 2.5mm. Another five samples (W26/7, 9, 11-13) have an average wall thickness of 3.5mm and therefore they can be characterised as thin-walled finewares, yet not eggshell. Finally, five sherds (W26/1-2 and 4-6) have a wall thickness of 4mm and above. This wall thickness is standard in utilitarian vessels; therefore, these sherds quite possibly come from small table-wares or amphorae, even though Adams (1986b, 438) does not include table-wares in this category.

The present examination of form and wall thickness shows that sherds W26/2, 7-10 and 14-15 (Plate 2 and Figure 3) come from small cups or goblets. Sherd W26/10 is from a plain white cup that has no exact parallels in Adams’ typology but shows some resemblance to Adams’ cup A15 (Adams, 1986b, 437, figure 254). Rose (1996, figure 4.26, illustration P17a) records a similar vessel from the Qasr Ibrim hinterland as Meroitic. Sherd W26/15 comes from a typical Meroitic semispherical goblet, which is the same as Adams’ cup 22A (Adams, 1986b, 438, figure 255). Sherds W26/6 and 13 are from Meroitic semispherical bowls that have also been recorded in Adams’ classifications (see C20 and C25 in Adams, 1986b, 437, figure 254).

From all 15 samples of ware W26, ten carry painted decoration, three carry a combination of painted and incised decoration, and only two base sherds are undecorated. The majority of the sherds are decorated with a variety of motifs which usually come in combinations of various different colours: dark red-brown, red, orange, light red, light brown and black. According to Adams (1986b, 439) the decoration of ware W26 is the same as the decoration of ware W25 of the N.I Group of Nubian wheel made wares from Nile mud. This can be considered a typically Meroitic kind of decoration. The motifs can be divided into nine major groups:

  1. Linear motifs (both incised and painted) that combine a thick band with two or three thinner lines or stripes directly below (sherds W26/1, 6, 10, 12, 13) (Plates 5, 7, and 8 and Figure 3). These motifs come in a red or dark red-brown colour. Similar decoration has been recorded by Adams on jars and amphorae of ware W25 of Group N.I (Adams, 1986b, 453, figure 262) and it is also recorded as decorative element A2 of style N.IA (Adams, 1986a, 271, figure 123).

Plate 5: Decorative motifs of samples W26/1, 4, and 5

  1. A combination of linear painted motifs that divide the vessel’s surface into horizontal zones, which contain stamped decoration. This decorative motif is found on a single sherd (W26/2) (Plate 6) that exhibits stamped decoration of rhomboid patterns. No parallels to this have been previously recorded by Adams.

Plate 6: Decorative motifs of sample W26/2

  1. Vine wreath motifs. Sherd W26/4 (Plate 5) is a characteristic example of elaborate vine wreath decoration in different colours. This decoration has been recorded on jars of ware W25 of Group N.I (Adams, 1986b, 453, figure 262, illustration w2) and it is the same as the distinctive representational motif GH0-14 of style N.IA (Adams, 1986a, 280, figure 132). This vine wreath motif is also represented in Aswani decoration; however, the variation in colours, the thin delicate lines and the natural wavy movement of the wreaths indicate that more effort and time was required for the decoration of Meroitic finewares. It is interesting that Elhassan (2004) does not record the vine wreath as a Meoritic religious symbol.

  2. Circular motifs that resemble the representation of a human eye (sherd W26/5) (Plate 5). Adams illustrates Nubian W25 ware of Group N.I with similar eye motifs (see Adams, 1986b, 453, figure 262, illustration u22) and he also records this element as GH y-3 of style N.IA (1986a, 283, figure 135). Elhassan (2004, 13) describes this element as the wedjat-eye or ‘the sound uninjured’ eye of Horus. Identical elements are recorded on various other Meriotic sherds (Elhassan, 2004, 139, figure 3, illustration f-3). It is an Egyptian religious symbol that was used for a long time, from the Old Kingdom until the Roman period, and it originates from Egyptian myths related to the god Horus (Elhassan, 2004, 13).

  3. Painted curvy lines that form symmetrical rhomboid motifs. The decorated areas are also defined by double linear margins in different colours. Sherd W26/8 (Plate 7) is an example of this kind of decoration. Adams records the exact same pattern on a Meroitic lekythos (1986b, 437, figure 254, illustration M13) and categorises this element as EG e-9 of style N.IA (1986a, 273, figure 125), which is also the same as element G 6-9 of style N.III of the early Christian period (Adams, 1986a, 304, figure 156).

Plate 7: Decorative motifs of samples W26/6, 8, 9 and 10

  1. X-shaped motifs creating rhomboid spaces that carry spear-shaped elements. This decorative motif is found in a variation of three colours (sherd W26/9) (Plate 7). Even though this specific element is not recorded by Adams, various other X-shaped decorative elements are categorised as EG f2-f6 of style N.IA (1986a, 273, Figure 125). Adams records the exact same element without the leaves as G 6-10 of style N.III, which belongs to Nubian pottery of the early Christian period (Adams, 1986a, 304, Figure 156). Elhassan (2004, 125, figure 3, illustration b.3) interprets a similar spear-shaped element as a lotus bud. Generally, the lotus flower and the elements related to its different representations draw from Egyptian mythology and relate to the god Nefertem, who was described as the god of fragrance (Elhassan, 2004, 9).

  2. Random motifs that cannot be repeated on the surface of the same vessel twice because they are quite large and cover most of the vessel’s decorated surface (e.g. W26/11) (Plate 8). This specific motif – of a painted square containing a T-shaped symbol – has no similar parallels in Adams’ typology. However, Elhassan (2004, 175, plate 51) records the same symbol on the body of a Meroitic fineware cup, and on the internal surface of a Meroitic base sherd (Elhassan, 2004, 138, figure 3, illustration e-7). This symbol is interpreted as an offering table or an altar. In fact, altars are associated with the idea of offering, also drawn from the Egyptian religion (Ehassan, 2004, 16).

Plate 8: Decorative motif of sample W26/11

  1. M-shaped motifs painted in between an area defined by two linear margins (W26/14) (Plate 9). This decoration has no parallels in Meroitic pottery, although it resembles the lotus flower motifs. It may represent a rather basic attempt to imitate the lotus bud in the form of an M-shaped pattern, or it could originate from reliefs on Meroitic hand-made wares (e.g. element QR 5-2 of style D.I in Adams, 1986a, 261, figure 113).

Plate 9: Decorative motifs of samples W26/12, 14 and 15

  1. Internal decoration of a fishbone motif running along the vessel’s body, in a specific zone marked by two linear margins (W26/15) (Plate 9 and Figure 3). This specific motif has no exact parallels in Adams’ typologies, although a similar fishbone body frieze is categorised as element G 6-2 of style N.IIA (Adams, 1986b, 294, figure 146). Again, this style relates to X-Group Nubian pottery and does not match the period of this specific sherd.

As noted above, the decoration of Meroitic finewares comes in various motifs and various combinations. Welsby (1998), Edwards (2004) and Elhassan (2004) all suggest that the decoration of Meroitic finewares is basically symbolic and related to religious practices and beliefs. The elaborate decoration and variation of motifs and colours, along with the attempt to build thin-walled vessels, are two main indications that this ware was manufactured for both everyday use and specific use during special occasions (e.g. religious ceremonies).

In contrast, ware R30 appears to be a commercial ware. The forms of these vessels are much more complex than for W26 and exhibit larger variation. This may be an indication that other Roman traditions from different parts of the Empire inspired the potters of Aswan and led them to experiment in the production of more complex forms. It should also be borne in mind that the R30 wares are thicker and made out of a different quality clay that becomes distinctively hard after firing. This is another factor which made greater experimentation in the production of new forms possible. On the contrary, W26 eggshell vessels with wall thickness between 2mm and 2.5mm are technically difficult to be produced in complex forms.

Many of the R30 decorative motifs appear to be low quality imitations of the Meroitic motifs. This is observable when examining body stripes, linear patterns and vine wreaths. However, the vine wreath motif appears along with other forms of complex pottery decoration on Egyptian vessels from the Ptolemaic period. The decoration of these vessels draws from Greek ceramic traditions of the fifth and fourth centuries BC that arrived in Egypt after Alexander’s conquests (Pollitt, 1986; Aston, 1999). Possibly, the vine wreath representations were introduced into Aswan and Lower Nubia a long time before the Roman period. Therefore, it cannot be assumed that the Aswani potters copied the Meroites regarding this specific motif. It is also important to note that more complicated and symbolic forms of Meroitic decoration are not copied on R30s. Finally, it appears that Adams’ N Family stands stylistically between the M and A Families, which may present a useful opportunity for future research related to the stylistic connections between all three groups. An example of this is illustrated above in relation to the dot-decorated Aswani bowls.

Notes

  1. Adams (1986b) initially believed that the production of Meroitic finewares was largely if not entirely confined to Lower Nubia, with some samples coming from Meroe and Musawwarat es Sufra. After recent work in central Sudan, it was proven that this opinion was clearly a mistake and that the production of these wares took place in various different locations (Adams, 2004).
  2. Other visual examination techniques, such as use-wear analysis and surface alteration analysis (Skibo, 1992), may also provide insight into the possible form and function of vessels used in food preparation. Furthermore, chemical examination techniques, such as residue analysis (Rice, 1987), may be usefully applied in the examination of ceramic containers used for transportation or storage of foodstuffs; however, these techniques are only of limited use in relation to the small sample of sherds examined in this project.

Figures

Figure 1: Illustrations of ware R30 sample (forms and decoration) (1.R30/6, 2.R30/9, 3.R30/1, 4.R30/2, 5.R30/5)

Figure 2: Illustration of R30 samples (forms and decoration) (1.R30/4, 2.R30/7, 3.R30/8, 4.R30/3)

Figure 3: Illustration of R30 and W26 samples (forms and decoration) (1.W26/13, 2.W26/15, 3.W26/6, 4.W26/10, 5.R30/15, 6.R30/12)