Economy, trade and ceramic production in Meroitic Lower Nubia

The economic potential and the autonomy of Lower Nubia

Lower Nubia has been an area with poor agricultural resources. Generally, the valley several kilometres south of the First Cataract is relatively infertile, in contrast to the valley north of Aswan (Jackson, 2002; Welsby, 1998; Edwards, 2004). According to Welsby (2002, 8), Lower Nubia is characterised by the total absence of rainfall and, apart from a strip alongside the River Nile, the environment is completely hostile to plants, animals, and humans. In ancient Egyptian records, Nubia is described as “miserable” or “wretched” (Smith, 2003). The ancient Egyptians were attracted mainly by its high-quality mineral resources, such as the diorite stones near the Second Cataract and the gold mines at Wadi al Allaqi and Gabgaba (Jackson, 2002; Friedman, 2002; Welsby, 1998; Edwards, 2004).

Little is known about the economic activities of the Kushites, and, unfortunately, appreciation of the potential of their land is generally based on rather modern geographical observations. To the north, agriculture and animal husbandry were limited to the banks of the River Nile, since water is essential for such activities. However, the high banks of the river may have contributed negatively to local agricultural production (Welsby, 1998, 153). During the Hellenistic period the importation of the saqia from Mesopotamia and Egypt possibly contributed to local agricultural productivity by making irrigation more efficient (Trigger, 1965, 123) (1). According to Strabo, the main crop grown in Nubia was millet, presumably the Sorghum vulgare or dhurra in Arabic, which is still the main crop in modern Sudan (Shinnie, 1967, 159) (2). Olive trees, which were once among the main crops of North Africa during the Roman period, do not grow at all in Lower Nubia due to desert conditions. Wine production, especially near the area of Qasr Ibrim, is unlikely to have existed (Welsby, 1998, 158) (3).

Even though local grain production seems to have existed at Qasr Ibrim, it is most likely that other agricultural products (e.g., wine and olive oil) reached Qasr Ibrim from other areas, either from the heartlands of Kush or from the southern parts of Roman Egypt. It can be postulated that social contact between the Meroites and the Romano-Egyptians in Lower Nubia was generally dictated by necessity, since the arrival of foreign agricultural products was probably linked to the survival of local populations.

The Lower Nubian settlements and population size

Most of the inhabited centres of Lower Nubia were located along the Nile. However, the surrounding desert could support sizeable nomadic groups, which often threatened the security of the settlements (Welsby, 1998, 153). Meroitic settlements in Lower Nubia may have had a rather different character from those in the southern heartlands. Although their importance as agricultural centres must have been small, they seem to have been significant as part of the major transportation route connecting Meroe and Egypt (Edwards, 2004, 156).

Studying population sizes in Lower Nubia is quite challenging. Anthropological studies conducted in Wadi Halfa have reported a low life expectancy among the Meroitic population (Martin et al., 1984). It has been suggested that even though the Meroitic Empire reached a level of important political development, the populations of the north did not benefit at all (Armelagos et al., 1981, 53). Recent reanalysis of archaeological material from other sites along the Nile demonstrated that during the Meroitic period, populations remained relatively small and were concentrated in a few settlements along the river, perhaps no more than ten to fifteen (Edwards, 1996). When bearing in mind that the Lower Nubian settlements were mainly isolated outposts in a rather inhospitable environment, it is easy to understand the problems that would have affected the area's demography. Firstly, a certain amount of logistical support towards these populations would have been required, at least until these settlements could become self-sustainable. A second problem would have been various military campaigns or epidemic diseases, which might have affected the population size. Finally, the limited agricultural production would not necessarily have been able to sustain all members of the local population (e.g., non-food producers, such as members of the administrative mechanism and priests). In fact, it has been suggested that the whole population of Lower Nubia was partly dependent on imported food (Edwards, 2004, 156-159).

These observations demonstrate that the traditional image of Lower Nubia as a rich agrarian province able to support large populations (Adams, 1977; Trigger, 1965) differs from recent interpretations. If an authority were prepared to support a population in such a barren area, they would do so if the population were primarily something other than food producers. In fact, most Lower Nubian settlements seem to have served primarily as outposts or chain stations of a large trading network that lay north of the Second Cataract (Edwards, 2004, 156). Qasr Ibrim was also a sacred site and major cult centre from early Kushite antiquity (Edwards, 2004, 162). It is most likely that there was a perceived need for these settlements to administer the area, regulate trade, maintain the cult sites, and support social contacts.

The Romano-Meroitic contacts through trade

The issue of Romano-Meroitic contacts and the way these reflected on trade activity is complex and has sparked considerable debate amongst modern scholars. A detailed study of the approaches to investigating Romano-Meroitic trade is presented in a separate article.

Generally, ancient trade activity could have taken place either by land or by water. In the case of the Romano-Meroitic trade, both land and water transportation were in operation during the Meroitic era, especially in Lower Nubia. The most important and possibly cheapest transportation route was the River Nile, particularly for the transportation of low-value goods such as relatively low-quality ceramic containers (Peacock, 1977; Peacock, 1982; Peacock & Williams, 2007). Trading activities along the Nile have been recorded on inscriptions and ostraca for both Romans and Meroites (Török, 1984; Welsby, 1998).

The Romano-Egyptians seem to have exported a variety of products, including olive oil, wine, and honey (Welsby, 1998, 174). Romano-Egyptian ceramic containers recovered as far as Wadi ben Naqa prove that Egyptian agricultural products travelled quite far into the Meroitic Empire (Vercoutter, 1962). The existing Aswani ceramic evidence from Qasr Ibrim (Adams, 1981; Adams, 1986a) corroborates this observation, indicating that the fortress was indeed a destination for Egyptian products and possibly a middle station for their onward transportation further south.

The evidence for Meroitic trade is quite sparse. The absence of a structured monetary economy in Kush makes it difficult to understand the Meroites' trade activities (Welsby, 1998). The elucidation of the nature of trade between the monetary economy in Egypt and the non-monetary economy at Kush requires even greater attention. The presence of fine pottery and other luxurious goods (both Meroitic and Egyptian) in burials across Lower Nubia indicates that local populations not only maintained themselves but also had the economic means to purchase such goods (Welsby, 1998, 173). It is likely that some wealth circulated in the area through trade or migrant labour.

Another form of ‘embassy trade’ system appears to have functioned since Napatan times among members of the elite (Edwards, 1996). During the Meroitic period, some members of the administration are recorded as ‘Apote Aromelis’, meaning ‘envoy to the Romans’. This title has been considered to relate to diplomatic and trading duties, and it therefore appears to indicate regular contact between the Meroites and the Romans (Edwards, 2004, 161-162; Haycock, 1974, 69). However, Edwards (2004, 167-168) suggests that there is no direct historical evidence which proves that private or commercial trade between these nations took place. On the contrary, most of the evidence refers to long-distance exchange, mainly of imported items found in burials, which were being channelled through elite or royal networks (Edwards, 2004, 167-168).

According to these observations, it is certain that a large circulation of Egyptian and Meroitic pottery existed in Lower Nubia, even though it may not be attributed to organised and large-scale trade activity. These ceramic vessels were either containers used for the transportation of agricultural products or vessels with a utilitarian and/or symbolic function, mainly found in burial contexts. It is also likely that all of these vessels functioned as transmitters of ideas about pottery production patterns and possibly affected the patterns of local ceramic production, if it existed, of course.

Meroitic pottery production in Lower Nubia

Shinnie (1967) and Welsby (1998) employ ethnological evidence to suggest that during the Meroitic period, there were two forms of ceramic production in Lower Nubia: a) wheel-made pottery made by professional potters, which was often traded, and b) hand-made pottery that was made by women for local use. Under the influence of Hellenistic Egypt, Meroitic pottery became a form of art which sometimes copied objects made out of metal or glass. Meroitic fine white wares and eggshell wares were amongst the most competent products of the Nile Valley (Welsby, 1998, 163-165; Shinnie, 1967, 116).

Evidence for pottery production from the Kingdom of Kush is scarce. It is clear that during the Meroitic period, previous Napatan traditions inspired by Egyptian production were replaced by new types of pottery of varied origins (Edwards, 2004, 170). The diversity of the Meroitic wares (both hand-made and wheel-made) may reflect their manufacture in various centres of production and by many different potters, even though Meroe (4) seems to have been a central focus of their production (Adams, 1973, 204). The expansion of wheel-made pottery production appears to have begun around the first century BC. The new wheel-made products exhibited standardised and homogeneous forms, and decorative motifs (5). Their production appears to have increased to the point that they presumably displaced other handmade wares. Assemblages in Lower Nubia also tend to include quantities of imported Egyptian pottery, usually about 20% on sites north of the Second Cataract, although older studies have failed to distinguish local from imported pottery found at Meroitic sites (Edwards, 2004, 171-172).

The location of Meroitic pottery industries would have been affected by two major factors: firstly, the availability of fuel and secondly, the proximity to the market (Welsby, 2002, 191). Access to the market was generally available through river transportation. As mentioned above, Qasr Ibrim has been proposed as one of the production sites of the Meroitic period, despite no archaeological evidence to support this claim. During the X-Group period (also known as the Ballana period, between c. AD 400 and c. AD 550), Qasr Ibrim became a very important settlement with the largest population in its history (Adams, 1996, 5). During that phase, archaeological evidence indicates extensive pottery production and large-scale importations of wine and luxury goods from Egypt (Plumley, 1975, 6-13; Plumley et al., 1977, 42-47). Unfired medieval white wares have been found at Qasr Ibrim, which indicates that it was one of the production centres of the time (Welsby, 2002, 193). The main settlement at Qasr Ibrim was located inside the fort on the top of the bluff, but during some periods, there was also a settlement at the foot of the bluff, close to the river’s bank (Adams, 1996, 1). This part of the settlement has been identified as operating in the Middle Ages under the name ‘Lower Ibrim’ (Browne, 1991) and perhaps earlier as well. During that period, industrial activities that required fuel and water, such as pottery production, must have been carried out close to the river (Clarke Sommers, 1912, 78-81). Unfortunately, these areas of Lower Ibrim were not excavated before their final destruction by the construction of the Aswan Dam (Adams, 1996, 1), so it is not known if similar production existed there during the Meroitic period. To understand Meroitic pottery production in Lower Nubia, it is necessary to focus on other known production sites.

Meroitic pottery kilns have been excavated in Meroe, in the northern area of the ‘Royal’ city (Garstang, 1912, 46). It seems that the most elaborate finewares were manufactured in the capital of the Meroitic Empire (Török, 1997). More recent evidence from Meroe indicates that both hand-made and wheel-made wares were fired in kilns, and it is possible that the same potters employed both techniques in their workshops (Robertson & Hill, 1999). Pottery workshops have been discovered in Musawwarat es Sufra, in the area beside the north wall of the city’s enclosure. Large dumps of kiln wasters indicate that high-quality Meroitic finewares were being produced there, along with quantities of local coarse wares. However, excavations in other parts of the site revealed a scarcity of finewares; therefore, it has been suggested that fineware production in Musawwarat es Sufra might have been largely unsuccessful (Seiler, 1998; Edwards, 1999). The paucity of finewares in the area may also be attributed to a fineware production that was mainly oriented towards other markets at a further distance, such as Lower Nubia and Qasr Ibrim. Furthermore, in terms of technological means, evidence from Musawwarat es Sufra shows that Meroitic pottery production closely resembled Romano-Egyptian production patterns. As well as operating the same type of circular kilns, the presence of Romano-Egyptian-style decoration stamps and a potter’s wheel indicates the use of similar manufacturing technologies (Edwards, 1999), possibly reflecting the diffusion of ideas between the potters. Other possible Meroitic kilns have been identified by Adams (1962) at Abdel Qadir and Argin in Sudanese Lower Nubia. However, Edwards (1999, 41) argues that the kilns at Argin are urban in nature and do not contain any manufacturing debris; therefore, their identification as pot kilns is unlikely. A more detailed study of Merotic production sites is presented in one of the following articles.

According to the distribution of sites excavated to date, it can be said that Meroitic Pottery production existed in two main areas: one close to Lower Nubia and the north regions of Kush, and another in the south, around Meroe and Musawwarat es Sufra. A similar division of manufacturing centres has been suggested by Laurence Smith (1997; 1999) based on fabric analysis examinations. Smith (1997; 1999) points out that the clay sources used in Meroitic pottery production can be divided into two main clay groups: the northern clays of Aswan, Meinarti and Kalabsa, and the southern clays of Meroe and Musawwarat es Sufra.

A theoretical model of pottery production

How pottery production may have functioned in Qasr Ibrim and the surrounding areas

As suggested earlier, pottery production in Lower Nubia (and probably Qasr Ibrim) is likely to have occurred in two ways: first, household production of hand-made pottery for domestic use (6); and second, workshop production of wheel-made vessels for trade. The Aswani production of wheel-made pottery operated almost certainly on an industrial scale, and specialised workshops produced standardised forms. A similar industrial production scale may have existed for the Meroitic finewares.

The existence of these two types of Meroitic pottery production falls within Peacock’s ethnological model of pottery production in the Roman world (1982, 6-11). According to Peacock (1982), household production can occur without the need for complex facilities, technologies, or skills. Vessels are usually hand-made by women, fired in open conditions, and used for domestic activities, even though some ethnological examples from the Balkans show that such pottery can also reach the open market. Industrial production can exist at the household industry or individual workshop level. It requires some professional skills and the use of more complex technologies and facilities than those of individual household production, such as turntables, fast wheels, and larger firing kilns. These vessels are mainly intended to be sold in the open market (Peacock, 1982) (7).

The agricultural resources of Qasr Ibrim and the surrounding areas of Lower Nubia were of relatively low potential; local populations would have been able to produce some foodstuffs to sustain their communities, but many other agricultural products would have been imported from other locations. Even if local inhabitants were self-sustainable, it is still unclear how non-food producers, such as potters, maintained themselves. However, various ethnographic observations have demonstrated that pottery production can be practised as a seasonal activity, alongside agriculture and animal husbandry, for extra income (Peacock, 1982; Sinopoli, 1991).

Pottery production may result as a natural by-product of food production: ceramic containers are needed for storing seasonal surpluses, cooking pots and serving vessels may be necessary for processing and serving food, and larger containers are useful for transporting local goods to markets for trade (Sinopoli, 1991; Skibo, 1992). It is rather unlikely that agricultural activities in Qasr Ibrim produced any surplus to be traded in long-distance markets, as it would have been carried in ceramic containers; certainly, in the case of grain, it could be transported in other containers, such as sacks and basketry. Similar organic materials have been recovered at Qasr Ibrim due to the dry nature of the area, which enables their preservation over time (Adams, 1985; Adams et al., 1983). The transportation of water from Qasr Ibrim to the nomadic populations of the desert could have occurred in ceramic containers, as was the case during the medieval period (Welsby, 2002). However, skins could also have been used for the same purpose.

The location of Qasr Ibrim appears appropriate for supporting pottery production: its proximity to the River Nile enables easy access to clay sources and water, though obtaining fuel may be a problem in an area that cannot fully support trees and plants. Dean Arnold’s ethnographic Ceramic Resource Threshold Model suggests that at production sites associated with sedentary or semi-sedentary societies, 37% of both clay and temper resources are produced within 1km of the production area (Arnold, 1985). This specific model would probably match the situation at Qasr Ibrim, as it also matches most of the production sites along the River Nile.

In summary, Qasr Ibrim has been proposed as one of the ceramic production sites in Meroitic Lower Nubia. So far, it has been demonstrated that there was a local need for pottery, that the area could maintain potters to provide most of the necessary raw materials, and that it operated as a local market. If such production indeed existed in the Meroitic period, it would be interesting to determine which cultural group controlled it. Was it the Meroitic potters, who are likely to have employed similar production patterns to those used further south in the Batn al-Hajar region and Meroe, or was it the Romano-Egyptian potters who possibly travelled from Aswan along with the Roman legions during the time when the fort belonged to the Romans? Furthermore, is there a distinct discrimination of ceramic production patterns based on the ethnicity of these potters, or is it possible that a mixture of similar production patterns existed through social contact, regardless of the potters’ ethnicities?

How the diffusion of ideas through social contact could have affected pottery production in Qasr Ibrim and the surrounding areas

Jones (1997, 84) suggests that ethnic groups within ‘Western’ social scientific traditions from the late 1960s onwards have been defined on the basis of self-identification and identification by others. Eriksen’s (1992, 3) processual approach to the definition of ethnicity holds that ethnic groups are culturally ascribed identity groups based on the expression of a real or assumed shared culture. The process of ethnicity involves a consciousness of difference among groups of people who perceive themselves as culturally distinct (Eriksen, 1992, 3). In the case of the Romano-Egyptians and Meroites, cultural differences between these groups were used to distinguish them as distinct ethnicities, even though this view has been traditionally overstressed in Roman literature, which viewed all cultural groups living outside the Empire as ‘barbarians’ (Woolf, 1998).

Smith (2003) stresses that ethnic discrimination between the ‘Ethiopians’ and the Egyptians existed in Lower Nubia from early Pharaonic times. However, during the Hellenistic and Roman periods in Egypt, and as a result of the complexity of the administrative mechanism, various ethnic groups (Greeks, indigenous Egyptians, Jews, and Romans) appear to have lived in a generally multicultural environment. In this environment, discrimination between ethnic groups does not appear to have affected civilian contact (Bilde et al., 1992). At Qasr Ibrim, evidence for the coexistence of different ethnic groups is strongly evident in the archaeological record. Adams (1986a), Edwards (2004), Welsby (1998), and Rose (1996) all agree that ceramic evidence indicates the presence of both Meroitic and Romano-Egyptian pottery, though this does not necessarily imply that each type of ware was used solely by a distinct ethnic group. As demonstrated in an earlier article, the Roman army had a strong presence in the area, and, moreover, some soldiers in the Roman legions were Greek speakers (Wilkins et al., 2006). Greek papyri dated to the late Ptolemaic period have been recovered at Qasr Ibrim, bearing requests to the oracle of Amun, while various ostraca from the same period record Nubian names (Ray, 2005). If the strategic importance of Qasr Ibrim as a middle station in a long-distance trade network along the Nile is also considered, the possibility of distinct ethnic groups that did not communicate at all is extremely low. If Meroitic and Romano-Egyptian potters lived together in the area, an exchange of ideas regarding pottery production patterns would have been possible through social contact.

Communication between ethnic/cultural groups can also result from trade (Renfrew, 1969). As demonstrated in the following article, trade existed in Qasr Ibrim and the surrounding areas, no matter who occupied the fort (Romans or Meroites). The possibility is that Qasr Ibrim functioned as a major market (as it did during the X-Group period) or as a meeting point and middle station between other markets. As described in Renfrew’s model of trade and exchange (Renfrew & Bahn, 1991, 322), such commercial activities require a place where people and traded goods meet. It is certain that imported Romano-Egyptian pottery (8) reached the Qasr Ibrim market, even if some local production may have existed in the area. In this case, it can be postulated that the ceramic containers used in the Romano-Meroitic trade served as both transporters of goods and conveyors of social ideas.

Communication between the Meroites and the Romano-Egyptians may also have occurred through other activities. Local immigrant workers might have been crossing the borders to work in the Roman gold mines and stone quarries, and Egyptian potters may have migrated to Lower Nubia, too. A second case is the presence of pilgrims travelling between the two states to visit locations of specific religious importance. In the case of Meroitic pottery, a variety of finewares recovered either in the hinterland of Qasr Ibrim (Rose, 1996) or in many Meroitic graves across Lower Nubia (Adams, 1986b; Edwards, 2004) verify that the vessels were being locally produced or at least imported to the local market (Adams, 1986b). According to this observation, it is quite possible that even if Qasr Ibrim had a local production of some finewares, others may have arrived there through trade or pilgrimage from other places in Kush or Egypt. In this case, it can be postulated that the ceramic vessels purchased from foreign markets transmitted the utilitarian and social aspects of pottery production from different areas, which were then incorporated into local pottery production.

It appears possible that, because of the religious and commercial nature of Qasr Ibrim, this specific area of Lower Nubia was a hub of communication and exchange of ideas, which could have affected the pottery production patterns of both cultural groups. In other words, a number of similarities would appear in the ceramic record in terms of vessel forms and decorative motifs, and if production took place in Qasr Ibrim itself, the similarities might also relate to the exploitation of the same clay source. Meroitic and Egyptian potters also appear to have employed similar technological means, such as circular kilns, fast wheels, and stamps. Again, if such communication never existed in Lower Nubia, the pottery of each cultural group would probably have appeared distinct, with very few similarities.

In order to elucidate the level of social contact between the Meroitic and the Romano-Egyptian potters, it is necessary to conduct several tests on two groups of samples from Adams’ A and M families, focusing on forms, decorative motifs, and clay resources. These families have been considered in the past to be distinct ware groups of the Romano-Egyptian and Meroitic traditions.

Notes

1. The saqia is a vertical wooden wheel rotated by animal power. It carries large pots (qadus in Arabic), which are used to lift water from a lower point to a higher point above the river’s bank. Even though the saqia is an important irrigation mechanism, it remains questionable to what extent it improved agricultural production in Lower Nubia (Welsby, 1998, 157). Edwards (2004, 159) argues that the saqia was introduced in Lower Nubia during the very late Meroitic period, and that before then the region's economic potential was so limited that it could not even cover basic subsistence needs.
2. In the early levels of Qasr Ibrim, only wild sorghum (Sorghum bicolor verticilliflorum) has been discovered. The only cultivated sorghum arrived in the settlement during the post-Meroitic period (Adams et al., 1983, 59). However, Rowley-Conwy et al. (1999) argue that domestic sorghum appeared at Qasr Ibrim around c. AD 100.
3. Even though climatic conditions were not ideal for growing vines, limited production did occur in the south (in the city of Kawa and in Oasis Bahriyah) during the reign of Taharqo (early seventh century BC) (Welsby, 1998, 158). In fact, a possible wine-pressing tank was found at the Oasis Bahriyah (MacAdam, 1955, 220), even though the amphorae recovered on the site suggested that it was used during the third century AD (Adams, 1966). A series of twelve vine-presses has been located in the northern area of Sudan, between Meinarti and Ikhmindi, possibly related to the period after the Romans withdrew from the Dodekaschoinos (Welsby, 1998, 158-160).
4. A source of fine, white kaolinitic clay has been identified in the quarries east of the northern cemetery at Meroe (Robertson, 1992, 47).
5. Meroitic decoration drew on symbolism and the official religion. By the late Meroitic period, variations in pottery became evident. In Lower Nubia, decorated vessels were far more abundant, with a larger variation of elaborate decorative motifs. It has been suggested that this kind of pottery was produced in the north of Kush, since Lower Nubia is the only place where it is found in large quantities (Edwards, 2004, 172).
6. The production of hand-made wares for domestic use has been ethnographically observed in modern Sudanese traditions (Barley, 1994). However, other ethnographic examples from the Andes in South America suggest that handmade pottery can also become a traded good within the confines of a specific cultural group (Sillar, 1997).
7. Of course, some archaeological examples from the Roman period (e.g., the production of Romano-British Dorset Black Burnished Ware) indicate that coarse, open-fired, handmade pottery can also be produced on a large scale and traded over long distances (Gillam, 1976; Tyers, 1996).
8. These imports were mainly coarse Aswani amphorae, which, according to Adams (1986a), arrived in Meroitic Lower Nubia along with new wine-drinking habits. However, Pamela Rose argues that the ‘Aswani’ ware R31 was manufactured locally, somewhere in Lower Nubia (Adams, 2004).