1. Introduction
The societies of the Horn of Africa were important players in the Indian Ocean trading system at least from the turn of the era and remained so until the late sixteenth century when a combination of local and external events led to the drastic reduction of contacts and external influences. The relevance of the Horn in long-distance trade derived from its strategic location at the gates of the Red Sea and a variety of much-sought local products, such as incense, ivory, gold, and slaves. From the tenth century onwards, Islamic polities developed on the eastern Ethiopian plateau and the Harar highlands and with the towns and other permanent settlements, which contributed to the development of trade (Curle, 1937; Fauvelle-Aymar and Bertrand Hirsch, 2011; Insoll, 2017; Pradines, 2017; Chekroun and Hirsch, 2020). These inland settlements were served by caravans that, departing from the coasts of present-day Eritrea and Somaliland, penetrated deep into the interior of Ethiopia, down to the southern forests of the country. Our knowledge of medieval trade in the Horn from an archaeological point of view, however, is scarce when compared to Egypt, the Arabian Peninsula, or the Swahili coast, and one may be led to believe that the region was less connected to the Indian Ocean network than others. Our work in Somaliland since 2015 proves that this was not the case.
In this article, we would like to present evidence from long-distance trade from the eleventh to the sixteenth century coming from sites documented in the area of Berbera (Fig. 1). A large percentage of the documented material, including up to 90% of the trade ceramics, comes from Asia—from Arabia to China. Although the region of Berbera (Barbaria) is mentioned already in the classical sources, it is only described in Arabic texts (Ibn Sa’id and Ibn Battuta) from the thirteenth century onwards (Dubois, 2014: 381–383), where it appears as a place of secondary importance compared to the other main port of Somaliland, Zeila (Taddesse Tamrat, 1977: 139; Fauvelle-Aymar et al., 2011). Its status as a proper town is unclear before the fifteenth century. It is mentioned as a settlement in 1505 by Venetian traveler Ludovico di Varthema and a few years later, in 1518, it is sacked by the Portuguese (Tedeschi, 1980: 279–280). We were unable to locate the medieval town, as it lies in all likelihood under later constructions. Nevertheless, we could document several medieval sites around Berbera, including places of trade and villages (González-Ruibal et al., 2017; González-Ruibal and de Torres, 2018). We will present here four sites of different nature and chronology that we documented between 2016 and 2020: Bandar Abbas, a place likely fulfilling commercial and ritual roles between the eleventh and twelfth centuries; Siyaara, a beach fair of the eleventh to early nineteenth centuries; Farhad, a fair some 3.5 km inland from the port of Berbera, which remained active between the thirteenth/fourteenth and sixteenth centuries; and Biyo Gure, a village that was involved in long-distance trade between the fifteenth and early sixteenth centuries. We will discuss the variety of imported goods that appear in these sites and interpret the evolution of trade for half a millennium. The four sites were places of entry for foreign commodities into the Horn of Africa. With the partial exception of Biyo Gure, they were not the final destination of the commodities, but simply conveyed them to the interior and, therefore, provide a good picture of the kind of goods that were being consumed in the wider region.
A note on methodology: the present work is based on finds from surface survey. Test pits were conducted in Biyo Gure, Bandar Abbas and Farhad. The results of the excavations suggest that the archaeological levels are all exposed on the surface (Biyo Gure, Bandar Abbas) or are very close to the surface and partially disturbed (Farhad). The sites do not seem to have been deeply stratified, which has much to do with the nature of the places—open, seasonal and lacking permanent structures (except Biyo Gure). We documented a total of 1485 items (pottery, glass, beads, stone objects), of which over 90% are imported goods. Local, hand-made pottery amounts to only 8.5% of the assemblage that we collected. While we might have unwittingly privileged imports during the survey, the truth is that local pottery actually represented a very minor part of the assemblage (always below 10%) with the exception of Bandar Abbas, where they appear in greater quantity (around 30%). We surveyed the entire sites in detail and in two cases (Siyara and Bandar Abbas) twice and with different surveyors so as to correct potential biases. We collected diagnostic sherds (shapes and decorations) of fine wares, storage jars, local pottery and glass, as well as beads and stone artifacts. Quantification is based on a minimum number of individuals, not sherds, but considering the selection already done in the field, the difference between fragments and MNI is usually negligible. Quantification is still problematic, as it is based on survey materials, the provenances are not always clear and there is some overlap in chronologies. We have selected only the pottery that has reliable chronological ranges and provenances. The results have to be taken with caution, but they provide at least a first glimpse into trade patterns.
[su_button url=”https://saxafimedia.com/asia-horn-indian-ocean-trade-somaliland/3/” style=”soft” size=”12″ wide=”yes” center=”yes” text_shadow=”0px 0px 0px #FFFFFF” rel=”lightbox”]CONTINUE READING ON THE NEXT PAGE >[/su_button]
