Iberia - part 3: The Moorish Conquest

Originally published in Slingshot issue 287, March/April 2013 – the magazine of the Society of Ancients

Moorish troops and servants, Las Cantigas de Santa Maria, 13th century AD

Prelude

After the death of King Wittiza in 708 AD, the Visigothic nobility was once again divided in two factions.  One faction supported Wittiza's son Agila and the other one supported Roderic.  The latter faction would eventually gain the upper hand. What happened precisely, however, is exceedingly obscure and it is possible that the ruling elite had split, with Agila ruling in the north and Roderic in the south. 
Contemporary chronicles report that Hispania was destroyed by the attacks of the Arabs and by “internal fury”.  Moreover, a number of crop failures, and consequent famines, combined with the plague in 707 and 709 AD had taken a toll on society.  The resulting radically decreased commerce and income ruined innumerable men.  Additionally, slaves were fleeing in droves and Jews were being persecuted.  It is very likely that also the Visigothic army suffered from this general desolate state of affairs.

Most of the source material for the history of the Moorish conquest comes from the writings of Arab historians and geographers.  Several of these worked centuries after the events they described and were not themselves natives of al-Andalus, as the Arabs called the peninsula.  This name may be derived from the name Andalusia or Vandalusia, after the tribe of the Vandals, which settled in Iberia for a short period in the fifth century AD and then crossed the Strait of Gibraltar to colonize Northern Africa.

Like some other Muslim conquests, the invasion of Iberia seems to have been a local initiative, without the approval of the hierarchy, in this case Mūsā bin Nuşayr, the governor of Ifrīqīya (North Africa) based at al-Qayrawān (Kairouan), in modern Tunisia.  The author of the Chronicle puts the blame for the success of the Muslims with the ambitions of Roderic and the treachery of King Egica’s son Oppa, who supposedly conspired with the Arabs.  Also the cowardice of Archbishop Sindered of Toledo is mentioned, who fled to Rome rather than remaining with his flock.

In July 710 AD, the Berber commander Malik bin Tarif crossed the Strait of Gibraltar with a reconnaissance force of 300 infantry and 100 cavalry, all Berber units from North Africa, and landed near modern Tarifa – hence the name.  The lack of resistance to their incursion encouraged the Muslims to organise an invasion force the following spring.

The Moorish Conquest

In April 711 AD, a larger Umayyad-Berber army passed the Strait from Ceuta with a force of 4,000 infantry and 1,000 light cavalry to test the ground.  It landed on the beach near modern Gibraltar under the leadership of the Berber Tāriq bin Zeyad.  Tāriq established a base camp at the foot of the Mons Calpe, later renamed Jebal al-Tāriq – Mountain of Tāriq – and currently known as the Rock of Gibraltar; a corruption of Jeb(a)laltar, the Arab name less the –iq; the “l” and “r” are interchangeable between languages, cf. blanco in Spanish and branco in Portuguese.

Transport through the strait went on for several weeks and one wonders why apparently nobody sounded the alarm.  When Tāriq felt strong enough to venture deep into enemy territory, his men marched to Carteya (modern Algeciras).  The governor of Córdoba quickly organised troops to oppose the invading army and the forces collided somewhere along the Roman road from Córdoba to Seville, ending in a clear defeat for the Visigoths.

Modern aerial view of the Rock of Gibraltar

After this battle, a small Muslim force under Mugīth ar-Rūmī proceeded further north to take Córdoba. The city seems to have been decrepit.  The Roman bridge was broken and there was a major hole in the ramparts, which the Moors used to enter the city.  Most people had already fled and only a small number of military men had remained in an improvised fortress inside the walls.  When they finally capitulated after three months, they were all put to death.

In the meantime, Tāriq had pulled his forces back to Cadiz, waiting for reinforcements from North Africa.  In May the governor of North Africa Mūsā bin Nuşayr sent an army of Luwata Berber forces, possibly 10,000 infantry and 2,000 cavalry, across the Strait.  According to Ibn Khaldūn (d. 1406 AD), of the 12,000 invaders only a maximum of 300 were Arabs, since not many Arab warriors would have served under a non-Arab Muslim client general (mawlā) as Tāriq.

It is noteworthy that the Caliph al-Walid bin ‘Abd al-Malik had forbidden Mūsā to risk the lives of Arabs during the Spanish adventure.  Not only was the crossing a dangerous undertaking, but also a defeat on a battlefield in enemy territory could prove disastrous if a quick escape had to be made across the Strait.

Meanwhile, King Roderic was fighting the Vascones in the area of Pamplona.  Evidently this area was not ruled by Agila, king of the north.  He only heard about the presence of the Umayyad army two or three weeks after the invasion.  When Roderic’s army reached Toledo, it was reinforced by contingents from local nobles, including Sisbert and Oppa, brothers and loyal supporters of King Agila.  Hoping that the conflicts between the Visigothic clans could be put on hold in order to defeat the invaders, Roderic accepted their help. The Visigothic army may have numbered up to 25,000 to 30,000 troops, some 5,000 of them cavalrymen, but this is pure guesswork.

The battle of the Guadalete

The exact location of the battle is unknown.  Most historians agree that the battle occurred near Arcos de la Frontera, somewhere along the river Guadalete (Wādī Lakkah in Arab chronicles), which was dry during the summer.  Roderic formed his army according to the traditional Visigothic battle order: two lines, with the (heavy) cavalry in front and slightly on the flanks, and the infantry in the centre of the second line, commanded by the king.  The cavalry was to breach and destroy the enemy formation, after which the infantry could finish the job.  The flanks of the army may have been commanded by Oppa and Sisbert, but this is not certain.

The Muslims may have combined the characteristic manoeuvres of the early Umayyad armies with Berber tactics.  In that case, the front line would have been made up of infantry, with cavalry in the second line and on the flanks, ready to exploit any tactical opportunity.  Typical combat tactics would have been harassing the enemy with projectiles and simulated cavalry charges and retreats (al-karr wa-‘l-farr), followed by a fierce charge of infantry, intended to break the enemy lines.

The Visigoths attacked, driving back the Umayyad centre, but the slow retreat of the Berbers was only a ruse.  Tāriq had wanted to tempt the Visigoths into focussing their attacks on the centre of his army, so that his cavalry could push back Roderic’s flanks.  It is often suggested that Oppa and Sisbert, commanding the flanks, abandoned their king at this stage of the battle, but this cannot be confirmed.

Whatever happened, the Berber cavalry attacked the Visigothic flanks, while Tāriq’s infantry launched an assault.  As a result, Roderic’s infantry in the second line fled in disarray and Roderic’s army was destroyed, losing more than half of its men.  In his History of the Kings of al-Andalus Ahmad bin Muhammad ar-Razi reports that of the Muslim forces 9,000 men were left to receive any booty.  If most of the total invading army was involved in the battle, this would mean that around 40% must have perished.

Nothing is known for certain about the fate of King Roderic himself.  A reference in the Chronicle of 754 AD to his widow Egilona in ca 715 AD suggests that he died in battle, or soon after.  The Chronicle of Alfonso III (King of Asturias from 866 to 910 AD) states that he took refuge in Lusitania and was buried in Viseu, 100 km southeast of modern Porto, where a tombstone was found with the inscription Rodericus Rex.  Unfortunately, this tombstone has not survived the ages, so there is no way to ascertain the truth.

Whereas Arab sources state that the conquest was essentially the product of a single and rather bloody battle, the Chronicle of 754 AD gives the impression that there had been considerable prior raiding by Muslim forces from Africa.  Also, the presence of probably two Muslim armies in the peninsula for anything up to a year before the crucial battle with Roderic is mentioned, which may have been the breeding ground for the version that Roderic’s political opponents had engaged the Muslims to depose the king.  The Chronicle attributes the defeat to the Visigothic army’s flight and states that this was due to its having accompanied the king “in rivalry”, “deceitfully” and “out of ambition for rule”.

Aftermath

Though there are no reliable accounts on the exact course of the Muslim campaigns, it is generally assumed that, after the victory at the Guadalete, Tāriq split his army.  One column went to Lusitania, another stayed in the conquered city of Córdoba, while he himself marched for the Visigothic capital of Toledo, where he spent the winter of 711-12 AD.   

Hearing the news of Tāriq’s victory and quick expansion along the Guadalquivir valley, governor Mūsā landed either at Cadiz or Algeciras with an army of 18,000 Arabs in June 712 AD to take full control of the invasion.  Instead of proceeding directly to Toledo, he chose to occupy the towns which Tāriq had bypassed: Medina Sidonia, Carmona, Alcalá de Guadaira and Seville.  Seville fell after a siege of a few months, but revolted after his departure, only to be subdued again.  Advancing into Lusetania, Mūsā met with stiff resistance at Mérida, where the Visigothic nobility held out through the winter.  The city was forced to surrender on 30 June 713 AD. In accordance with the terms of the capitulation, the Muslims only seized the goods of the Church and of those who had perished or fled. Then Mūsā turned to Toledo. Near Talavera he is said to have met Tāriq, whom he struck on the head with his riding crop, upbraiding him for insubordination and greed.  Though Arab sources on the conquest of al-Andalus sometimes flatly contradict another, the rivalry between Tāriq and his superior Mūsā appears to be a common theme.

Moorish cavalry, Las Cantigas de Santa Maria, 13th century AD

At Toledo Mūsā executed an unspecified number of members of the indigenous aristocracy for their involvement with the brother of a former king, who had probably been chosen or consecrated himself as king at this time.  After wintering in Toledo, campaigning began again in the spring of 714 AD, which led to the nominal subjection of Galicia and the Ebro valley.  It is striking that after the invasion little attempt seems to have been made to defend the cities – only Córdoba, Mérida and Seville put up any effective defence – or to raise a second army.  No-one seems to have defended Toledo, a superb natural fortress.  A partial explanation for this might be that the sharp division in Visigothic society between military and non-military classes prevented the non-arms-bearing part of the population from participating in the defence of the city.

Recalled to report on the conquest by Caliph Walīd I, Mūsā returned to Damascus in September 714 AD, taking Tāriq with him and leaving his son ‘Abd al-‘Aziz as governor of the newly conquered lands.  In Damascus Mūsā found that the caliph had died and had been succeeded by his brother Sulaymān, who prosecuted him on charges of embezzlement.  Stripped of his honours, Mūsā died in ignominy.  Tāriq, the hero of the Iberian conquest, was also forgotten and died in obscurity.

Over the next five years, ‘Abd al-‘Aziz completed the subjugation of the western region of the peninsula.  Subsequently, he conquered the north-east, the fertile lands of the south and the Levante in the east, skilfully using a diplomatic strategy of cooperation with the Visigothic nobility on the one hand, and brutal acts of violence to terrorize opponents on the other.  In spite of the rapid conquest of the Iberian peninsula the Arab hold of the conquered land was precarious, as the Muslims formed only a very small percentage of the population.  Indeed, in around 718 AD Caliph 'Umar II is said to have seriously considered the idea of abandoning the conquest.

The Battle of Poitiers

After conquering most of the Iberian peninsula, the Muslims made repeated advances into France, taking Narbonne in 719 AD after a long siege, but failing to capture Toulouse.  The population of Narbonne was treated harshly: the men were slaughtered and the women and children enslaved.

In 725 AD they advanced up the Rhône valley and sacked Autun in Burgundy, before they were driven back again.  In 732 AD they crossed the western Pyrenees at Roncesvalles and took Bordeaux, advancing into central France.  The Frankish maior domus Charles Martel (“The Hammer”) defeated the invaders led by ‘Abd ar-Rachmān al-Ghāfiqī at modern Moussais-la-Bataille, 22 km north of Poitiers.  Charles earned his nickname as a result of this victory.  Referring to this event as the Battle of Tours, Carey renders a detailed description of the course of this battle in his book “Road to Manzikert” (pp.101-7).

Traditionally, this battle is dated to October 732 AD, but there are indications that the battle was fought a year later.  Yūsuf bin ‘Abd ar-Rahmān took Arles shortly afterwards in 735 AD, devastating Provence for a period of four years.  In response, Charles Martel launched a raid into western Provence in 737 AD, taking Avignon, though he was not able to retain it.   After an unsuccessful siege of Narbonne, he withdrew, ordering the destruction of the defences of the city of Maguelone (near Montpellier) and the amphitheatre of Nîmes.

Charles Martel fighting 'Abd ar-Rachman al-Ghafiqi at Poitiers

A second raid by Charles in 738 AD was not so much directed against the Arabs as against the most prominent of their local allies, the Duke Maurontus of Marseille. Having taken Avignon for a second time, he advanced on Marseille, causing the Duke to flee to a fortified island.  After these defeats, the Muslims did not advance further into Europe, mainly for reasons other than military weakness. Political instability on the side of the Moors and the economic structure of Western Europe, which yielded little interesting and easily removable plunder, made it difficult to motivate soldiers for campaigns in the north.

The importance of Poitiers has often been exaggerated; it was no Waterloo.  The Arabs felt no compulsion – religious or otherwise – to conquer western Christendom in the name of Islam.  “To them Europe seemed remarkably unattractive, with few opportunities to trade, little booty and a terrible climate” (Armstrong, Islam, A Short History, p. 44).  The defeat near Poitiers did, however, signal the beginning of end of the booty economy of the Muslim Iberian peninsula, as the pressure on resources in al-Andalus could no longer be relieved by raiding into Francia.  Instead, the Iberian Muslims were obliged to live off the resources of the peninsula. Competition for revenues and status would soon give rise to savage feuds.

 

Resistance

The general lack of resistance that the invaders met with during their conquest may be based on the fact that the invasion was originally taken for another transient Moorish raid.  Roderic’s opponents may have hoped that the invaders would defeat the king and depart, leaving them in charge.

The centralised character of the kingdom may have facilitated the rapid conquest of Iberia, similar to the situation of Anglo-Saxon England in 1066 AD.  Had it been divided into numerous local lordships and principalities, these would no doubt have put up strong resistance.  However, the defeat of the royal army at the Guadalete left the entire Peninsula open to the invaders.

North of the Pyrenees in Septimania, Roderic’s rival King Agila II (711-14 AD) had been able to survive the Moorish invasion.  Based on the distribution of Agila’s coinage, it seems that Agila’s influence was limited to the northeast, chiefly the provinces of Septimania and Tarraconensis. Agila’s rule was stable enough to secure the throne for his son Ardo, who succeeded him in 714 AD and held out until the Muslims pushed north across the Pyrenees in 721 AD, when he was finally defeated. Apparently, the Goths of Septimania made no attempt to negotiate with the invaders.  Cities like Nîmes and Carcassonne managed to hold out against the invaders well into 725 AD.

Moorish troops with captives, Cantigas de Santa Maria, 13th C. AD

Pelagius, (Pelayo in Spanish), son of Fafila, dux of Gallaetia, leader of the last and most prominent of the Gothic remnants, established the strong Kingdom of Asturias in the north, ruling it from 718 AD until his death in 737 AD.  Through his victory over the Moors at the Battle of Covadonga in 722 AD, he is traditionally credited with beginning the Reconquista.

The difficult terrain and remoteness of Asturias definitely aided his cause greatly, but he also benefited from being a secondary target anyway.  Eastern Iberia, and ultimately France, offered far more interesting opportunities than the isolated mountain hold-out Pelagius had created.  When a long famine, which began in 750 AD, caused many Andalusi to leave for North Africa, the Christian King Alfonso I (r. 739-57 AD) of Asturias saw his chance to establish some fortified outposts in the Duero plains in this period and to raid even further south.

Pacification

The Cordillera Central marked the most northerly limits of Muslim occupation in the western half of the peninsula. Coimbra, Coria, Talavera, Madrid, Guadalajara and Medinaceli were all frontier settlements; the position of the frontier would hardly change for three centuries.  To the north of these settlements, the land seems to have been almost completely uninhabited, except by wandering shepherds.  The southernmost outposts of Christian settlements were places like León and Astorga.

In the East the frontier presented a different picture.  Here the Muslim settlement pressed up to and into the Pyrenees: Pamplona, Tudela, Huesca, Girona and Narbonne were in Muslim hands.  Only in Narbonne, taken by Charles Martel’s son Pippin III in 759 AD, was their rule challenged.  The northern outposts of Muslim settlement would remain intact until well into the twelfth century.

In general, the Muslims offered generous terms, which certainly made surrender a more attractive option, whereas unsuccessful resistance could lead to death.  Theodemir, comes of Baetica in the south, who had been defeated in open battle in 713 AD, was able to negotiate a treaty with the invaders.  Still firmly possessing the vital cities, he forced the Muslims to accept his rule over the province.  Of course, Theodemir was a subordinate to the Emir of al-Andalus, but the independence of the province lasted until the 780s AD, maybe even longer, and thus survived his death by at least forty years.  Because of this treaty, Theodemir’s region would be known by the Muslims as “Tudmīr” for centuries afterwards.

In the later phases of the conquest, many Visigothic lords in the Ebro valley were allowed to retain their lands and status and soon converted to Islam.  Converts from Christianity to Islam came from two categories. The first group came from the aristocracy and influential families, who saw political advantages in joining with the new rulers.  Amongst the best known of these are the Banū Qasi (the Sons of Cassius) of Tudela, and the Banū ‘Amrūs of Huesca, who formed dynasties which dominated the area for two centuries after he Muslim conquest.  Cassius was a nobleman ruling over some territory on the northern Ebro.  His move secured his reign, that of his son Fortunus and of further descendants.  Still in the tenth century, the Banū Qasi proudly claimed their Gothic ancestry ruling over a by then rather large territory.

Far more numerous were converts amongst the “normal” Visigothic or Hispano-Roman Christians, who typically occupied low positions in the social hierarchy.  Most Christians, however, retained their old faith, though many adopted Muslim customs and learned Arabic, giving rise to the term Mozarabs (Spanish mozárabes from Arabic musta’rib), meaning “Arabised ones”.

As was the case with the Visigoths, the Muslims never constituted a large part of the Iberian population.  Whereas by 750 AD there were perhaps some 30,000 Muslims in the peninsula, the Iberian population at the time is estimated at anywhere from four to eight million.  However, conversions to Islam later added to this number.

Unlike in the rest of the Islamic world, the conquerors did not inhabit certain garrison towns.  Instead, settlement seemed to have been haphazard and determined by the interests of the settlers rather than by any overall scheme.  Moreover, the conquerors settled down as property owners, marrying the daughters of the previous Visigothic owners, as when ΄Abd al-΄Aziz bin Mūsā married Roderic’s widow, and when Sara, granddaughter of Wittiza, married two Arab husbands in succession and founded a dynasty which produced, among others, the tenth-century historian Ibn al-Qutiya.

As a result, the Andalusi Muslim invaders do not seem to have received the aţā (pension) that soldiers normally received in early Muslim society.  This also meant that there was no need for an elaborate bureaucracy to service and process the lists of those entitled to pensions.  Córdoba (Qurtuba in Arabic), with its rich agricultural hinterland, abundant mineral resources and its dense population of Muslims, became the capital in about 716 AD.

Slave market

Moorish settlements

On the whole, the Arabs settled in the main cities and the fertile irrigated areas of the Guadalquivir valley, the Levante around Murcia and Zaragoza and the middle Ebro valley.  In some places there were concentrations of men from the same tribes.  Other tribes were more dispersed.  Apart from the Qaysīs, almost all of them belonged to the Yemeni group of Arab tribes.  Some of these had lived in the steppe lands of Jordan and southern Palestine, while others came from Yemen proper.  Yemen was a land of cities, well-built villages and carefully tended farms, so that they would have been familiar with urban and agricultural life.  Many of them were second or third generation immigrants to North Africa and reared in urban settlements such as Qayrawān.

In contrast to the Arabs, the Berbers were widely distributed throughout al-Andalus.  Many settled in the central Meseta, Extremadura and the whole of the north and west apart from Zaragoza and its surroundings.  Toledo, Merida and Valencia lay in areas with a predominantly Berber population.  As these lands were less rich and inviting than the areas settled by Arabs, it has been suggested that the Berbers were obliged to accept inferior lands.  There is no evidence for this, even though there are indications that the Arabs held the Berbers in utter contempt.

The Berbers

The Berbers (from Latin barbari and Greek βάρβαρoi – barbarians) are the indigenous inhabitants of North Africa.  The Chronicle of 754 AD refers to Berbers as Mauri (Moors), being the inhabitants of the three Roman provinces of Mauretania (modern Morocco and Algeria), whereas the Arabs are called Saraceni.  Today, Berbers mostly refer to themselves as Amazigh, Imazighen or Imaziγens (“the free”) and speak a language that is quite distinct from Arabic or Indo-European languages.  Whereas both Arabic and Berber belong to the Hamito-Semitic language group, they constitute different branches.  The Berber script is probably based on a Punic variety of the Phoenician alphabet.

Modern Libyan Berbers

The Berbers were never a homogeneous ethnic or political group, encompassing a range of cultures and ancestries.  The one unifying force was the Berber language, Berber land and history.  Organised in tribes, some were nomadic and others living in settlements in areas with an agricultural base.  Some of them strongly opposed to Arab rule, others rapidly converted to Islam and became part of the Muslim forces.

According to Arab sources, at the time of the invasion the Andalusi Berber tribes were divided into two groups, the Butr and the Barānis.  The difference between Butr and Barānis seems to have had little effect on the politics of al-Andalus, unlike the murderous disputes between the Arab tribes.  Most of the Berbers who joined the Muslim conquest and settled in al-Andalus, came from the Butr group.  Berber troops received a share of the booty, but apart from a few exceptions, they do not seem to have occupied positions of political importance.

The Early Years (714-741 AD)

The governors (wālīs) of al-Andalus were in general appointed by the governor of Ifrīqīya in Qayrawān, or sometimes by the Caliph in person.  They were outsiders to the province with no local power base or following, which did not always mean that the wishes of the local people were ignored. If the locals did not like a governor, or if he attempted to undermine their autonomy and privileged fiscal status, they would write to the caliph, who would then send them someone who pleased them.

Between 714 and 756 AD al-Andalus was ruled by some twenty wālīs and their short tenures made it almost impossible to quell rising in-fighting among the various factions of the leadership.  Rivalries that had already existed in the Arabian Peninsula, such as between the Qaysīs/Mudar Arabs and the Yemenite Kalb, were exported abroad.  Factional frictions and strife in Damascus gave rise to bitter feuds in Iberia as well.

The Arrival of the Syrians

In 740 AD there was an anti-Arab revolt by the Berbers of North Africa, as the governor of Egypt ‘Ubayd Allāh bin al-Habhāb tried to impose land tax to increase the revenue required to pay the Syrian army, which was the backbone of the caliphate.  Further fuel was added to the fire by the practice of taking Berber children for the harems of the Umayyad elite.

In a very short period of time the entire Magreb (“the West”) had slipped from the control of the governors.  In response to the complete defeat of the local forces, the Caliph Hishām recruited a new army in Syria. On their way to North Africa, the Syrian soldiers were joined soldiers from Egypt. However, this military expedition against the Berbers was not a success.  The Syrian army was defeated and the 10,000 survivors fled north to Ceuta in modern Morocco.  Cut off from their homelands, they appealed to the wālī of al-Andalus for help, but he refused.

Meanwhile, a similar Berber uprising had started in al-Andalus.  The governor of Egypt had appointed ‘Uqba bin al-Hajjāji as-Salūlī as wālī of al-Andalus to implement his strict fiscal policies. The Andalusi Arabs were also fiercely opposed to him and in 740 AD there was a coup d’état in which as-Salūlī was forced to resign and was replaced by ‘Abd al-Malik bin Qaţan al-Fihrī, a venerable figure chosen by the people. However, the change was too late to prevent an uprising.  In the autumn of 741 AD, there was a major Berber revolt in the north-west and the Arabs were driven out of all the lands north of the Cordillera Central.  The Berbers marched south-west towards Córdoba and al-Fihrī was unable to resist them effectively.

Belmez castle near Córdoba, 1245 AD

Looking across the Strait of Gibraltar for help, al-Fihrī started to negotiate with the leader of the Syrians, Balj bin Bishr, who had fled to Ceuta.  The Syrians were so desperate that they were prepared to agree to almost any conditions which would save them.  Ibn Bishr accepted that he should fight the Berbers and then return to North Africa when the work was done. In the spring of 742 AD the Syrians were finally helped to cross into al-Andalus.

The battle-hardened Syrians joined the Andalusi Arabs led by al-Fihrī and defeated the Berbers in a fierce battle near Toledo. However, contrary to the deal made with al-Fihrī, the Syrians were reluctant to return to North Africa.  Relations broke down between Ibn Bishr and al-Fihri and Ibn Bishr launched a coup, which left the old wālī dead and himself in control.  Of course, this could not go unavenged and the Andalusi Arabs launched a counter-attack led by two of al-Fihrī’s sons, which left Ibn Bishr mortally wounded in the battle of Córdoba. Nevertheless, the Syrians remained in control of the city and chose a new wālī from their own ranks, who defeated the opposition.  Many Arabs of distinguished lineage were sold as slaves.

Reorganisation

In 743 AD a new wālī, Abū’l-Khaţţar al-Husām bin Dirār al-Kalbī, a member of the Kalb tribe, one of the leading Yemenite tribes of Syria, was sent from Qayrawān, apparently in response to a petition from al-Andalus of all parties who wanted peace.  Having first secured the release of the Arab and Berber captives, he set about providing for the Syrians who now clearly had to be accommodated in al-Andalus. Al-Kalbī settled the Syrians in a methodical way, each jund (army, based on tribal origins) in a different area.  The jund of Damascus was settled in Elvira, the jund of Jordan in Rayyu (Málaga and Archidona), the jund of Palestine in Medina Sidonia, the jund of Hims in Seville and Niebla, and the jund of Qinnasrīn in Jaén.  The jund of Egypt, possibly the largest, was divided between the Algarve (from al-Garb: ”Where the sun sets”) in the west and Murcia in the east.

It is said that the Syrians were given a third of the property of the local people to live off, though it is not clear whether revenues or actual lands to cultivate are meant.  The junds were obliged to pay the government a fixed sum from the revenues they collected.  In exchange for this livelihood, the Syrians had to do military service.

Discord

Unfortunately, the short period of peace would be disturbed as the Caliph Walid II in Damascus was murdered in 744 AD, which automatically led to a revolt against al-Kalbī, his appointee in al-Andalus. Being an outsider to al-Andalus with no power-base of his own, al-Kalbī was led to rely on and favour the Andalusi Yemenis.  The Qaysīs would not tolerate a governor who favoured their hated rivals and they chose the hard-bitten, brutal and fanatical as-Sumayl bin Hātim al-Kilābī as their new leader.

War could no longer be avoided. The Qaysīs were fewer in number, but stronger in the Córdoba area. Al-Kilābī set out to divide his enemies and managed to win over the leader of the Yemeni tribes which had been settled in Syria for a long time and may have felt more in common with other Syrians than with the Yemeni tribes of South Arabia.  The coalition revolted in April 745 AD and after two years of civil war, the rebels were finally able to eliminate al-Kalbī and his supporters.  Tawāba bin Yazīd, one of the original leaders of the revolt, was proclaimed the new governor.  On Ibn Yazīd’s death from natural causes in the following year, authority passed to Yūsuf bin ‘Abd ar-Rahmān, most probably another leader of the revolt against al-Kalbī.

Change

The events of 741-43 AD had profoundly changed the political character of Muslim Iberia.  It substantially increased the Arab, and especially the Syrian, element in the population, particularly in those rural areas in the south, which were to be the heartland of al-Andalus for centuries to come. Many Syrians, who came from Syria and the steppe lands of Mesopotamia in northern Iraq, had long-standing loyalty to the ruling Umayyad family.  

A significant number of probably non-Arab native Syrians or prisoners of war from other regions of the Muslim empire arrived.  These had no tribal affiliations and were mawālī (“clients or freedmen”) of the Umayyad family, owing their loyalty only to the ruling dynasty. Their presence would prove vital to the later success of the Umayyads in establishing themselves as rulers in al-Andalus.

In January 747 AD an outside candidate, Yūsuf bin ‘Abd ar-Rahmān al-Fihrī, was put forward by al-Kilābī, which was a shrewd choice.  Yūsuf al-Fihrī was already an old man and al-Kilābī could expect him to be a pliable instrument.  Al-Fihrī’s tribe the Quraysh (the Prophet’s tribe) had always occupied an intermediate position somewhat outside tribal divisions and could hope to attract loyalty from all parties.  In addition, the family had good contacts with the Berber groups. 

La Aljaferia, Zaragoza

Al-Fihrī started to exercise his power by excluding Yemenis from important positions.  When in 755 AD Zaragoza was besieged by Yemeni Arabs supported by dissident Berbers, however, al-Fihrī was powerless to help and Zaragoza was only saved by an expedition of Qaysī volunteers from the south. When some of al-Fihrī’s forces had proceeded towards Pamplona, where Basque raids had been reported, they were defeated.  The old governor could not come to their assistance, for at this point a certain ΄Abd ar-Rachmān bin Mu’āwiya of the Umayyad family had landed from Africa, and al-Fihrī had to hurry south, only to face the subsequent establishment of the Umayyad emirate in al-Andalus.

Under the newcomer ΄Abd ar-Rachmān Iberia would be the first region in the hitherto politically united “Arab Empire” to break decisively free from the secular authority of the caliphs in Damascus in 756 AD.

Arms and Armour of the Muslims

The nomadic desert Arabs had long been known to the peoples of Syria and Persia as fierce and warlike hit and run raiders, mounted upon fast horses or dromedaries. With little or no armour at all, they carried hide shields and were armed with spear, straight sword, and bow and arrows.  Much of their equipment, particularly armour, was acquired by looting settlements and forts on the northern borders of their domain.

Little is known about the first Umayyad contingents that crossed the Strait of Gibraltar in 711 AD.  We do know that the vast majority was infantry.  Logistic difficulties prevented the transfer of large cavalry units and mounted troops consisted only of a few light units, armed with spear and bow, riding without stirrups.

Berber soldiers were dressed in a turban, which covered most of the head, a tunic, wide trousers and a large woollen or cotton cloak (haik), which was wrapped around the body.  The clothing was mostly all in the same (dark) colour.  They were equipped with a spear, javelins, a sling with stones, a short sword, – usually carried on the back – a leather cuirass, cork-soled sandals and a small round leather or wooden shield.  

Before battle, many Berbers shaved their heads in a purification ritual.  Arab soldiers had adopted the North African dress code, but usually wore a cuirass, sword and oblong shield, in line with the Middle Eastern heavy infantry trends of that time.

Armour

Though early Islamic sources treat the coat of mail (dir’) as a standard piece of military equipment, which could be worn under a cloak (qabā’) to disguise it, probably only Berber and Arab officers had coats of mail or some other type of armour, with helmets of Byzantine origin.  Otherwise, for most soldiers this type of body armour would have been rare, as it was quite expensive.  They would have worn leather or felt padded armour or none at all.

Records show that in 706 AD coats of mail were sold for 700 dirhams, while spears and shields fetched between 50 and 70 dirhams.  There are also references to the practice of wearing two coats of mail (dir’ayn), the one under being shorter or even made of fabric or leather.

Helmets

The conical bayda-helmet is described by a contemporary Arab as “a helmet of iron which is composed of plates like the bones of the skull, the edges whereof are joined together with nails; and sometimes of one piece”.  This description is confirmed by some of the surviving Byzantine and Sassanid examples.  By Umayyad times, the conical helmet was forged in one piece.  Lamellar armour probably came with the Turkish invasion in 1163-91 AD.

Helmet with neckguard, from Al.Sufi's Kitab al-Sufar, 1224 AD

Tapering helmets were considered typically Turkish. Sometimes helmets had nose-pieces, but in general, nose-guards were unusual.  Another common feature was the mighfar, either an aventail, consisting of a piece of mail or fabric attached to the rim of the helmet, or a mail coif or hood hanging down behind to protect the neck.  More commonly, at least in the early period, it was a mail hood covering the top as well as the sides and back of the head.  The helmet would be put over the mighfar, which would hang down upon the shoulders.

Shields

Shields were worn by both cavalry and infantry, though in many illustrations figures are shown without them. They seem to have been small disks made of wood or leather and were certainly less than a metre in diameter.

Swords

The main offensive weapon was the straight, hilted sword (sayf), typically hung from a baldric as in early Roman fashion. There seems to be no evidence of the curved sword or scimitar, though Khurāsāni soldiers from eastern Persia in the early ninth century are said to have had curved scabbards.

The best swords came from India (hindī swords), followed by those made in Indian fashion in Yemen, which, along with Syria and Khurāsān, was the most famous centre of manufacture.  The best steel seems to have come from Sri Lanka, though that may have simply been the entrepôt for metal imported from further east.

Yemeni swords sold between 50 and 100 dinars, whereas at the bottom of the market, Egyptian swords could be had for just 10 dirhams.  Surviving examples of early Islamic swords are few in number and difficult to date.  A number discovered in northern Iran have straight blades, 100 cm in length, apparently without hand guards.

Horse-armour and Stirrups

Stirrups were not used in the Middle East or Europe until the seventh century and there is no reason to suppose that they were known to the Muslims at the time of the early conquests. Early stirrups were made of leather (loop-stirrups) or wood and broke easily.  Heavy cavalry mounts were protected with felt barding, while their riders wore coifs and hauberks and were armed with maces, straight swords and large daggers.  To better stabilise these heavier warriors on their horses, the leather loop-stirrups were replaced by metal ones.

Whereas in the ninth century it was widely believed that iron stirrups were a new invention, we know that in south-western Iran they had already been in use at the end of the seventh century.

Spears and Maces

Early Muslim spears (rumh, plural rimāh) almost always appear as an infantry weapon.  For want of stirrups, cavalry duels were typically fought with swords.  Mounted shock combat was not executed by heavy lancers, but consisted of charges with sword and light lance.

Spears were thrusting weapons with an iron point often used with devastating effect against cavalry charges.  In infantry encounters, before the fighting closed to sword length, early Muslim spearheads with long edged blades were used for slashing from side to side as well as thrusting straight ahead.

Javelins (harba, plural hirāb) were short with a long blade and were mainly used as a symbol of authority to be carried before a caliph or governor.  Together with spears, iron maces or bars were typical shock weapons.

Bows

Archers mostly fought on foot during the Umayyad and early ‘Abbasid periods.  From the ninth century, however, the mounted archer largely replaced the footman and became the backbone of the forces of the caliphate.  Abbasid generals increasingly used Khurāsāni and Turkish horse archers protected by leather cuirasses and round leather shields and armed with lance and short composite bow.

Though archery formed an important part of Muslim warfare, we know little of the bows (qaws, plural aqwās) and arrows of the early Muslim period.  There seems to have been a division between the lighter Arab and the heavier, more effective Persian bow, though it is not clear that there was any difference in design or construction.

The leather or wooden quiver is described as a quiver with a broad open top which tapers towards the bottom, so that when arrows are put in point downward their feathers are not damaged.  The length of a bowshot is put by the majority of lexicographers as between 180 and 240 m.   It is likely that the soldiers from Transoxania introduced a shorter composite bow like the type from Seljuk and Mongol times.

Arab Military Organisation and Tactics

There was no “Umayyad army”, but rather a number of different armies at different times and places. The most important army was the Syrian army (jund as-Shām – army of The Levant).

Warriors were supposed to provide their own arms and there seems to have been little difference between cavalry and infantry equipment.  Although the caliphs regarded the settled and urban Arabs as a more reliable source for troops, regular payment of the nomadic Bedouin (badawiyyūn) helped to settle and control them more easily.  It also creamed off the best fighting men so that rebellions were less of a threat.

A notable feature of the Muslim conquest was the founding of new garrison-towns. The first of these garrisons were known as amsars and many would develop into major cities.

Payment

Armies were maintained at the expense of the rest of the population by levying taxes. Non-Muslims paid a heavier tax rate, but were excused military service.  Next to a monthly pay (aţā), active soldiers were also rewarded by the booty (ghanīma) won by a man himself and the official share of spoils distributed by the government (fay).  Aţā stipends as well as military duties were inherited by a man’s son.

Ultimately, the pay structure was based on irafa units, in which men were grouped according to function and status, each irafa receiving a sum to be distributed amongst its members.

Umayyad silver dirham, 743 AD

Organisation

The last Umayyad Caliph, Marwan II (d. 750 AD), is credited with standardising his army, breaking up the conventional line of battle, usually 5 men deep and composed of unequal tribal contingents, and sub-dividing his forces into smaller units.  The smallest unit was the saff al-maqatir, consisting of 16 spear armed infantry or 8 archers.  The usbah was always composed of 32 men (2 infantry or 4 archers groups).  The miqnab had 64 men, the kardus 128, the jahfal 256, the kabkabah 512, the zumrah 1,024, the taifah 2048, the jash 4,096, the khamis 8,192 and the al-askar al-azam 16,384.  Phil Barker notes that this system was probably directly derived from the old Macedonian system and may never have been used in practice.  Each unit would have included heavy infantry, light infantry archers and heavy cavalry lancers.  Also a purely mounted force could be arranged in squadrons to give more flexibility.  The kardus was considered the most appropriate tactical unit for cavalry emerging from ambush.

Other units, which seem incompatible with the Macedonian system, were the khamsah (5 horsemen), ashrah (10-20 horse), tablikhanah (80), and alf (1,000 horse).  The katibah consisted of 500-800 infantry and the faylaq of 5,000.  The urafā (singular arīf) were in charge of ten to fifteen men and were responsible for assembling and paying the men, but not for leading them in battle. The urafā were also used as a military police in that they had to screen the men for their loyalty to the caliph.  They did not belong to the tribal elites and may have been employed precisely to restrict the authority the tribal nobility.

Officials whose function it was to count the number of troops and to make sure that all the men were at their battle-stations were called urād (singular: ārid).  These officials also decided who was fit to fight and to have their names recorded in the diwān (here: payroll).

Turkish influences grew in the ninth century as junds were augmented with ghulām (plural ghilmān, “pages”, originally largely used for menial tasks, but also occasionally armed and sent into battle). Later these ghilmān, purchased as slave children, were trained and converted to Islam and would serve the Abbasid caliphs as their personal bodyguard.  Ghulām heavy cavalry would be the finest in the Muslim world, marrying elements of steppe warfare (use of lasso and bow from horseback) with those of Arab lance and sword.

Tactics

The early Muslim armies relied on razzia tactics learned in Arabia.  The strategy was to weaken a foe with raids before a more serious attack or invasion.  Moreover, in the early days horses were too valuable to be used in frontal assaults.  Instead, the cavalry were reserved for flanking manoeuvres or to attack broken infantry.  In the early Muslim armies, the distinction between infantry and cavalry was blurred: the cavalry often fought on foot, while the infantry was transported on camels and horses to meet the enemy.  Mounts enabled them to travel more quickly to the scene of battle and to conduct reconnaissance expeditions.  Muslim infantry fought in the Byzantine fashion, placing better-armoured foot-soldiers in the front. Unlike Byzantine horsemen, however, Muslim cavalry would dismount and fight as infantry, using their lances as pikes.  Armies were conventionally drawn up for battle in lines, with a centre, a left and a right wing.  Lines could be straight or crescent-shaped, with the wings advanced and the centre further back.  The cavalry would be on the left and right wings and the infantry in the centre, lined up in three ranks: one with swords, one with spears and one with archers.

A line could also be curved with the centre advanced and the wings behind.  This was considered the worst variety and should be reserved only for cases of dire necessity.  In this case two squadrons of horse should be deployed to support the centre.

Commanders usually stationed themselves at the centre.  Curiously, they often chose to command sitting down, either on a carpet or on a chair.  This way of command may reflect Sassanid practice, but could also have been practical in that a stationary commander is easier for his subordinates to find than one who is dashing all over the battlefield.

Generally speaking, many conflicts started with the initial skirmishing between cavalry, whereas the decisive encounters were between foot soldiers, fighting first with long spears and later at close quarters with drawn swords.  After the battle had reached the close combat stage it was difficult for any commander to have much control over the course of events.  In the hand-to-hand fighting it was the small groups gathered around their banners which were the real fighting units, men from the same tribe or locality who knew each other and who were prepared to die with each other.

The infantry spear wall to defeat a cavalry charge became a common practice in the warfare of the Marwanid period (684-750 AD).  The infantry would kneel down pointing their spears at the enemy until it was almost on them.  Then they would rise and thrust their spears, advancing steadily against the opponent.  The ability of the Syrian troops to act in this fashion in the face of an oncoming cavalry charge is a sign of training and professionalism.  It was this discipline and training which gave them the advantage over enthusiastic but disorganised armies.  The most striking example of the effectiveness of these tactics  comes from the fall of the Umayyad Caliphate, at the Battle of Tell Kushaf, when the ‘Abbāsid troops under ‘Abd Allāh bin ‘Alī acted exactly in this way to defeat Marwān II’s army.

The next instalment of this series will focus on the consolidation of Muslim power in Iberia

Sources and Further Reading

• Armstrong, Karen, Islam, A Short History, Phoenix Paperback, Orion Publishing Group, London, 2001
• Barker, Phil and Richard Bodley Scott, DBM Army Lists 2,WRG, Devizes, 1998
• Carey, Brian Todd, Road to Manzikert, Byzantine and Islamic Warfare 527-1071, Pen and Sword, Barnsley, UK, 2012
• Collins, Roger, The Arab Conquest of Spain, Blackwell Publishers, Oxford, UK, 1989, 1994
• Collins, Roger, Caliphs and Kings (796-1031), Wiley- Blackwell, Chichester, UK  2012
• DESPERTA FERRO, No 7, Al Andalus, de la conquista a la quiebra del califato, Desperta Ferro Ediciones, SLNE,    Madrid, 2011
• Al-Andalus, Orto y ocaso de un Estado militarizado, Francisco García Fitz
• La batalla de Poitiers, Philippe Sena
• La supervivencia de los reinos cristianos, José Sánchez-Arcilla
• El guerrero andalusí, Yeyo Balbás
• Un ejército al servicio del califa , ۢۢAbd-al-Rahmān III y su política military, Vergilio Martínez Enamorado
• Yihad durante el califato de al-Andalus, Cristina de la Puente
• Tolerancia, convivencia y coexistencia en al-Andalus, ¿mito o realidad?, Alejandro García Sanjuán
• Las campaňas de Almanzor, Xavier Ballestín Navarro
• Fletcher, Richard, Moorish Spain, Orion Books Ltd., London, 2001
• Kennedy, Hugh, Muslim Spain and Portugal, Longman, London & New York, 1996
• Kennedy, Hugh, The Armies of the Caliphs, Routledge, Abingdon & New York, 2001
• MEDIEVAL WARFARE, Volume I, Issue 3:
• Iván Gimenez, The Rise of the Saracens
• Murray Dahm, Arab Sources on the Conquest of al-Andalus, lberto Raúl Esteban Ribas, The Battle of Guadalete
• Kai Grundmann, The fracture, downfall and remnants of the Visigothic Kingdom
• David Balfour, A Turning Point for Europe and Islam
• Gareth Williams, Fortifications of Western Europe, 700-1100 AD, Karawansaray Publishers Luxembourg, 2011
• Nicolle, David, Armies of the Muslim Conquest, Men-at-Arms 255, Osprey Publishing, Oxford, 1993, 2008
• Nicolle, David, The Moors - The Islamic West, 7^th-15^th Centuries AD, Men-at-Arms 348, Osprey Publishing, Oxford,  2001
• Nicolle, David, Armies of the Caliphates 862-1098, Men-at-Arms 320, Osprey Publishing, Oxford, 1998
• O’Callaghan, Joseph F., A History of Medieval Spain, Cornell University Press, Ithaca and London, 1983
• Phillips, William D. Jr. and Carla Rahn Phillips, A Concise History of Spain, Cambridge University Press, New York, 2010
• Russell Robinson, H., Oriental Armour, Dover Publications, Inc., Mineola, New York, 1967, 1995
• Slingshot No. 55, A Medieval Arab Military Manual,  Phil Barker, pp. 13-16, September 1974
• Slingshot No. 99, Origins of the Mamluk Military Institution, John Feilla, pp. 11-2, January 1982