Warfare

The sources regarding pre-Islamic warfare in Arabia contrast between South-east and South-west Arabia, the former with a rich array of copper-alloy weapons and the latter de facto with a true epigraphic tradition. Other such sources include Ancient South Arabian inscriptions, Greco-Roman history, archaeological finds, contexts such as buildings and oral history which was written down centuries after the fact.

Warfare channels and conditions the social mentality and behaviour of the ancient world. In ancient south-western Arabia, to judge by analogy with synchronic and diachronic neighbours, it lies at the centre of notions surrounding the societal role of man versus woman. Pre-Islamic social mores transpire normally in a tribal structure which binds together resources for a common defence. Failure to fulfil tribal or clan duties of retribution belonged to the worst crimes — tantamount to a denial of the tribal system itself, to judge from Islamic texts. Adult men (more than ca. 15 years of age) all are expected to participate in the common defence. As early civilisations grew, a system for the retribution of wrongs, such as [feuds] and [vendettas][feuds], threatened the social fabric, the so-called mirror punishment. Fear of retribution held this in check, often supported by marital alliances. The rough interdicting topography of the region contributes to the heterogeneity of cultural manifestations, including conflict culture. But open desert expanses also opened up a horizon to its denizens to conduct distant raids, and fighting became a function of society.

The sources for pre-Islamic warfare are spotty. Numerous Ancient South Arabian texts from South-west Arabia contrast with the rarest, undeciphered, most superficial or least verifiable ones from South-east Arabia. On the other hand, metallic weapons since the early 2nd millennium BCE recently have become numerous in south-eastern Arabia, while those in the west are still rare (overview: Al-Jahwari et al. 2021: pl. 18). Bronze Age and Iron Age south-eastern Arabia both show an array of weapons in contrast to the lack in the west. The happenstances of find-preservation have proven to be an insuperable filter for our knowledge. The role of archery in south-western Arabia is inferred by the military tactics used by contemporaries and the abundance of its trappings in south-eastern Arabia.

Philology

With the onset of monotheism in the 4th century CE, South Arabian epigraphic sources fade. Greco-Roman and Arabic sources provide a basic historical framework. In a pioneer article (1976) and the Sabaic dictionary, Beeston contributed 120 relevant vocables. Ancient South Arabian (ASA) texts, such as one from the reign of Yathaʿʾamar Watār (Ṣirwāh, end of the 8th century BCE) already show a ‘military’ tone, with its martial vocabulary. This includes Sabaic words such as ‘seize the property’ (bʿl/hbʿl), ‘beat the enemy’ (bḍʿ/hbḍ), ‘devastate the fields’ (gbḏ), ‘rout’ (ks³ḥ/hks³ḥ), ‘be entrusted to the administration of the conquered territory’ (mḫḍ/hmḫḍ), ‘obtain booty’ (mly/s¹tmly); ‘take revenge on someone’ (nqm), ‘besiege’ (s³wk/s³wkt) or ‘attack’ (wrw; Arbach 2019: 30).

Early on, Beeston described the different kinds of south-western military units and occupational titles. Thus, the ḫamīs, a native Sabaean force, fought under the command of a king or one of his marshals (1976: 7). If this was a professional army of sorts, it probably waxed and waned with political, military and economic fortunes. Levies supplied by the Sabaean highlands form a second force (Beeston 1976: 8). Parallel to the Roman and later, Byzantine armies of the 1st to 7th centuries, the cavalry were proportionately lesser in number than in the late 1st millennium BCE (Yule 2019b: 80–81). To hold parity with the tactics of the day, the proportion of mounted soldiers must have increased over time.

Qayls are military, royal vassals. Their families and names correspond initially with their place of origin (Beeston 1976: 4). Key terms include ‘fielded army’ (mṣr), ‘overlord’ (qayl), ‘officers’ (maqtawi, sg. mqtwy pl. mqtt), ‘Bedouin auxiliaries’ (ʾṣḥb), ‘Aksumite and Sarāt troops’ (ʾḥzb), ‘ad hoc task force’ (gys²), ‘vanguard’ (mqdmt), ‘raiding column’ (ġzwy), ‘warrior’ or ‘man’ (ʾs¹d), ‘cavalry’ (ʾfrs¹), ‘persons mounted on horses’ (rkbt/ʾfrs¹m), ‘mercenary captain’ and ‘troops’ (sg. nḥl pl. ʾnḥl; Beeston 1976: 7–12; Yule & Robin 2005: 261). The warriors’ payment consists of a share of the captured spoils.

Important are mentions in Classical Arabic poetry as early as the 6th century CE to give at least some orientation for the most important weapons. Beeston pointed out that the only weapons to appear in ASA texts are the lance (rmḥ) and dagger (s²zb; Beeston 1976: 12–13). It contradicts historical and archaeological evidence for the main weapons used militarily. First, as stated above, Beeston never seriously considered the role of archery of the Greeks, Romans and Parthians in the centuries around the turning of the ages from BCE to CE (Yule 2019a: 154). Moreover, at the time he wrote, Arabian archaeology had hardly revealed yet the panoply of the Early and Late Iron Ages (LIA) particularly in eastern Oman. Hoyland nicely updates this philological discussion, but misses the abundant published excavated sources which show archery to be the main weapon especially for the LIA, i.e., late 1st millennium BCE to 300 CE (2003: 188–192; bibliography: Yule 2019a). The reason that archery remains textually mute is that archers consist of the lower social layers, far below mounted knights, and thus were not of interest. Robin expands the panoply to include vocables of the armies of the Prophet (Yule & Robin 2005: 268–269). In the 6th century CE, the flexible sling is one of the major weapons used (Schwarzlose 1886: 314; Robin 2019: 145). Mentions of the use of the spear in Classical Arabic poetry co-exist ambivalently with the notable absence of metallic spear heads in late pre-Islamic archaeological context. Strabo writes that the horse is introduced into Arabia in the 1st century CE confirmed by Sabaic mentions of horses at about this same time, and then mount.

Military campaigns in some details have come down to us in, more so in late pre-Islamic times in western Arabia, including the battles of Najrān (525 CE), Abraha’s (r. 533–after 558) campaign against the Bānū Amir, Maʿadd and Lakhmids north of Mecca (552 CE) as well as the Sasanian invasion of Wahriz (ca 570 CE). Prior to hostilities one would seek divine guidance and support. Discussions of ancient warfare contrast the ‘western way of war’ with asymmetric guerrilla strategy. It can be excluded that Ḥimyarite armies were influenced by the military tactics of their Greek, Roman, Parthian, Sasanian neighbours. Forces thus are expected to have consisted of heavy infantry, archers, peltasts as well as cavalry, light and heavy. Treachery is a practical alternative to siege, as in the case of the fall of Shabwa in which 30 Sabaeans breached the defences of Shaqir castle despite a defending force of 4,000 (Beeston 1976: 48). In another case, rock inscriptions recount how Ḥimyarite forces burnt a church at Ẓafār which contained 300 Aksumites (e.g., Ry 508; The book of the Ḥimyarites).

The narrative is often biased and may exaggerate the size of the enemy to underscore one’s own bravery. In several cases the numbers of soldiers are given, broken down, by cavalry, infantry and sometimes mounted on camel — ranging from three to 16,000 (Yule 2019b: 81–82; Yule & Robin 2005: 267). Most are dwarfed by the campaign in 25 BCE of the Roman Aelius Gallus (26–24 BCE) with two legions — more than 10,000 strong, in addition to 3,240 cavalry taking to the field against Sabaʾ. Forces of this size require military standards (Ar. liwāʾ) in order to know where the leader is and that the melée is still ongoing.

The Kašf al-ġumma is recorded oral history which explains the arrival of Azd-Arabs from South-west Arabia to Oman (Klein 1938: 21). Malik bin Fahm’s 6,000, “horse and foot” vanquish the Persian “40,000 men, some say 30,000” (Ross 1874: 114) “Asāwira” and “Marāziba”. They reportedly placed elephants at the front of their formation. Yāqūt in his Muštarik names ten Persian foundations by the ancient name of Dastagird (possibly at Falaj al-Sūq; Klein 1938: 23). Other ancient forts mentioned include one at Damā (Klein 1938: 62). The seat of the ruling al-Jalandā, Ṣuḥār, must have been fortified as were other ancient towns in the region such Dibā, Manaḥ and al-Shihr.

Archaeology

Authors frequently point to numerous examples of epigraphically dated rock art in Arabia to show different weapons (e.g., Robin 2019). A rare Ḥimyarite depiction (4th–5th cent.) shows a noble with body armour who on foot brandishes a fenestrated axe and mounted, a spear (Fig. 1; Yule & Robin 2005: 358, figs 157–158; Robin 2019: 146–147 fig. 1; Yule 2019b: 64, fig. 87, 142). Elite Ḥimyarite warriors, such as Ḥaẓiyān, depicted in Fig. 2, show the use of expensive and prestigious B-shaped bows and ample arrowheads. The grave stela of Kāthibat (Fig. 4) show a bow and to the figure’s left, a quiver. A hunting scene for ibex with a dog is shown. Hunts often take place in a military setting to cement social rank and structure.

Our sample of weapons recovered in regular excavations is spotty. The archaeology of LIA South-east Arabia offers next to no artefactual comparisons with that of the south-west. The graves of the highest-ranking individuals in south-eastern Arabia are also most thoroughly robbed ones, such as the king of Oman in Mleiha/al-Milayḥa (U.A.E.) attributable to the Pré-Islamique Récent (PIR) period, dated 222/221 or 215/214 BCE (Stein 2019: 129, fig. 6.6). The other examples show a range of social stratification in the warrior culture. At top rank are interred with larger graves, with more, rank-indicating and more qualitative grave goods (Fig. 5). Women’s graves never contain weapons.

A depiction of horse-back riding engraved on a phiale crafted from copper alloy of a helmed warrior riding a horse came to light in a grave in the central part of Oman which dates the bowl possibly to the PIR.B, second half of the 2nd cent. BCE to 1st cent. CE (Fig. 6).

Defensive and offensive archaeologically contexted finds are integral for the topic of warfare (Fig. 7). The fortifications of south-western Arabia develop over the centuries without having to defend themselves from siege machines. Prominent here are tall defensive buildings, up to six floors in height, as at Qaryat al-Fāw in southern Saudi Arabia (Fig. 9a), built within elaborate walled defences (Fig. 9b). They grow heavier toward the end of the pre-Islamic period with inner and outer walls (Figs. 9c and 9d). Ẓafār shows the heaviest fortification at the main entrance, to the south in al-Ḥayfa.

In south-eastern Arabia fortifications are known as early as the early–mid 2nd millennium BCE at Tell Abraq, but first become common in the EIA (Figs. 8, 10b). The rampart at Hīlī (Fig. 10a) is particularly regular in shape, built largely from mudbrick. This tradition continues into the PIR period at al-Milayḥa and ed-Dur. Far less elaborate in eastern Oman are EIA forts such as that al al-Hawqayn (Fig. 10b). Examples from the EIA and Samad LIA (such as al-Ṣunṣuna, Fig. 10c) show no development from the previous period. Those of the PIR are more elaborate and regular in form, in part owing to being built on the plain.

In 2012, an excellent example of a symmetrical late Sasanian fort came to light at Fulayj in the Bāṭina, ca. 31 km south-east of Ṣuḥār (Fig. 10d). Subsequently, excavation began.

EIA weapons in south-eastern Arabia never occur in primary context, a major deficit. For the Samad LIA, the main weapons include archery, daggers, swords and rarely, the horse. Bows and shields are known only from rock art depictions, not from graves. Given the extremely thin population, ate pre-Islamic warfare amounted to raiding. One assumes a Bedouin economy, although some doubt exists to what extent, when and where (Yule 2018: 460–468). Many of the ferrous weapons show geochemical correspondences with contemporary objects from Luristan in western Iran, in particular the ores from Sanandaj-Sirjan (Stepanov et al. 2020).

In contrast to the Wadi Suq period, the skeletons of the Samad LIA show no combat wounds at Samad in eastern Oman (Yule 2018: 457).

Paul A. Yule

References and suggested reading

• al-Ansari, A. M. 2010. Qaryat al-Faw, in A. al-Ghabban, B. André-Salvini, F. Demange & M. Cotty (eds) Roads of Arabia: 310–363. Paris: Musée du Louvre.
• Al-Jahwari, N., D. Kennet, S. Priestman & E. Sauer 2018. Fulayj: a late Sasanian fort on the Arabian coast. Antiquity 92,363: 724–741. DOI: 10.15184/aqy.2018.64.
• al-Jahwari, N., P. Yule, Kh. Douglas, B. Pracejus, M. al-Belushi & A.T. elMahi 2021. The Early Iron Age metal hoard from the Al Khawd area (Sultan Qaboos University) Sultanate of Oman, vol. 6. Oxford: Archaeopress.
• Arbach, M. 2019. Les mots de la guerre en Arabie du Sud au I^er millénaire avant l’ère chrétienne, in J.-F. Breton & F. Villeneuve (eds) La guerre en Arabie Actes des 22^e Rencontres sabéennes Paris, 21–23 juin 2018: 21–41. Paris: Geuthner.
• Beeston, A.F.L. 1976. Warfare in ancient South Arabia (2nd–3rd centuries A.D.) (Qahtan fascicle 3). London: Luzac & Co.
• Beeston, A.F.L., M. Ghul, W.W. Müller, J. Ryckmans 1982. Dictionnaire sabéen. Louvain-la-Neuve, Beirut: Libraire du Liban.
• Breton, J.-F. 2019. Le sac de Shabwa par les Sabéens vers 225–230, in J.-F. Breton & F. Villeneuve (eds) La guerre en Arabie Actes des 22^e Rencontres sabéennes Paris, 21–23 juin 2018: 85–110. Paris: Geuthner.
• Darles, C. 2019. Fouilles de Shabwa V. Les fortifications (BAH, 216). Beirut: Presses de l’IFPO.
• Hoyland, R. 2003. Arabia and the Arabs. New York: Routledge.
• Kennet, D. & N. al-Jahwari 2015. The Rustaq-Batinah Archaeological Survey Preliminary Report on the First Season 2013-14. Windows on our past, archaeological research in Oman 2012–2014: 265–354. Muscat: Ministry of Heritage and Culture.
• Klein, H., 1938. Kapitel 33 der anonymen arabischen Chronik Kašf al-Ġumma al Ǧāmiʾ li-ahbar al-Umma betitelt Ahbār ahl ʿOmān min auwal islāmihim ilā ʿhtilāf kalimatihim (Geschichte der Leute von ʿOmān von ihrer Annahme des Islam bis zu ihrem Dissensus). Diss. Uni. Hamburg.
• Moberg, A. 1924. The book of the Ḥimyarites. Lund: C.W.K. Gleerup.
• Robin, C.J. 2019. À propos de l’armement des troupes ḥimyarites au VI^e siècle de l’ère chrétienne, in J.-F. Breton & F. Villeneuve (eds) La guerre en Arabie Actes des 22^e Rencontres sabéennes Paris, 21–23 juin 2018: 145–173. Paris: Geuthner.
• Ross, E. 1874. Annals of ʿOmán, from early times to the year 1728 A.D. from an Arabic MS by Sheykh Sirhāʾn b. Saʿīʾd bin Sirhāʾn bin Muḥammad, of the Benú Alí tribe of ʿOmán, translated and annotated. Journal Asiatic Society of Bengal 43(1): 111‒196.
• Schwarzlose, F. 1886. Die Waffen der alten Araber aus ihren Dichtern dargestellt..., Leipzig: Hinrichs'sche.
• Stein, P., 2019. Languages and scripts in the Arabian Gulf in the Hellenistic period the epigraphic evidence from Mleiha (Sharjah, U.A.E.), in G. Hatke & R. Ruzicka (eds) Ancient South Arabia through history: kingdoms, tribes, and traders: 118–142. Cambridge: Cambridge Scholars Publishing.
• Stepanov, I., L. Weeks, K. Franke, B. Overlaet, O. Alard, C. Cable, Y. Yousif Al Aali, M. Boraik, H. Zein & P. Grave 2020. The provenance of early Iron Age ferrous remains from southeastern Arabia. JAS 120: 105–192. DOI: 10.1016/j.jas.2020.105192.
• Yule, P. 2018. Toward an identity of the Samad period population (Sultanate of Oman). ZOrAr 11: 438–486.
• Yule, P. 2019a. Defense during the Samad period – a first attempt at an archaeology of conflict in south-eastern Arabia. JOS 20: 143–176.
• Yule, P. 2019b. Tarikh al-yaman al-qadim ḥmyr — Himyar Late Antique Yemen. Aichwald: Lindensoft.
• Yule, P. & C.J. Robin 2005. Ḥimyarite knights and infantrymen, two reliefs. Arabia 3: 261–271, 358–363.

_{Alternate spellings: War, Conflict}