Aksum and Nubia: Warfare, Commerce, and Political Fictions in Ancient Northeast Africa

2. The Question of Aksumite Trade with Nubia

2. The Question of Aksumite Trade with Nubia Though excavations at the hill of Bēta Gīyōrgīs indicate that the district of Aksum, the town after which the kingdom took its name, was occupied at least as early as the fourth century BCE,63 it was not until the first century CE that the kingdom of Aksum was first mentioned by Graeco-Roman authors. That it attracted the attention of the Mediterranean world at this time is undoubtedly the result of the increase in western trade with India via the Red Sea following the conquest of Egypt by Augustus in 30 BCE, a development which brought Ethiopia into commercial contact with the Roman Empire. Though the Ptolemaic kings of Egypt had sent numerous expeditions as far south as Somalia in search of elephants for use in warfare,64 their contact with the indigenous peoples of the Horn of Africa remained limited, that with India even more so.65 With the establishment of regular trade with the Red Sea and Indian Ocean under the JulioClaudian dynasty (27 BCE-68 CE), however, Graeco-Roman authors began to show much greater interest in the commerce of the Horn of Africa.66 Chief among these authors was an anonymous merchant from Egypt whose Periplus of the Erythraean Sea, written in Greek sometime in the mid-first century CE as a guide to fellow traders, provides a wealth of information on the commerce of the Red Sea and the western Indian Ocean.67 Given Ethiopia’s close ties to the Graeco-Roman world even at this early date, it is no surprise to learn from the Periplus that the 63 Fattovich 2004: 74. 64 Casson 1993. 65 Sidebotham 1986: 7-12. 66 Ibid.: 13-47, 116-41. On the new pattern of seafaring, based on sailing in open water with the monsoon winds, which began c. 120-100 BCE and facilitated the growth of Indian Ocean trade, see Curtin 1984: 96-100. That Roman trade with the Indian Ocean world greatly exceeded that of the Ptolemies probably has much to do with the fact that the population of the Roman Empire, estimated at approximately fifty to sixty million, was far greater than that of Ptolemaic Egypt, which was probably no more than seven to eight million (Sidebotham 1996: 289). 67 Casson 1989. On the Egyptian origin of the anonymous author of the Periplus, see ibid.: 7; Seland 2010: 14. 26 Aksum and Nubia Ethiopian king of the day, one Zoskales,68 was proficient in reading and writing Greek.69 That later Aksumite kings erected inscriptions in Greek70 demonstrates that Zoskales was by no means an exceptional case in this regard. Though the author of the Periplus does not specifically refer to Zoskales as a king of Aksum and mentions him instead in the context of the commerce of Adulis, he calls Aksum (Ἀξωμίτης) a “metropolis” (μητρόπoλις) but refers to Adulis as simply “a fair-sized village” (κώμη σύμμετρoς).71 This strongly suggests that Aksum, not Adulis, was the political center of Ethiopia by this time. In addition, Zoskales is described in the text as the only ruler of the region between Ptolemaïs Theron on the Sudanese coast72 and “the rest of Barabaria” (=northern Somalia).73 In this the testimony of the Periplus is mirrored in the archaeological record, which has yet to yield evidence for the existence of any state other than Aksum in the Horn of Africa during this period. Finally, given that the coastal plain of Eritrea is only about 40-60 km wide and is largely unsuitable for agriculture, any state as large as Zoskales’ realm would have to have been based in the more fertile highlands in which Aksum was located.74 The archaeological record similarly substantiates the Periplus’ description of extensive commerce between Ethiopia and the Roman world by way of the Red Sea.75 While similar evidence exists for Kushite trade with the Mediterranean world,76 such trade seems to have reached Nubia almost exclusively by way of the Nile Valley. To date the only place on the Red Sea coast where possible evidence of a 68 The oft-repeated identification of Zoskales with the Za-Haqelē of late medieval Ethiopian king-lists is unlikely given that the earliest of these lists post-date the Periplus by well over a thousand years (cf. Casson 1989: 109). 69 Periplus 5:2.21-2. 70 Bernard 2000. 71 Periplus 4:2.6. 72 Casson 1989: 100-1; but see...