In the following sections I plan to examine how pottery production patterns can be used to investigate connections between two cultural groups that have been historically recorded as enemies. The two cultural groups examined are the Romano-Egyptians and the Meroites, who lived together in the same geographical region of Lower Nubia, between the first and the middle fourth centuries AD. A key point in this article will be the investigation of ceramic evidence from the Lower Nubian settlement of Qasr Ibrim.
Problems and research questions
The excavations of 1980 revealed five phases of deposition at Qasr Ibrim, timed between 100 BC and AD 100. Almost 35,000 fragments of pottery were recovered from the site, most of which related to the chronological period when the Romans occupied the fortress (Adams, 1985).
Preliminary ceramic examinations (Adams, n.d.) revealed the co-existence of Romano-Egyptian wares along with local Nubian and Meroitic types of pottery. Of course, the attribution of specific wares to distinct cultural groups is by definition problematic. It is well known that vessel shapes, decorative features and ceramic technologies can travel and be shared among potters of different cultural groups. Furthermore, knowledge, skill and even potters themselves can travel and share their ideas and repertoire with other potters. In Adam's (1985; 1986a; 1986b) study, the attribution of different ware types to distinct cultural groups was conducted in terms of broader ceramic traditions. Knowing the problems and difficulties behind such attributions, the present study accepts Adam's (1985; 1986a; 1986b) cultural divisions between Romano-Egyptians, Nubians and Meroites in order to examine a theoretical model of pottery production that will be presented below.
Officially, the relations between the Romans and the Meroites in Lower Nubia between the first century BC and first century AD were considered to be extremely hostile. This perspective can be verified by both historical information and archaeological evidence that were mentioned in the the previous article. This observation creates the first question that will be addressed in this research paper: Were the relations between the Romano-Egyptians and the Meroites indeed hostile or were there contacts, exchange of ideas, trade and possibly a more peaceful symbiosis? A second issue is to what extent these contacts affected pottery production itself.
Generally, there is limited information on pottery manufacture in Lower Nubia during the Meroitic era. Qasr Ibrim was proposed to be one of the production sites, despite the absence of direct evidence. Kilns may have been missed as a result of time pressure during the excavations (Adams, 1986a) or they might have been located in areas that were not investigated thoroughly, such as the front side of the cliff by the river (Adams, 1996). Until today, it remains a mystery whether any ceramic production ever existed in Qasr Ibrim.
Presuming that no ceramic production existed in the area, it is quite possible that all the ceramic finds recovered from Qasr Ibrim arrived in the settlement as imported wares, which were manufactured somewhere else. Adams (1986a, 100) suggests that the Egyptian utility vessels that were imported into Nubia during the Meroitic times were mainly wine amphorae, connected with the high development of wine-drinking in the area (Adams, 1986a, 91). Ceramic vessels often serve as containers in long-distance trade and, particularly in the Roman period, the wine and olive oil trade was conducted on a large scale between various locations in and outside of the Roman Empire (Sinopoli, 1991; Peacock, 1977; 1981; 1982). Adams’ observation is likely to indicate that the social interaction between the Meroites and the Romano-Egyptians in Lower Nubia led to the appearance of new social practices such as wine drinking, accompanied by some social contacts through trade.
If any form of ceramic production ever existed in Qasr Ibrim, it would be interesting to know in what way it functioned. Were there only Meroites producing pottery, or were there also some Egyptian potters that migrated to Qasr Ibrim along with the Roman legions? On this issue, Adams (1986b, 456) suggests that the Nubian silt-ware R34 (of Family N) shares identical forms with the Aswani ware R30 (of Family A); therefore, it can be postulated that it was made by Romano-Egyptian potters living in Nubia, who had no access to the special clays of the Aswan region. And even if there was only one cultural group of potters in the area, would their production be affected by the arrival of other imported vessels (such as wine amphorae), which carried different production patterns (e.g. form, decorative motifs, etc.)?
This article investigates a series of questions related to the level of social contact between potters in Lower Nubia. Even though ceramic evidence from Qasr Ibrim proves the use of culturally different types of pottery during the same period, can we be sure that the production of these vessels was destined to be culturally distinct? In other words, if there were contacts, trade and exchange of ideas between the Meroites and the Romano-Egyptians, would it not appear in the pottery production patterns of both sides? What connections were there between Meroitic and Aswani potters?
Previous research and related issues
William Adams (1986a; 1986b) was the first to attempt the construction of a general typology of Nubian pottery, mainly for providing a dating tool and a guidance manual for future studies of this specific material. Adams’ pottery classification from the Meroitic period in Lower Nubia is divided into five different ware families:
- Family D, represented by hand-made wares of Nubian traditions and group D.I. This family is closely akin to hand-made wares from the Napatan, Ptolemaic and Roman levels of Qasr Ibrim (Adams, 1986b, 411).
- Family M, represented by wheel-made finewares of Nubian traditions mainly from the Meroitic period and only one ware (W30), which belongs to X-Group times (Adams, 4986b, 435). (For more details on chronologies see Figure 1 below).
- Family N, represented by ordinary wheel-made wares of Nubian traditions made out of Nile mud. Multiple places of manufacture have been proposed. Group N.I belongs to the Meroitic period and exhibits the largest variation in forms, surface treatments and decorative motifs (Adams, 4986b, 440-441).
- Family A, represented by imported wares (both finewares and utility vessels) from the Aswan region in Egypt. Group A.I belongs to the Meroitic and early X-Group period, of between AD 100 and AD 500. It represents Graeco-Roman traditions that were being replicated in Aswan during that time (Adams, 1986b, 526).
- Family T of ‘Theban’ wares, represented by Middle and Lower Egyptian traditions imported into Nubia during the Meroitic period and later times (Adams, 1986b, 561).
Figure 1: Chronological chart of ware groups by period, adaptation after Adams (1986, 10)
Adams (1986a, 91) notes that the pottery from the Meroitic period exhibits a greater variety of forms than the pottery of the subsequent periods, except the X-Group. It seems as if each ware has its own distinct forms, even though some forms are found in two or more wares at the same time. The majority of these vessels are liquid containers: cups, goblets, small pots and bottles, lekythoi, oil bottles, beakers, jars and amphorae. Adams (1986a, 91) suggests that their abundance is undoubtedly connected with the high development of wine-drinking because of the easy accessibility of Egyptian wine in the Graeco-Roman age. More recent ceramic analyses performed by Rose (1996) in the Qasr Ibrim hinterland survey show that Adams’ classifications are quite general compared to the diversity of the recovered ceramic material. In fact, similar forms are found in different ware groups, not only in the same ware family, but also across families that are traditionally believed to be culturally distinct, such as the M and A Families. In the present study it is suggested that further comparisons of forms between vessels from Families M and A could be used to investigate if any similarities can be related to the diffusion of ideas between the potters of the two cultural groups.
Regarding imported Egyptian wares, Adams (1986a, 100) suggests that the Aswani utility vessels – unlike the decorated vessels – are not closely paralleled in Nubian pottery. The majority of imported vessels into Nubia during the Meroitic and Early Christian times are wine amphorae, which appear in a variety of forms and wares (Adams, 1986a, 100). Generally, large coarse vessels with thick and hard walls appear more suitable to be used in the transportation of such products (Howard & Morris, 1981). According to these observations, it is possible that the development of 'Egyptian' wine drinking practices affected the forms of Meroitic pottery, also accompanied by a large importation of Egyptian vessels into Nubia, which served as wine containers. Of course at the same time, along with the traded amphorae, other types of Egyptian decorated wares reached the Lower Nubian market. This study suggests that the appearance of new imported forms and decorative motifs in Lower Nubia could be examined to show whether Meroitic pottery production was affected by the arrival of Aswani vessels.
Adams divides his five main families into two broader fabric groups. The Meroitic Family M and the Aswani Family A are both described as made out of fine clays that exhibit a high level of fabric density and very fine to fine texture. In contrast, Families D, N and T are described as made out of Nile mud, exhibiting a porous fabric (yet not always for Family T) and a generally medium texture (Adams, 1986a, 78). These observations indicate that two of the main wheel-made wares of the Meroitic period in Lower Nubia (Families M and A) were made out of fine clays and exhibited fabric differences compared to the other three ware families. However, a major problem in examining fabrics from Lower Nubia is that the recovered material is much more diverse than what it traditionally believed to have been in the past.
Edwards (1999, 14) suggests that the existing overviews of Meroitic pottery by Adams can only be regarded as introductory study. Adams’ typologies have limited applicability outside Lower Nubia and his ceramic chronologies still need to be developed. Another major problem is that many typical forms of Meroitic pottery (M Family) from Lower Nubia are not encountered further south. For this reason Lower Nubia has to be viewed as a place with its own local production. On the other hand, field work in the heartlands of central Sudan has been relatively limited; therefore, a more detailed study of the existing material is needed in order to establish the general characteristics and overall distribution of both Lower Nubian and Southern Meroitic pottery (Edwards, 1999, 14).
Edwards (1999, 15) summarises a major problem: “In the past, a failure to clearly distinguish between imported (e.g. Aswani) and locally manufactured Lower Nubian wares has flawed many studies based on generalised stylistic/art historical approaches. In many cases, it seems likely that the provenance of a number of imported vessels, currently disputed, could easily be established by fabric analysis”. Recent work at Qasr Ibrim (Rose, 1996) and Gabati, an area near Meroe (Edwards et al., 1995; Edwards & Rose, 1997), has contributed a great deal towards establishing fabric series, and in identifying both chronologically significant groups of wares and potentially comparable products distributed across the Meroitic Kingdom (Edwards, 1999, 15). Finally, recent petrological analyses by Laurence Smith (1997; 1999) of Meroitic and Aswani samples found across different sites in Nubia have been used to examine the provenance of these ware families. The new question that derives from the above observations is whether any fabric similarities between Meroitic and Aswani vessels from Qasr Ibrim could indicate the exploitation of similar clay sources in Lower Nubia.
Finally, bearing in mind the knowledge, the technological means and the level of skill required for the manufacture of finewares (Rye, 1981) along with the level of specialisation and standardisation in the production of wheel-made pottery (Rice, 1991), it could be possible that the Meroitic and the Egyptian potters employed similar manufacture techniques (2). The question remains whether the employment of these techniques can be attributed to a direct or indirect contact between the potters of the two cultural groups.
For the purpose of this study, there will be examined a small assemblage of Romano-Egyptian and Meroitic wares from Adams’ M and A Families from the Qasr Ibrim collection. The objective is to investigate whether or not these types of pottery can reveal information related to similar production patterns between the two different cultural groups. Although the theoretical issues relating to the association of specific ceramic traditions with cultural or ethnic groups are complex (Jones, 1997) and will not be examined in detail here; instead, the study will focus on the practical aspects of pottery production that can be related to an exchange of ideas between potters. Any suggestion of a diffusion of ideas regarding pottery manufacture technologies, vessel typologies and decorative motifs, could be useful in elucidating the relationships between these populations. Furthermore, this approach can operate as a future methodological tool which employs ceramic analysis techniques to verify or re-evaluate the existing image of the past according to textual sources.
Finally, the present ceramic analyses are intended as a case study which contributes towards defining the connections in Romano-Egyptian and Meroitic pottery production patterns, rather than an attempt to cover the whole topic of social and cultural interaction in Meroitic Lower Nubia. The questions laid out above are too complex to be answered fully; however, this case study will contribute towards understanding the effects of social contact between potters, and towards elucidating some aspects of pottery production within the wider context outlined earlier, also in regards to the complex debate on the production patterns of the Romano-Egyptians and Meroites.
Methods and analysis
In the following sections, number of different methods and analysis techniques will be employed in order to approach the complicated issue of pottery production in Lower Nubia during the Meroitic era. Firstly, it is necessary to reconstruct the general image of Lower Nubia before trying to suggest any patterns related to pottery production and exchange of ideas. For this purpose, textual and archaeological evidence will be collected, compared and correlated in relation to: the economic potential and the autonomy of Lower Nubia; the Lower Nubian settlements and the population size; the Romano-Meroitic contacts through trade; and finally, the Meroitic pottery production in Lower Nubia. In this way, there will be proposed a hypothetical model of how pottery production functioned, and how the diffusion of ideas through social contact might had affected pottery production in Qasr Ibrim and the surrounding areas.
Secondly, focus will be placed on archaeological evidence from the excavations at Qasr Ibrim in order to investigate how different groups of pottery can be useful in determining similar production patterns and possible exchange of ideas between potters of different cultural groups. More specifically, there will be an examination of samples from two different ware groups: the W26 'eggshell' ware of the Meroitic Family M and the R30 Aswani ware of the Romano-Egyptian Family A.
The ceramic collections under examination were produced by William Adams after the excavations of the Egypt Exploration Society at Qasr Ibrim during the 1980s and they were first transported to the University of Kentucky in the United States. Today, the collections have moved to the UK and various parts are located in three institutions: the University of Cambridge, the University of Southampton and the British Museum in London. For the purposes of this study, the University of Southampton granted access to the two ware groups under examination. The techniques employed for the investigation of the pottery include an analysis of external characteristics, fabric examination under a compound microscope, and thin section analysis.
In relation to the analysis of external characteristics, the similarities and differences of the two ware families will be investigated by examining their form and external decoration. The analysis of form will focus firstly on the identification of different types of vessels within the sample groups that can be matched with existing typologies created by William Adams (1986b). Secondly, the analysis of forms will inform us over the function of the vessels by examining external properties such as shape, size, rim diameter and wall thickness (Rice, 1987, 207-243). The analysis of external decoration, also known as stylistic analysis, will aim in isolating different decorative motifs and in identifying their origins and connections with specific traditions (Meroitic or Romano-Egyptian). Stylistic analysis will also be used to examine the symbolism behind some decorative elements, contributing to the elucidation of ideas and beliefs that were once embedded in the pottery of specific cultural group (Rice, 1987, 244-273). The final aim of the analysis will be to provide information related to whether or not function and symbolic representations can be attributed to an exchange of ideas between the potters of the two different cultural traditions. Similar approaches have been used by Oliver Gosselain (2000) to investigate modern interaction between potters of different cultural groups in Sub-Saharan Africa.
The second analysis technique will be the examination of sherds under a compound microscope, which is a common method of identifying the composition of a vessel’s fabric. For this technique it is necessary to create fresh breaks on the selected sherds and to examine all the samples under a binocular compound microscope with magnification of up to x30. The colour of the fractures can provide information about the firing conditions under which the vessel was fired. Furthermore, the visual identification of particles of temper in the clay matrix can provide information about clay selection and manipulation processes (Orton et al., 1993, 67-75). In the present analysis, the identification of such inclusions will be performed according to the guide published by Peacock (1977). This includes the identification of voids and visible particles according to their shape, hardness and colour, in combination with their reaction to 10% dilute hydrochloric acid.
William Adams conducted preliminary fabric examinations on the material from Qasr Ibrim during the 1980 excavation. These descriptions were recorded in his unpublished field manual right after the excavation (Adams, n.d.) and the general conclusions of his fabric analyses were published in his two volumes of the Ceramic Industries of Medieval Nubia (1986). However, Adams (n.d., 4) noted the limitations of his preliminary examinations, which were conducted under great time pressure and only with the use of a normal hand-lens. For the purposes of this study, it was decided that new visual examinations were necessary, even though sherds were selected from boxes that had already been characterised by Adams and his associates. As compound microscope examination techniques were not available in the field during the 1980 excavation, new microscopic analyses are conducted to verify or disprove Adams’ classifications, or even add something new to his observations. Compound microscope examination is also used to compare the fabrics of Families A and M in order to investigate any patterns that suggest the exploitation of similar clay sources.
The final examination technique will be thin section analysis, which is an accurate technique for the identification of mineral inclusions. In this technique, a thin layer of clay is produced from the fracture of the vessel, which is then ground to a thickness of less than 40 microns. The extracted sample is then placed under a petrological microscope and examined under polarised light (Williams, 1983). The identification of mineral inclusions is performed according to the visual properties that are described in standard mineralogical manuals (Kerr, 1977; Gribble & Hall, 1992). Thin section analysis is also used to investigate the provenance of specific fabrics. After the identification of mineral inclusions, it is sometimes possible to tie in the fabric’s provenance on the geological map of a certain area, since clays are typically derived from parent rock formations, whose geological composition contains quantities of specific and known minerals (Williams, 1983).
For the purposes of this project, thin section analysis is conducted on a small assemblage of eleven samples for two main reasons: firstly to identify any mineralogical similarities between the clays of the two examined fabrics, and secondly to investigate their provenance. For the second purpose, thin section analysis results will be compared with previous studies from other sites in Lower Nubia, and will be examined in correlation with the geological maps of Sudan and Egypt in order to identify possible clay sources.
In the last part of this study, there will be a short summary of the approach followed and the current analysis results will be combined to show any similarities or differences in the ceramic production patterns of the two different cultural groups. These results will be used to test the hypothetical production model described earlier. Finally, there will be a short discussion on the potentials of the Qasr Ibrim ceramic material for future research.
- Adams (1986b, 456) suggests that the Nubian silt-ware R34 (family N) shares identical forms with the Aswani ware R30 (family A); therefore, it can be postulated that it was made by Romano-Egyptian potters living in Nubia, who had no access to the special clays of the Aswan region.
- Standardisation and specialisation does not only occur in wheel-made pottery. Ethnological examples from the Philippines have demonstrated that handmade pottery can also exhibit standardisation and specialisation related to the developed skills of the potter (Longacre, 1999).