With the death of the Prophet Muhammad in 632, leadership of the umma had passed to one of his earliest followers, Abu Bakr, whose piety won him the allegiance of the new faith’s followers in both Medina and Mecca. Beyond this core of believers, however, as noted earlier, a series of rebellions and revolts arose amongst various of the other tribes that—whether by negotiation or conquest—had joined the Prophet’s fledgling commonwealth. It was as part of the ensuing so-called ridda, or ‘wars of apostasy’, that the commander Khalid Ibn al-Walid initiated his campaign against the tribes of the Iraqi desert fringe in 633, as a concerted effort was made to bring all Arabs within the embrace of the faith. By 634, as we have seen, this process was sufficiently complete for the Muslim high command in Medina to turn its attention to the conquest of Roman Palestine that Muhammad had ordained, initiating the campaigns of that year. From the first, these endeavours were characterised by a high degree of central control of the Arab armies, which combined the warlike spirit and rapid mobility of the nomadic Bedouin with the more organised military traditions of the settled populations of the southern Arabian littoral.
After the defeat of the Roman army east of Gaza in early 634, and the subsequent failure of imperial forces to contain the second, more easterly Arab column, the invaders were able to establish control of much of Palestine, focusing their attacks on villages and winning mastery of the countryside. This in turn enabled them to isolate the cities of the region, a number of which began to submit to Arab rule, agreeing to pay tribute in return for security. By Christmas, Bethlehem was in Arab hands, making it impossible for the Christian clergy of Jerusalem to perform their customary pilgrimage. The Patriarch of Jerusalem, the hard-line Chalcedonian Bishop Sophronius, whose lament on the fall of the city to the Persians was cited earlier, complained of how: ‘As once that of the Philistine, so now the army of the godless Saracens has captured the divine Bethlehem and bars our passage there, threatening slaughter and destruction if we leave this holy city. ‘The Saracens’, he declared, ‘have risen up unexpectedly against us because of our sins and ravaged everything with violent and beastly impulse and with impious and ungodly boldness.’ By the end of 635, not only much of the Holy Land but also probably Jerusalem itself was under Arab control. The death of Abu Bakr and his succession by another of the early companions of the Prophet, Umar ‘al-Faruq’, or ‘the Redeemer’, did nothing to stem the Arab advance. Bearing the title of amir almu’minin, or ‘Commander of the Faithful’, Umar would eventually make a triumphal entry into Jerusalem, whence the remains of the True Cross had been spirited away to Constantinople. As the Chronicle of Theophanes records: ‘Sophronius, the chief prelate of Jerusalem, negotiated a treaty for the security of all of Palestine. Umar entered the holy city clad in a filthy camelhair garment. When Sophronius saw him he said “In truth, this is the abomination of the desolation established in the holy place, which Daniel the prophet spoke of.” With many tears, the champion of piety bitterly lamented over the Christian people.
The Arab armies now pressed on to the west bank of the Jordan, advancing on Roman Syria. In a series of engagements along the northern fringes of the volcanic Hawran plain, by the River Yarmuk, the Arabs defeated a large Roman army led by the Emperor’s brother Theodore. As Heraclius reconfigured the forces available to him in southern Syria, resistance stiffened somewhat, but in a decisive encounter the Arabs broke the East Roman field army in open battle between Emesa and Damascus. The remnants of the Roman forces were obliged to withdraw to defensive positions in Cilicia, the foothills of Armenia, and northern Mesopotamia. An Arab raid across the Euphrates targeted the region’s famed monastic communities. As one near-contemporary Syriac source records: ‘the Arabs climbed the mountain of Mardin and killed many monks there’. Another account relates how ‘these Arabs went up to the Mardin mountains and they killed there many monks and excellent ascetics, especially in the great and famous abbey on the mountains above Rhesaina’.
In response, the commander of the Roman forces in Upper Mesopotamia attempted to buy peace, leading to his dismissal by the Emperor. The Roman refusal to pay elicited a concerted Arab response, and in 636 the amir Umar’s armies advanced in force across the Euphrates into Upper Mesopotamia. The civic notables of Edessa and Harran surrendered, but first Tella and then Dara were taken by storm. In the latter, we are told, every single Roman found in the city was executed. Rather than face this fate, the inhabitants of Amida and a number of other cities rapidly came to terms. Extending their control over the old Roman-Persian frontier zone, the Arabs now struck east into Sasanian territory, sweeping across the ‘black lands’ of the Iraqi alluvium and bearing down on the capital of the shahs at Ctesiphon. A massive Persian army was assembled to meet them under the leadership of the general Rushtam. In 637 this force, which included significant Transcaucasian contingents and which the author of the Armenian History reckoned at some eighty thousand men, managed to drive the Arabs away from Ctesiphon and propel them back across the Euphrates, inflicting the first significant defeat on the Arabs since their campaigns of conquest had begun.
From the length and breadth of Arabia the forces of the umma were rallied, and in January 638 at al-Qadisiyya, near the old Nasrid capital of al-Hira, the massed ranks of the Arab and Sasanian field armies clashed. The result was a cataclysm for the Persians. Broken in open battle, a retreat rapidly became a rout. As the Armenian History records, ‘The Persian army fled before them, but they pursued them and put them to the sword. All the leading nobles were killed, and the general Rushtam was also killed.’ As the Arabs once more focused their attention on Ctesiphon, in the following year a desperate attempt to evacuate the Persian high command and the royal treasury ended in disaster as the baggage train was ambushed and its precious cargo seized. Shah Yazdgerd III, a grandson of Khusro II who had ascended the throne in 632, managed to flee east to the rocky fastness of the Zagros Mountains, but there was little he could do to save his capital, which the Arabs were now able to occupy almost unopposed.
In the west, the Romans took advantage of the Arab preoccupation with Persia to launch raids on Arab-held northern Syria and northern Mesopotamia, whilst, on the Palestinian coast, the garrison at Caesarea continued to hold out, receiving supplies by sea. The Arab general Iyad was, however, able to drive back these imperial assaults, which are recorded to have done much to alienate the local population from the Roman army, and the commander of the Muslim forces in Syria, Mu’awiya, finally overcame the defences of Caesarea and put both the garrison and the population to the sword. ‘The city’, we are told, ‘was plundered of vast quantities of gold and silver and then abandoned to its grief. Those who settled there afterwards became tributaries of the Arabs.
With the loss of Caesarea, all of Syria and Palestine was now in Arab hands. In 640 Muslim forces advanced into south-western Armenia while the general Amr ibn al-As, pursuing retreating Roman forces out of Palestine, initiated the conquest of the Nile valley. Aided by reinforcements directed to him from Medina, the general was able to take first Oxyrhynchus, then in 641 the Roman military base at Babylon, where he seized abandoned siege engines, before fighting his way up towards the Nile Delta, attacking the estate complexes of the local aristocracy where resistance was seemingly concentrated. In 642 the general’s armies initiated the siege of Alexandria. Parleying on behalf of the imperial authorities, the Patriarch Cyrus negotiated an armistice whereby the Arabs received an annual sum of 200,000 solidi and the city was demilitarised, the Roman army and administration being obliged to withdraw to Cyprus. A subsequent attempt in 646 on the part of the Roman general Manuel to reoccupy the city and use it as a bridgehead for the reconquest of Egypt (to which we shall return in the following chapter) impelled the Arab army to enter the city and slaughter the garrison. Manuel and the Patriarch Cyrus fled to Constantinople.
As the final remnants of Roman resistance in the Near East beyond Anatolia, Asia Minor, and the islands were snuffed out, the Arabs pressed ahead with their conquest of the crumbling Sasanian Empire. Assaults were launched across the Persian Gulf in 641 and into the Zagros Mountains. In 642 the Arabs advanced through the Zagros into the Parthian territory of Media. With the defeat there of the Persian army at Nihawand, the Arabs spread over the remaining regions of the Persian world in what was essentially a mopping-up exercise, breaking down the resistance of the regional lords and princes one by one. Finally, in 652, Yazdgerd III was assassinated as he attempted to flee to the steppe, the last son of the House of Sasan ignominiously a fugitive in the realm of Turan. The medieval Persian epic the Shahnameh, or ‘Book of Kings’, describes how the assassin, a miller by the name of Khusro, approached Yazdgerd ‘his heart filled with shame and fear … his cheeks were stained with tears, and his mouth was dry as dust. He came up to the king like someone about to impart a secret in a man’s ear and plunged a dagger beneath his ribs. The king sighed at the wound, and his head and crown fell down to the dust, beside the barley bread that lay before him.’
REASONS FOR ARAB SUCCESS
What had begun as an attempt on the part of Muhammad’s followers to claim and occupy what they regarded as their divinely promised patrimony in Palestine had therefore escalated into an extraordinary wave of conquests; these wiped the ancient empire of Persia off the face of the map and once more drove Heraclius and the Romans back behind the mountains of the Taurus and ante-Taurus that defended the Anatolian plateau. From their initial focus on Palestine, and on uniting all Arabs within the embrace of the new faith, the armies of Abu Bakr and Umar had acquired a momentum of their own: they would continue to march on and conquer until either they were defeated or the Day of Judgement came. The Arab armies were clearly aided in their success by the relative exhaustion of the two great superpowers that they had set about dismembering; it was the perspective of the author of the Armenian History, for example, that it was the destructive pride and overweening ambition of Khusro II which had opened the gates of Hell and unleashed the Saracen scourge. In Persia, as we have seen, political circles in Ctesiphon had gone into meltdown as a result of Heraclius’ victorious campaign of 628. Heraclius’ daring descent into Persian territory and his ravaging of the lands to the north of Ctesiphon may well have done lasting damage to the agricultural resources and administration of a region that had been the economic powerhouse of the Sasanian state. Political paralysis and administrative chaos may also have critically limited the ability of the Persian authorities to respond to the Arab threat.
Likewise, Heraclius, it should be noted, had gambled everything on his last throw of the dice against Persia. Already drained of their resources by the demands of his war effort, or reduced to ruin by Persian assault, the cities of Asia Minor simply may not have been in a position to finance and support a sustained defence of Syria, Palestine, and Egypt, where the restoration of Roman rule in the aftermath of the Persian withdrawal of 628–30 is likely, in any event, to have been largely symbolic at the time when the Arab armies began to appear: long-standing traditions of Roman control had been fractured and disrupted and were yet to be fully restored. Indeed many of the ‘Roman’ armies that the Arabs encountered are likely to have been little more than gendarmes, or local levies, hastily gathered together by civic notables and landowners to defend their cities and estates.
Moreover, it could be argued that the Eastern Roman Empire’s extensive desert frontier was in any case its Achilles heel. The Romans had never successfully resolved the problem of how to police and defend the frontier: walling it off was impossible; maintaining security through the services of conflicting networks of client chiefs had proved untenable; and relying upon the services of a single client chief had been unworkable. At the end of the day, all that had perhaps rendered Roman imperialism practicable and sustainable in Syria and Palestine had been the absence of a concerted threat from along the desert fringe. The Palmyrene revolt of the 270s had demonstrated how fragile Roman control of the region might be if faced with such a challenge. ‘Divide and rule’ thus remained the key to Roman survival. Given the objective military and geographical circumstances, the unity that the religion of Muhammad provided to the tribes of north-central Arabia, and the cultic and military focus towards Roman Palestine that the Prophet ordained, may of itself have been sufficient to seal the fate of Roman power in the East. Constantinople, too, suffered political problems of its own: the death of Heraclius in 641, and the power struggle that ensued, did much to distract attention from the Arab march on Alexandria and to detract from the effective coordination of Roman resistance. Likewise, disaffection on the part of Jewish and other religious minorities, and alienation on the part of the peasantry and the poor, are also likely to have played their role in encouraging communities to come to terms with the invaders in their midst.
But the triumph of the Arab armies was also the work of the Arabs themselves. The combination of Beduin mobility and the more organised military and political traditions of the sedentary populations of the southern Arabian littoral, such as the Yemenis, created a formidable war machine, whilst the wealth of the Roman and Persian territories provided a clear material incentive for military expansion (especially for tribesmen whose ability to profit from trade with the sedentary empires to the north had perhaps been disrupted by warfare). Tactically, the strategy we see at work in Palestine in 634, of Arab armies attacking ‘soft targets’ such as villages, engaging in conspicuous massacres of the rural population, and then offering terms to the leaders of civic communities, promising security in return for tribute, was a psychologically canny one that permitted strikingly rapid conquests and the avoidance of entanglement in lengthy sieges. When faced with resistance from cities such as Dara, the favoured Arab strategy was simply to storm them, throwing men at the walls until enough of them got in, rather than bedding down to a long drawn-out war of attrition. The brutality shown to the inhabitants of those cities that did resist sent a clear message to the leaders of other communities that it would be in their manifest interest simply to surrender and ‘pay tribute out of hand’, as the Qur’an directed, rather than risk suffering a similar fate. As the Armenian History records of the Arabs, ‘then dread of them fell on the inhabitants of the land, and they all submitted to them’.
It is, however, the ability of the Arab commanders to storm cities—to order their warriors to advance and attack and advance again until a city fell, irrespective of the casualty rate—that perhaps alerts us to the ultimate factor behind Arab success: zeal. Driven on by religious fervour and certain of paradise, the Arab armies appear to have had a far higher ‘pain threshold’ and to have enjoyed morale superior to that of either their Persian or Roman adversaries. Dead or alive, Allah would reward them. Confident in the power of their God, the authority of the Prophet, and the imminence of Divine Judgement, the forces of Islam swept all before them. Both Roman and Persian armies, by contrast, had recently experienced defeat, and, as the military theorist Carl von Clausewitz realised, in warfare it is morale that is often the decisive factor.
Between the initial raids into Roman Palestine in the early 630s and the death of Yazdgerd III in 652, the ‘Arabs of Muhammad’ and their allies had not only managed to drive the Roman forces of Heraclius back into Anatolia and Asia Minor but had also destroyed once and for all the ancient empire of the shahs of Persia. Of the two great powers that had for so long dominated the politics and culture of western Eurasia, one was no more and the other was palpably on the ropes. In 652–3 the Arabs extended their control into the Transcaucasus, exacting oaths of loyalty from the Armenian prince Theodore Rshtuni and his vassals, who had hitherto fought on Constantinople’s behalf.
This was the necessary precursor to the intensification of the jihad against Byzantium, for it secured for the Arabs control over the lines of east-west communication across the valleys of Armenia that led onto the Anatolian plateau. As the Armenian History records:
In that same year the Armenians rebelled and removed themselves from [allegiance to] the Greek kingdom and submitted to the king of Ismael. T’eodoros, lord of Rshtunik, with all the Armenian princes made a pact with death and contracted an alliance with hell, abandoning the divine covenant. Now the prince of Ismael spoke with them and said: ‘Let this be the pact of my treaty between me and you for as many years as you may wish. I shall not take tribute from you for a three-year period. Then you will pay tribute with an oath, as much as you may wish. You will keep in your country 15,000 cavalry, and provide sustenance from your country: and I shall reckon it in the royal tax. I shall not request the cavalry for Syria; but wherever else I command they shall be ready for duty. I shall not send amirs to [your] fortresses, nor an Arab army—neither many nor even down to a single cavalryman. An enemy shall not enter Armenia; and if the Romans attack you I shall send you troops in support, as many as you may wish. I swear by the great God that I shall not be false.’ In this manner the servant of Anti-Christ split them away from the Romans. For although the emperor wrote many intercessions and supplications and summoned them to himself, they did not wish to heed him.
As this passage reveals, however, and as both the Arabic and non-Arabic sources agree, the tightness of Arab rule over the recently acquired territories varied enormously, and would continue to do so to the end of the seventh century and beyond. In general terms, Arab rule was at its most secure in lowland zones and inland areas, such as the jazira of Mesopotamia. The Arab armies of conquest were clearly at their least confident when fighting on mountainous terrain. As a result, all they were really able to acquire from Theodore Rshtuni and the Armenians was a loose acknowledgement of Arab suzerainty and a promise to provide a military levy. The Arabs would only intervene in Armenia if the Romans, under their new Emperor, Constans II, did. What the Arab commanders were effectively promising was a guarantee of Armenian autonomy, something that the princes of the region had hitherto sought to achieve by playing off the rival powers of Rome and Persia.
Likewise, at no point in the seventh century did the Arabs manage to conquer or occupy by force the mountains of the Lebanon, which, as shall be seen, were to be home to bands of Christian insurgents known as the ‘Mardaites’ who maintained a campaign of guerrilla warfare against the Muslims and who would periodically descend from Mount Lebanon to attack Arab forces in the plains and cities below. To the east, the Zagros Mountains remained outside direct Arab control and would provide a place of refuge for every malcontent, heretical or disaffected group that early Islam would generate. The Arabs’ grip on coastal zones was similarly precarious. As noted in Chapter Seven, supplied by sea, the city of maritime Caesarea in Palestine had resisted Arab conquest for years. There are indications that many of the cities and communities along the Syrian coast drifted in and out of Arab rule over the course of the seventh century, paying tribute to the Arabs when they perceived them to be strong, but turning off the tributary tap when they sensed the power of the Muslims to be waning. In the late seventh century the inhabitants of Cyprus would pay tribute to both the Arabs and the Romans, sending cargoes of copper west to Constantinople and east to Syria.2 In Egypt the original deal agreed between the Patriarch Cyrus and Amr ibn al-As was that Alexandria would pay tribute and the Nile Delta effectively be demilitarised with the evacuation of the Byzantine garrison. Only Constantinople’s subsequent attempt to reoccupy the city had led to a more forceful assertion of Arab might.
Even within the lowland zones and inland, however, the rapidity of the Arab advance, and the willingness of the conquerors to cut deals with the leaders of provincial society, offering security and religious and property rights in return for the payment of tribute (as recorded for example, with respect to the Christian Arabs of al-Hira in 633), necessarily meant that for much of the seventh century, Muslim rule rested relatively lightly over the lands of the Near East. So long as tribute was paid to the Arab authorities, then long-established local elites could get on with running the communities they had so long dominated. A classic example of this emerges from Egypt where, according to the Chronicle of John of Nikiu, the first Arab-appointed Prefect, or governor of the city of Alexandria, was a certain John of Damietta, who had previously been the Byzantine general in charge of Roman resistance to the Arabs. In the mid-eighth century, the great Christian Orthodox theologian John of Damascus could claim descent from a family of Roman imperial administrators who had continued their mandarin lifestyle under their new Arab masters. In terms of administration, economic activity, and even, to some extent, religion, life in the conquered territories continued as it had before. Essentially, at a grassroots level, the same people were effectively collecting the same taxes in the same manner; the difference was that thereafter these taxes were being handed over to the Arabs rather than to the representatives of the Roman Emperor or the Persian shah. As in the post-Roman west, the provincial-level regional aristocracies and elites remained in place. It was at the level of the grandest families, most closely implicated in imperial rule or most heavily dependent on the trans-regional structures of empire, that discontinuity was at its most apparent, with such families either fleeing or being snuffed out.
The high degree of continuity in social and administrative structures evident in the wake of conquest was also facilitated by the fact that the Arab armies, carefully supervised by the high command in Medina, were, for the most part, cantoned separately from the populations over whom they now ruled. These armies were living in newly established garrison towns such as Fustat in Egypt (modern Cairo), or Kufa and Basra in Iraq, where they received stipends (ata) derived from the tribute of the local population. Only in Syria do rulers and ruled appear to have lived cheek by jowl, but here in any case there were long-standing Arab populations. The conquerors were thus able to maintain their identity, effectively living as a separate and privileged military caste feeding off the resources of their Christian, Jewish, or Zoroastrian tributaries. The conversion of subject populations was not a priority, and in order not to alienate either the subject peoples or the non-Muslims who had fought in their own ranks in the armies of conquest, the early leaders of the umma such as Umar (r.634–44) and his successor Uthman (r.644-56) adopted the religiously ambiguous and multivalent title of amir al-mu ’minin—‘commander of the faithful’—rather than anything more stridently or aggressively Islamic.
The documentary papyri that survive from Egypt in the seventh century convey much of the paradoxical nature of this nascent ‘early Islamic’ world. The densely inhabited and intensively cultivated landscape of the Nile Valley ought to have been amongst the easiest terrains for the Arab conquerors to dominate. Yet they barely register in the documents we have concerning rural life. Instead day-to-day life was dominated both by members of the local Christian elite, bearing the old Roman title of ‘pagarchs’ (pagarchoi), and the personnel of the Miaphysite Church; together, these took up much of the slack left by the withdrawal of Roman governors, the flight of the Chalcedonian patriarch, and the disappearance of the great households of members of the upper echelons of the senatorial aristocracy, such as the Flavii Apiones of Oxyrhynchus. Indeed, from the ‘bottom up’, for much of its post-Heraclian history, seventh-century Egypt would have looked very much like a Coptic Christian theocracy in which society was dominated and taxes were collected by the Church and by Christian notables—before the lion’s share of these revenues was then handed over to the Arab Muslim administration and its army based at Fustat, whose demands for tribute were pressing and insistent. A similar situation pertained among the Christian communities of Iraq, where the disappearance of the institutions of the Sasanian state led to a major expansion of the juridical and administrative authority of the episcopacy and city-level Christian elites. In both Arab-controlled Egypt and Iraq, ‘Muslim rule’ paradoxically meant stronger bishops.
As the documentary evidence from Egypt and Iraq reminds us, administration was not just continued by the same sorts of people in the same manner; it continued in the same languages, above all Greek (with some Syriac and Coptic) in the former Roman provinces, and Syriac and Pahlavi (Middle Persian) in the ex-Sasanian ones. Roman gold coins and Persian silver coinage continued to circulate in their respective currency zones, with the Muslims minting mock-imperial coinage, albeit ultimately devoid in the Roman case (perhaps from the 660s onwards) of Christian religious imagery. It must have seemed to many that although some seismic shift had clearly occurred, the new world order looked remarkably like the old. To all intents and purposes there existed a ‘sub-Roman’ Arab Empire in the former Roman territories, and a ‘sub-Sasanian’ one in the lands hitherto subject to the shah. Indeed many clearly anticipated an imminent Byzantine counter-strike, whereby the provinces would be restored to the empire of Christ just as Heraclius had won them back after twenty years of Persian occupation in the late 620s. It was a sense shared by certain of the Arabs themselves: a proverb cautioned that ‘Islam has started as a foreigner [to all lands] and may again become a foreigner, folding back [on Mecca and Medina] like a snake folding back into its hole.’