Due to the largely accepted idea that `the Arab conquests were made possible by the opponents’ weaknesses rather than by the power of the nascent Muslim armies’, pre-conquest Arab forces have received very limited attention. How such forces were brought together, organised and led have yet to be studied in any real detail. The main reason for this is the state of the source material. Unsurprisingly, aside from their deployments as scouts within their own armies, the Romans and Persians are silent about the military organisation of the Muslims, while the Arab accounts present their own problems. Their religious nature often attributes victory to the convictions of those involved and their submission to the Will of God rather than military organisation, skill and bravery. Events can be distorted to further an agenda or by the employing of literary topoi to bolster an otherwise unknown part of the narrative. Later Islamic sources also tended to portray their predecessors in anachronistic terms, projecting the social, political and military organisation of their periods back onto that of early Islam, imposing `a false sense of organisation and method on military manoeuvres, which were, in reality, much more chaotic’. Such an abundance of potential problems makes any attempt to reconstruct any aspect of the early Muslim military fraught with danger and undermines any chances of firm conclusions.
The earliest Muslim military actions would have been a combination of caravan looting and raids against neighbouring Bedouin tribes to bolster resources, seek vengeance, discourage potential enemies, claim strategic points or enforce religious conversion. Such raids reflected the enemies that the fledgling Muslim army faced and how rare true pitched battle was in Arab warfare. They also `contributed a great deal to the Muslim community in terms of wealth, experience and the achievement of political and strategic goals.’ How- ever, as the enemies of Islam grew in size and stature such an unstructured army would not have been successful, forcing Muhammad and his advisers to improvise and incorporate a more structured approach to administration and organisation.
Perhaps the most immediate change brought about by the rise of Islam came in the realm of army leadership. Aside from tribal leaders, who owed their status to their ancestry and personal success, pre-Muslim Arab war parties had little in the way of a command structure. Under Islam, ultimate military authority, itself something of a novelty across much of the Arabian Peninsula, lay with Muhammad and his caliphal successors; however, as campaigns became further removed from Medina, it became necessary to appoint individuals to military command. In choosing men of certain tribes for certain commands, the Prophet and his caliphal successors demonstrated an under- standing of tribal politics while the appointments of men like Khalid and Amr, later converts to Islam, showed that Muhammad was willing to promote military talent ahead of standing within the Muslim community. It should also be pointed out that the repeated instances of rapid communication and dictation of military movements attributed to the caliphs in Medina should be treated with scepticism. Some major redeployments may have been ordered by the caliphs but the majority of decisions will have been taken on the ground by those men the caliph had entrusted to achieve the strategic objectives of the campaign.
The leadership of skilled individuals such as Khalid may have encouraged the emergence of a more structured military beyond its tribal make-up. The Muslim army does seem to have used similar formations to late antique Roman and Persian armies with right and left wings and a centre. Advance guards, vanguards and rearguards are also mentioned. An even more organised structure is recorded at the Battle of Qadisiyyah, where the Muslim commander, Sa’d b. Abi Waqqas, had divided his force into sub-groups of ten. However, it is likely that such subdivisions were superimposed on the past by later writers for, even with this interposing of a religio-political hierarchy and the appearance of numerous independent corps during the Ridda Wars, there was little sign of what would be described as a regular, even semi-permanent army.
As with other antique forces, the early Islamic army was largely divided into cavalry and infantry. However, a tentative warning must be sounded regarding the blurring of the two as cavalrymen would often fight dismounted and infantry could be transported on horse or camel. The vast majority of Arab horse of the early period was light cavalry used as raiders and skirmishers or as lancers, rather than horse archers or heavy cavalry such as the cataphracts of the Roman and Persian armies. It is also worth noting that horses were not abundant in Arabia; a fact that might explain why Arab cavalry relied more on mobility and skirmishing to avoid costly casualties both in terms of men and horses. It might also partly explain why it was infantry that bore the brunt of the fighting in Arab warfare. The core of the Muslim infantry was made up of swordsmen who carried a straight, hilted blade – the sayf – that was used for thrusting and slashing. They also made use of iron-tipped spears and javelins. Another sizeable part of the Muslim infantry used the archery skills that hunting with a bow honed. The Arab bow seems to have been a smaller variant than its Persian counterpart but it is possible that the more rapid fire offered by the smaller bow allowed Muslim archers to more effectively shield their infantry and cavalry.
Little physical material remains of early Muslim defensive equipment, and that which does survive is difficult to date or source. Muslim sources rarely speak of military equipment unless the articles themselves were famous, such as the swords, shields, bows and lances of Muhammad, and it is likely that most Muslim soldiers will have fought without the full military panoply. Instances of Arab chainmail armour do survive, although how widespread its use was in the Muslim army before the conquests is difficult to gauge. Mail was expensive to buy or make, meaning that perhaps only the richest Arab soldiers or those who had served in the Roman or Persian armies will have had such armour. Helmets may have been less prevalent before the conquests with a hood of mail called a coif being used instead to protect the head. Shields were carried by both cavalry and infantry and, while they are not well described in the sources, the few surviving descriptions suggest that the normal Arab shield was wooden or leather made into a `small disk, certainly less than a metre in diameter’.
A less significant section of the Muslim army was that given over to siege engines. Most Arab settlements had some kind of fortifications but few were prepared for a prolonged siege so the Muslims will have had little experience of siege warfare. Siege equipment such as the swing-beam manjaniq, similar to the trebuchet of Europe, is seen in later Muslim armies; however, the extent to which such machines were used by the Arabs of the 630s is difficult to say. A manjaniq was deployed during the siege of Ta’if in 630, although its lack of success against modest defences is telling, which may explain why such machines were more likely to be used as anti-personnel weapons rather than against fortifications. There is also no evidence for the torsion-based predecessors of such machines, which further suggests that Arab siege craft was largely basic. However, while it is easy to downplay the siege abilities of tribal societies such as the Arabs and the Avars, they proved themselves to be quick learners and highly adaptive to such situations. The Arabs in particular seem to have quickly realised that `victory often depended on preliminary political success rather than sheer military power’. With this realisation, Muhammad, his successors and their commanders proved themselves adept at separating a settlement from its allies through negotiation or blockade and then offering `protection and toleration in return for a fixed tribute’. Through such a combination, even the most major of cities – Damascus, Ctesiphon, Jerusalem, Antioch and Alexandria – would prove to be within the grasp of Muslim forces.
With the advent of Islam’s temporal power, a vague outline of a recruiting process begins to emerge. Volunteers or prescribed tribes gathered at Medina or at a predetermined site, were formed into an army and then sent into the field. Most of the muqatila – `fighting men’ – who served in the Arab armies were of Bedouin origin, which is unsurprising given that raiding, fighting and familiarity with riding, spears, swords and archery were integral parts of their daily lives. However, the rapid expansion of the Muslim community brought with it a wider spectrum of potential soldier. There is some evidence that the Muslims equipped some of their more settled or poorer members to fight. Alliances with Jewish, Christian and other non-Muslim tribes played major roles in the military survival and successes of Muhammad and his `Umma in its earliest years. Clients and slaves were also present in Muslim armies with the likelihood being that not all of them were Arabic in origin. Defection also added to the military strength of the Muslim armies while at the same time undermining its opponents.
The recorded sizes of Muslim armies are often hard to accept due to their seemingly formulaic nature. They are usually portrayed as being particularly small in number throughout their earliest history, such as raiding parties featuring forces numbering less than 100. However, the rapidity with which Muhammad was able to field armies of up to and beyond 10,000 might be cause for some suspicion – 300 at Badr; 700 at Uhud; 3,000 at Mu’ta, 10,000 at Mecca and 12,000 at Hunayn. During the attacks on Roman and Persian territory, the Muslim armies are also regarded as being on the small side with perhaps as few as 6,000 fighting at Qadisiyyah and the garrisons in southern Mesopotamia perhaps only numbering up to 4,000.
This seeming paucity of Arab soldiers must also be tempered by the exaggerated reporting of the armies of their Roman and Persian foes. The Great Powers probably maintained a numerical superiority over the Muslims but it was almost certainly not as overwhelming as the suggestions of the Muslim sources, which at times attempt to put armies in the order of hundreds of thousands in the field. Many of the proposed numbers for Muslim armies need to be viewed from a contemporary perspective. The previous two centuries or more had seen a marked decline in the size of armies deployed by the Romans and Persians; so much so that Mauricius considered an army of 5,000-15,000 to be well proportioned and 15,000-20,000 to be large. The fact that the Muslims may have been able to field a force of anything between 20,000 and 40,000 at Yarmuk suggests that the numerical gradient they faced was not as severe as is usually thought.
However, in spite of some advances compared to the pre-Islamic period, the early Muslim military remained simplistic. Aside from perhaps the greater desert mobility that camels provided, they were at a technological disadvantage to their Roman and Persian adversaries and, while perhaps not overly serious, they were at a numerical disadvantage too. Organisationally, even after the successes of the Ridda Wars, the Muslim army was still closer to a tribal war party than it was to the professional forces that the Romans could field. They were not paid nor provided any benefits and their enrolling in the army was not recorded in any way. However, these men were fuelled by the prospect of booty, encouraged by the martial bonds of their tribe and buoyed by the morale offered to them by their religion, and, once they were brought together under the Muslim banner and led in battle by a cadre of skillful practitioners of war, they were to prove an increasingly irresistible force. And in the 630s, the Great Powers were about to find out how devastating such a force could be.