Aksum [and South Arabia]

Aksum is both the name of a kingdom located in northern Ethiopia and the capital of said kingdom. The Aksumite kingdom, whose lingua franca was the Ethiosemitic language of Gəʿəz, enters the historical record around the turn of the Common Era and came to an end around the seventh century. Throughout much of that period, the Aksumites were in regular contact – sometimes amicable, sometimes hostile – with the South Arabian kingdoms of Sabaʾ and Ḥimyar, at times invading and occupying parts of South Arabia.

In the Ancient South Arabian corpus, references to Aksumite Ethiopians are limited to Sabaic inscriptions, ranging in date from the second half of the second century CE to 559 CE. The third century is the best documented period for Ethiopian-South Arabian relations (in chronological order Robin-Umm Laylā 1, CIH 308, CIH 308 bis, Nāmī NAG 13-14, Ir 12, Ja 631, Ja 635, Ry 533, FB-Maḥram Bilqīs 2, CIH 314+954, CIH 350, Ir 19, Ir 69, Ja 574, Ja 575, Ja 576+577, Ja 585, MAFRAY-al-Miʿsāl 2, Ir 20, MAFRAY-al-Miʿsāl 5). For the fourth century, only two Sabaic inscriptions (Ir 28, Gr 27) allude to contact with Aksum, while for the fifth century no documentation survives. Less Sabaic documentation is available for the sixth century (Gar antichità 9 d, Ry 508, Ja 1028, Ry 507, Ist 7608 bis, Wellcome A 103664, CIH 621, DAI GDN 2002-20, CIH 541, Ry 509, Murayghān 3, Ja 547+Ja 546+Ja 544+Ja 545), despite this being a period in which Aksumite rule in South Arabia was both more extensive and intensive, as well as the period in which Abraha, an officer in the Aksumite army, seized power as king of Ḥimyar. The reason for this disparity in documentation is that, with the transition from pantheism to monotheism [see Christianity [in Arabia], Judaism [in Arabia]) in the fourth century, the habit of erecting inscriptions in temples came to an end, with the result that the number of Sabaic inscriptions overall decreased.

The Aksumites are sometimes referred to by name, as ʾks¹mⁿ (ʾaksūmān) but are more often designated by a term derived from the root ḥbš. In descending order of frequency, these are ʾḥbs²ⁿ (ʾaḥbūshān), ḥbs²t (ḥabashat), and ḥbs²ⁿ (ḥabashān), and ḥbs²yⁿ (ḥabashiyyān). The last two terms are attested only once each, in Robin-Umm Laylā 1 and Ja 576+577 respectively, with ḥbs²yⁿ serving as an adjective rather than a noun. From the root ḥbš is derived the name Abyssinia, first attested in the Hellenized form ABACCIN on an Aksumite coin dating from ca. 400 (Voigt 2003: 62), as well as the Arabic ethnonym ḥabasha, most commonly used in medieval sources in reference to Ethiopians but often times extended to include black Africans more broadly. It appears that ḥbs²t designates Ethiopians as the entire body politic of Aksum, e.g. ʾrḍ ḥbs²t “the land of the Ethiopians” and mlk ḥbs²t “the king of the Ethiopians”, while ʾḥbs²ⁿ denotes a group or groups of Ethiopians of unspecified size, e.g. mṣr ʾḥbs²ⁿ “the armed forces of the Ethiopians”, tnbltⁿ ʾḥbs²ⁿ “the Ethiopian ambassadors”, and ḏ-bn ʾḥbs²ⁿ “some of the Ethiopians”. In support of this thesis, it should be noted that “the land of the Ethiopians” is never called ʾrḍ ʾḥbs²ⁿ, nor does “the king of the Ethiopians” take the form mlk ʾḥbs²ⁿ. Although it can be safely assumed that the Ethiopians to whom Sabaic inscriptions refer hailed from the Aksumite kingdom, these same inscriptions make a distinction between “Aksumitesˮ (ʾks¹mⁿ) and “Ethiopiansˮ (ʾḥbs²ⁿ/ḥbs²t). Thus, for example, Ir 28, dating from the reign of the Ḥimyarite king Karibʾīl Watar Yuhanʿim (ca. 312-316), speaks of a diplomatic mission to “the land of the Ethiopians and the Aksumites” (ʾrḍ ḥbs²t w-ʾks¹mⁿ). That this reflects an actual dichotomy is substantiated by Gəʿəz inscriptions from the pre-Christian period of the reign of ʿĒzānā, i.e. pre-340s which similarly distinguish between “Aksumitesˮ (ʾks¹m^m, ʾksm, ʾks¹wm) and “Ethiopiansˮ (ḥbs²t^m, ḥbśt) in ʿĒzānā’s title (Drewes and Schneider 1991: 242, 243, 246, 247). A likely explanation for this is that ʾks¹mⁿ designates the inhabitants of the main core of the Aksumite state, while ʾḥbs²ⁿ and ḥbs²t designate Aksum’s vassals in the northern Horn of Africa.

Although Robin-Umm Laylā 1, dating from sometime between 160 and 190 CE, alludes to Sabaean tribes in northwestern Yemen taking up defense against the Ethiopians, relations improved during the reign of the Sabaean king ʿAlhān Nahfān (r. ca. 195–215 CE), who formed an alliance with the Aksumite king Gadarat. This alliance was seen as significant enough to merit mention alongside other major achievements of ʿAlhān Nahfān in no fewer than sixteen inscriptions erected at Jabal Riyām, CIH 308 preserving the fullest version. One of the conditions of the alliance was that each of the two parties would aid the other in times of war. Thus, it was that, when war broke out between Sabaʾ and Ḥimyar during ʿAlhān Nahfān’s reign, Aksum joined the pro-Sabaean coalition (Nāmī NAG 13–14 — Nami 1965: 60–63). Aksumite-Sabaean relations remained amicable during the first part of the reign of ʿAlhān Nahfān’s son, Shaʿr Awtar (ca. 210–230). This was not to last, however, as the Ḥimyarite king Liʿazīz Yuhanif Yuhaṣdiq (r. ca. 220–235) re-established – at least temporarily – the long-defunct Sabaean-Ḥimyarite union, with himself as its head. The Aksumites did not view this development with approval, as they invaded South Arabia and advanced to Ẓafār, only to be defeated by a Sabaean force (Ja 631).

Thereafter, Aksumite relations with Sabaʾ remained hostile for the rest of the latter kingdom’s history. During this period, the Aksumites established settlements in the Yemeni Tihāma, referred to in Sabaic as ʾʿṣd < ʾaʿṣād “courtyards, plots, villagesˮ (sg. ʿaṣad), and maintained close contact with local tribes like Sahratān/Dhū-Sahratam and ʿAkkum, with whom the Aksumites often acted in concert. References to children and wives (wld-hmw w-ʾʾnṯ-hmw) in connection with these settlements indicate that those Aksumites who settled in South Arabia during the sixth century came with their families and evidently intended to put down roots in the region. Śawām (modern al-Sawā) in southwestern Yemen seems to have functioned as the Aksumitesʼ administrative headquarters in South Arabia, as it was to that town that Sabaean diplomats made contact with the Aksumites (Ja 585). In times of war, Aksumite forces fought in units known as ʾḥzb < Gəʿəz ʾaḥzāb “peoples, tribesˮ (sg. ḥəzb), a term which reflects the organization of Aksumite army divisions along clan lines, and which is attested in military contexts in Aksumite inscriptions as well. Although several third-century Aksumite kings – Gadarat, ʿAdhēbā, Datwīnas, and Zaqrīnas – are known from Sabaic inscriptions, leadership of Aksumite troops in South Arabia seems to have most often been delegated to a son of the Aksumite king, referred to in Sabaic as “the son of the nagāśīˮ (wld ngs²yⁿ), nagāśī being the standard Gəʿəz term for king. The coregency of the Sabaean king Īlsharaḥ Yaḥḍub and his brother Yaʾzil Bayyin (ca. 236–255) is a particularly well-documented period, not least in terms of Sabaean-Aksumite relations (Ir 19, Ir 69, Ja 574, Ja 575, Ja 576+577, Ja 585). During this period, the Ḥimyarite king Shammar Yuhaḥamid (r. ca. 235–245), broke off the peace agreement which he had made with Sabaʾ and made common cause with the Aksumites. The mid-third century, in fact, marks the high point for Aksumite occupation of South Arabia, as it was at this time that the Aksumites briefly occupied the oasis of Najrān, the point of departure for all northward-bound caravan trade, which was administered by a governor of the Aksumite king (ʿqb ngs²yⁿ) (Ja 576+577). The Ḥimyarite-Aksumite alliance was not to last, for at a later point in the coregency of Īlsharaḥ Yaḥḍub and Yaʾzil Bayyin both Shammar Yuhaḥamid and the Aksumites surrendered to the Sabaean coregents. When next we encounter the Aksumites in a military context, they were targeted in a campaign by the Ḥimyarite king Yāsirum Yuhanʿim I of Ḥimyar (r. ca. 265–287), one that brought an end, at least for the time being, to Aksumite occupation in South Arabia (MAFRAY-al-Miʿsāl 5).

We next hear of Ḥimyarite relations with Aksum during the reign of Karibʾīl Watar Yuhanʿim (ca. 312-315), when a diplomatic exchange between the two kingdoms took place (Ir 28). Sometime after this, however, relations turned sour, for Gr 27 documents repairs to a residential structure at Ẓafār “after the Ethiopians burned itˮ (bʿdn ḏt dhr-hw ʾḥbs²ⁿ). Since Gr 27 can be dated to the first third of the fourth century CE on paleographic grounds, it would seem to indicate that the Aksumites made an attack on the Ḥimyarite capital. This attack was ultimately unsuccessful, given that Ḥimyarite rule continued unbroken, and indeed no fourth-century Aksumite inscription records an invasion of Ḥimyar – precisely because this venture failed (Hatke 2022). On the other hand, the Aksumite kings Ousanas (r. ca. 310–330) and his brother and successor ʿĒzānā (ca. 330–365) lay claim to Sabaʾ and Ḥimyar, along with their respective royal palaces of Salḥīn and Raydān, in their titles (Drewes 2019: 197, 208, 216, 229, 233), suggesting that the invasion of Ḥimyar served as grounds for the political fiction of conquest. A fourth-century development of greater long-term historical importance for both sides of the Bāb al-Mandab was the introduction of Christianity. ʿĒzānā embraced the new religion, establishing it as the state religion of Aksum, a status that Christianity has enjoyed in Ethiopia down to the present. By contrast, the ruling elite of Ḥimyar, while initially tolerant of Christianity, did not convert, though the religion spread throughout South Arabia, finding particularly fertile ground in Najrān.

Persecution of Christians by the Jewish elite of Ḥimyar began in the late fifth century (Conti Rossini 1910), and by the early sixth century Aksum, which had established a diplomatic presence at Ẓafār in 509 (Gar antichità 9 d), became involved in this conflict. In 518, the Aksumite king Kālēb (r. ca. 510–540) sent a campaign against Ḥimyar that brought to power a Ḥimyarite Christian, Maʿdīkarib Yaʿfur (r. ca. 518–522), who served as Aksum’s proxy ruler in South Arabia. His authority was reinforced by a garrison of Aksumite troops stationed at Ẓafār. Within four years after coming to power, however, Maʿdīkarib Yaʿfur was overthrown by Yūsuf ʾAsʾar Yathʾar (r. ca. 522–525), a Jewish Ḥimyarite rebel who proceeded to massacre the Aksumite garrison at Ẓafār, burn the local church, and fortify the Red Sea coast against Aksum (Ja 1028, Ry 507, Ry 508). Then, in the autumn of 523, Yūsuf attacked Najrān, offering the local Christians the choice of conversion to Judaism or death. Although no Sabaic inscriptions refer directly to this campaign, it is recorded in detail in sixth-century Greek and Syriac hagiographies (Beaucamp 2007; Moberg 1924; Shahîd 1971). In response, Kālēb invaded Ḥimyar once again in the spring of 529, this time leading the Aksumite army in person. Yūsuf was killed in the course of the Aksumite landing, an event alluded to obliquely in CIH 621, and the Aksumite army quickly took control of Ẓafār, Ṣanʿāʾ, Maʾrib, Wādī al-Jawf, and Najrān. Kālēb spent seven months in South Arabia, restoring churches that had been destroyed and welcomed back within the fold all Christians who had been forcibly converted to Judaism. Once again a Ḥimyarite Christian was brought to power as Aksum’s proxy ruler, in this case one Sumūyafaʿ Ashwaʿ (r. 529–post 531). Ist 7608 bis and Wellcome A 103664, two fragmentary Sabaic texts that may derive from the same inscription, date from Sumūyafaʿ Ashwaʿ’s reign and imply a greater degree of Aksumite control than under Maʿdīkarib Yaʿfur. This is clear from the fact that the dedicants of the inscription invoke “their lords, the kings of the Aksumitesˮ (ʾmrʾ-hmw ngs²t ʾks¹mⁿ), as well as references to “governors for the kings of the Aksumites” (ʿqbt^m l-ngs²t ʾks¹mⁿ) and submission to the Aksumite kings (ytʿbdnn l-ʾmlk ʾks¹mⁿ). A corpus of fragmentary sixth-century Gəʿəz inscriptions from Maʾrib and Ẓafār, along with some of unknown provenance, appear to document the Aksumite invasion of 529 (Müller 1972; idem 2012; Piovanelli 2013). Through their quotation from scripture, including Isaiah and the Book of Psalms, these inscriptions give a decidedly Biblical cast to the invasion.

Like Maʿdīkarib Yaʿfur before him, Sumūyafaʿ Ashwaʿ was overthrown—not, however, through a Ḥimyarite rebellion but by an Ethiopian Christian, Abraha, who seized power sometime after 531 and reigned until ca. 560. After two punitive campaigns sent by Kālēb failed to remove him from power, Abraha was left in place as king of Ḥimyar. Abraha reportedly agreed to pay tribute to Kālēb’s successor, a ruler identified with Wāʿzēb (r. ca. 540–560), who in his Gəʿəz inscription (in musnad script) from Aksum, RIÉth 192, lays claim to South Arabia (Drewes 2019: 256). However, from his own inscriptions, in Sabaic, there is little doubt that Abraha was effectively an autonomous Ḥimyarite king. To that end, he restored the Maʾrib Dam as a gesture of good will towards his South Arabian subjects (DAI GDN 2002-20, CIH 541, Ja 547+Ja 546+Ja 544+Ja 545), and in a series of military campaigns managed to not only restore Ḥimyar’s Arabian empire but extend its reach to the frontiers of the Roman and Sāsānid Empires (Ry 509, Murayghān 3). For all that, Abraha remained true to his Ethiopian Christian origins, proclaiming himself “the Ethiopian kingˮ (mlkⁿ ʾgʿzyⁿ) and invoking Messiah (ms¹ḥ) and the Holy Spirit (rḥ qds¹) in his inscriptions, celebrating Mass in the church of Maʾrib, and building a church in Ṣanʿāʾ. Ethiopians maintained a significant presence in South Arabia during Abraha’s reign, their numbers swelling through the Aksumite troops who, having been sent by Kālēb to remove Abraha from power, ended up siding with the latter. The dynasty founded by Abraha declined under his sons Yaksūm and Masrūq and was toppled ca. 570 by a Sāsānid invasion of Ḥimyar.

George Hatke

References and suggested reading

Sources

• Beaucamp, J. (ed. and trans.) 2007. Le martyre de Saint Aréthas et de ses compagnons (BHG 166). Paris: Association des amis du Centre d’histoire et civilisation de Byzance.
• Conti Rossini, C. 1910. Un documento sul cristianesimo nello Iemen ai tempi del re Sharâhbîl Yakkuf. Atti della Rendiconti della Accademia dei Lincei 5/19: 705‒750.
• Drewes, A.J. 2019. Recueil des inscriptions de l’Éthiopie des périodes pré-axoumite et axoumite. Tome III - Traductions et commentaires. B. Les inscriptions sémitiques (Aethiopische Forschungen, 85). Wiesbaden: Harrassowitz Verlag.
• Drewes, A.J. & R. Schneider 1991. Les inscriptions de la période axoumite. A. Les inscriptions guèzes, in E. Bernand, A.J. Drewes & R. Schneider, Recueil des inscriptions de l’Éthiopie des périodes pré-axoumite et axoumite. Tome I: Les documents: 215–358. Paris: De Boccard.
• Moberg, A. (ed. and trans.) 1924. The Book of the Himyarites: Fragments of a Hitherto Unknown Syriac Work. Lund: C.W.K. Gleerup.
• Nami Kh.Y. 1965. Nuquš ʿarabiyya ǧanubiyya 4. Hawliyyat kulliyyat al-Adab, gamaʿat al-Qahira 22(2): 60–63.
• Shahîd, I. 1971. The Martyrs of Najrān: New Documents. Brussels: Société des Bollandistes.

Studies

• Hatke, G. 2022. Political Fictions, Political Realities: Aksumite-Ḥimyarite Relations in the Fourth Century CE. Antiguo Oriente 20: 15–52.
• Müller, W.W. 1972. Zwei weitere Bruchstücke der äthiopischen Inschrift aus Marib. NESE 1: 59–74.
• Müller, W.W. 2012. Äthiopische Inschriftenfragmente aus der himjarischen Haupstadt Ẓafār. Aethiopica 15: 7–21.
• Piovanelli, P. 2013. The Apocryphal Legitimation of a ‘Solomonic’ Dynasty in the Kǝbrä nägäśt – A Reappraisal. Aethiopica 16: 7–44.
• Ryckmans, J. 1969. L’inscription sud-arabe Nami NAG 13–14. Eretz-Israel 9: 102–108.
• Voigt, R. 2003. Abyssinia, in S. Uhlig (ed.) Encyclopaedia Aethiopica. Vol. 1: 59‒65. Wiesbaden: Harrassowitz Verlag.

_{Alternate spellings: Axum, Axoum, Aksoum, Aksumites, Aksumite, Axumites, Axumite, Abyssinia, Abyssinian, Abyssinians, Ethiopia, Ethiopian, Ethiopians, Ḥabashat, Habashat, Kush}