Archaeological Approaches to Market Exchange in Ancient Societies

7: Housing the Market: Swahili Merchants and Regional Marketing on the East African Coast, Seventh to Sixteenth Centuries AD

141 AncientSwahilitownsontheeasternAfricancoastarebestknownfortheirdedication to long-distance exchange within the Indian Ocean trade system. Middlemen traders in these towns are commonly cast as savvy, entrepreneurial brokers, negotiating trade between hinterland areas and overseas merchants as well as overseeing emerging cosmopolitan city-states. Often implied but rarely demonstrated archaeologically, these merchants participated in a vast market network that stretched thousands of miles along the coastal corridor, 100 miles to the interior, and thousands of miles across the Indian Ocean to ports along the Red Sea, the Persian Gulf, and the western coast of India. In this chapter I discuss what we know about market exchange along the Swahili coast based on archaeological and historical sources. I follow this overview with a discussion of an apparent contradiction in market development in Swahili towns on Pemba Island, Tanzania: the increasing privatization of exchange relations between Swahili and overseas merchants and the growing presence of overseas trade goods in the assemblages of non-elite towns and villages. This pattern offers a contrast to normative models of how elite goods moved within Swahili town regions, where prestigious goods are often understood to have moved in restricted spheres only. With archaeological data from Pemba Island I explore whether the increasing prevalence of overseas goods is suggestive of a regional economy with market exchange and how such an economy might be affected by greater restrictions on external commerce by local merchants. This process of privatization of external commercialization—in Jeffrey B. Fleisher Chapter Seven Housing the Market: Swahili Merchants and Regional Marketing on the East African Coast, Seventh to Sixteenth Centuries AD JEffREy B. flEiSHER 142 which market exchanges move into the private homes of town merchants—coupled with evidence of market exchanges in the local region provides a unique example of howamarketsystemcanthrivethroughhighly“embeddedtransactions”(Granovetter 1985; Garraty, Chapter 1). Rather than overt political institutions becoming involved in regulating markets, the Swahili case shows how a shared religious tradition as well as highly personalized relationships between merchants aided the emergence of market transactions. The AncienT SwAhili: A Brief inTroducTion The ancient Swahili inhabited a 3,500-km stretch of the eastern African coast (Figure 7.1), extending from southern Somalia in the north to Mozambique, the Comoros, and northern Madagascar in the south. Although it is difficult to determine the precise period in which a coherent group of coastal peoples emerged, the nine centuries from AD 600 to 1500 witnessed dramatic growth in the form of dozens of new coastal towns and villages. During the eleventh and twelfth centuries AD, a number of settlements began to distinguish themselves from their rural neighbors by constructing more permanent monumental architecture. The largest of these settlements—commonly known as stonetowns because of the presence of coral rag mosques, tombs, and elite houses—participated directly in the trade of long-distance goods and likely owed much of their expansive development to that trade. These were the towns the Portuguese encountered—and were impressed by—when they entered the Indian Ocean world in the late fifteenth century (Horton and Middleton 2000; Kusimba 1999; Pearson 1998). Although the most explosive growth occurred from the seventh century onward, the coast was settled much earlier, and some of these settlements were involved in long-distance exchanges. The Periplus Maris Erythraei (Casson 1989), a first-century AD Greek description of coastal settlements, describes the trade opportunities available at coastal sites, where imported metal and glass were exchanged for coastal and hinterlandproductssuchasivory,rhinoceroshorn,andcoconutoil.Unfortunately,no archaeological sites mentioned in the Periplus have been located, and finds from this period are extremely sparse (Chami 1999). In fact, little archaeological evidence exists for settlements that predate the seventh century, although the presence of Roman pottery (Juma 1996) and other imported fourth- through sixth-century artifacts suggests that modest trade continued in the centuries that followed (Smith and Wright 1988). A textual hiatus extends from the Periplus of the first century to a series of Arab geographies that date to the tenth century and later. The sparse archaeological evidence of the first half of the first millennium gives way to a rich and increasingly well-understood set of coastal and near-hinterland settlements that date from the seventh and eighth centuries and later. Archaeologists have found dozens of archaeological sites (of what had probably been hundreds) that contained a distinctive local pottery tradition, the TIW/Tana Tradition (Chami 7.1. Eastern African coast. JEffREy B. flEiSHER 144 1994; Horton 1996...