A History of Warfare

by John Keegan

  The horse and human ruthlessness together thus transformed war, making it for the first time ‘a thing in itself’. We can thenceforth speak of ‘militarism’, an aspect of societies in which the mere ability to make war, readily and profitably, becomes a reason in itself for doing so. Yet militarism is a concept that cannot be applied to any horse people, since it presumes the existence of an army as an institution dominant over but separate from other social institutions. There was no such separation among Attila’s Huns, nor would there be among any horse people until the Turks espoused Islam. The fit and adult males of a horse people were the army, but not the sort of army by which Turney-High measured a society’s position above or below the ‘military horizon’. All the horse peoples who beat a path of conquest off the steppe into the lands of civilisation fought ‘true war’ by all the tests — lack of limitation in the use of force, singularity of purpose and unwillingness to settle for anything less than outright victory. Yet their warfare had no political object in the Clausewitzian sense, and no culturally transforming effect. It was not a means to material or social advance; indeed, precisely the contrary, it was the process by which they won the wealth to sustain an unchanging way of life, to remain exactly as they had been since their ancestors first loosed an arrow from the saddle.

  None of the horse peoples that kept a base on the steppe ever willingly altered its habits; at best, their most successful leaderships became absorbed as a ruling class in the settled societies they conquered, without abandoning their nomadic ethos, and that was true even of the islamicised Turks, despite the degree to which they sustained Byzantine forms of government within their empire after their capture of Constantinople in 1453. The Mameluke system, despite the degree of autonomy the Mamelukes enjoyed, was, as we have seen, merely a means of perpetuating the horseman’s way of life, with all the wealth and honour that military power brought in its train. Most horse peoples, moreover, for most of the time that the frontiers of China, the Middle East and Europe lay open to their assault, succeeded neither in finding individual employment nor in imposing themselves as conquering rulers on more advanced societies. The steppe life remained rooted in war, but the warpath was a hard one, blocked in almost every direction by the defences of states which fought all the more fiercely to keep the horse peoples confined within the limits of the steppe. They had learned the ghastly consequences of dropping their guard.

  In the wake of the Huns’ disappearance, there remained no strong horse people in contact with civilised powers in Europe or the Middle East. The most significant were the Ephthalites, the so-called White Huns, who seem to have been displaced on to Persia’s northern border by the Hsiung-nu when both lived far away on the fringe of China.52 The Ephthalites had at least one spectacular success, in part because Persia’s energies were devoted to waging its endemic war with Byzantium, but in 567 the Persians eventually succeeded in beating them off; deflected eastward, they appear to have made their way into Hindu India and planted the roots of future Rajput power.

  Meanwhile Byzantium was holding at bay various horse peoples extruded westward by the perpetual tribal discords in the steppe heartland. These included the Bulgars and the Avars, the former pushed by the latter, who in turn were displaced by the growing power of the Turks. The Bulgars eventually settled in the Balkans, where they were to be a cause of trouble until the Ottomans finally brought them under control. The Avars migrated to Hungary, caused widespread disruption, and, though occasionally allied to Byzantium, laid siege to Constantinople in 626; with Persian help, they very nearly succeeded in entering it. They were repulsed but remained a powerful source for harm until they were eventually overcome by Charlemagne in the eighth century; their place was then taken by the Magyars, the last horse people to migrate from the steppe into central Europe.

  Until they were turned westward, the Avars may, however, have already learnt the habit of warring against imperial power, if they can be identified with the Juan-Juan who at the beginning of the fifth century had fallen out with the dynasty known as the Northern Wei in northern China. The Northern Wei were one of a group of sinicised steppe people who, on the fall of the unified Han empire in the third century, had ruled north of the Yangtse; the circumstances of their rise to power are so complex that the period is known as the ‘Sixteen Kingdoms of the Five Barbarians’ (304–439). By 386, however, the Northern Wei had emerged as dominant and begun to reunify northern China. In the process, they came into conflict with the Juan-Juan people living above the Gobi Desert, and expelled them from their territory. They were helped in this by a subject class of the Juan-Juan who worked for them as ironsmiths; these were the Turks. The Turks bore a recent grudge: after helping their masters to put down a revolt by another subject tribe, their chief sought as his reward the hand of the Juan-Juan chief’s daughter. He was refused. The Northern Wei offered him a noble daughter of theirs and then together they fell on the Juan-Juans, who were broken. The Turks took their territory, and their chief the title of khagan or khan, which later became that of most steppe rulers.

  The Turkish khan and his successors founded a great empire. They were ‘the first barbarians to create a kingdom so extensive as to touch at different points the four great civilised societies of the day: China, India, Persia and Byzantium’.53 By 563 they had advanced as far as the Oxus River, on Persia’s eastern border, and with the Persians they made common cause against the Ephthalites. By 567 the Turkish khan, Istemi, had got a share of the Ephthalite lands as the spoils of victory. In the following year he was an important enough figure for the Byzantine emperor, Justin II, not only to receive an embassy from him but to send one of his own in return on the enormous journey to the centre of the steppe. Fatally, the Turks then fell to quarrelling over authority within their empire, the besetting fault of horse peoples and the main cause of dissolution of their structureless polities. In this period of disunion they lost much of their eastern territory to the rising power of the T’ang dynasty in China, which extended its control as far as the Oxus River by 659. By then, however, the Turks had encountered a new enemy in the west which was also reaching out into the steppe, making great conquests and striking to contest control of Central Asia with the Chinese. In the course of the next century and of this struggle for power for the steppe heartland, which culminated in the battle of the Talas River in modern Kirghizstan in 751, the Turkish empire was to be overthrown.54 The new enemy were the Arabs.

  Arabs and Mamelukes

  The Arabs were not a horse people, though they were to become their principal employers in the civilised world. If only for that reason they would deserve the attention of military historians, but they deserve it for very much more. First, at the time the Turks encountered them, they were just completing one of the greatest campaigns of conquest in history, a campaign which had transformed an almost unknown tribal group from the deserts of inner Arabia into rulers of most of the Middle East, and the whole of North Africa and Spain. They had shaken the Byzantine empire, destroyed the Persian, and founded one of their own. Only Alexander the Great — and he had been the first long-range conqueror in history — had captured territory of a similar extent and with an equivalent rapidity. Their pattern of conquest, moreover, was creative and unifying. Though the Arabs would later fall out among themselves, the original empire was a single whole and one which rapidly dedicated itself to the arts of peace. The Arab rulers were to become great builders, beautifiers and patrons of literature and science. Unlike the rough horse peoples whom they were later to enlist as soldiers, they showed an astonishing ability to emancipate themselves from the campaigning way of life, to embrace civilisation and to cultivate sophisticated manners of thought and behaviour.

  More than that, however, they stood out among military peoples because they demonstrated an ability to transform not merely themselves but warfare itself. There had been military revolutions before, notably those brought by the chariot and the cavalry horse. The Assyrians had established the principles of
military bureaucracy, on which the Romans had built. The Greeks, as we shall see later, had evolved the technique of the pitched battle, fought to the death on foot. The Arabs transfused warfare with a new force altogether, the force of an idea. Ideology, it is true, had played its part in warfare before. The Athenian Isocrates had urged a Greek ‘crusade’ against Persia during the fourth century BC, in which the idea of liberty was implicit.55 During the Emperor Theodosius’s struggle with the Goths in 383, the Roman Themistius had argued that the strength of Rome lay ‘not in breastplates and shields, not in countless masses of men, but in Reason’.56 The Jewish kings had fought in covenant with their single and almighty God, while Constantine had invoked the image of the cross to bring him victory against a pretender at the battle of the Milvian Bridge. These had all been muted or limited ideas none the less. Though the Greeks took pride in their freedom and despised the subjects of Xerxes and Darius for their lack of it, their hatred of Persia was at root nationalistic. The appeal to reason lacked force at a time when Rome’s armies were already heavily barbarised, their ranks filled with savage soldiers who had never heard reason’s name. Constantine, moreover, was not yet a Christian when he uttered the appeal to conquer in the sign of the cross; and while the warrior kings of Israel may have drawn strength from the old Covenant in their small and local wars, the Christians of the new Covenant were to agonise for centuries over the issue of whether warmaking was morally permissible or not. Christians, indeed, have never found unanimity in the belief that the man of war may also be a man of religion; the ideal of martyrdom has always been as strong as that of the justified struggle and remains strong to this day. The Arabs of the conquest years were not caught on that crux. Their new religion, Islam, was a creed of conflict, that taught the necessity of submission to its revealed teachings and the right of its believers to take arms against those who opposed them. It was Islam that inspired the Arab conquests, the ideas of Islam that made the Arabs a military people and the example of its founder, Muhammad, that taught them to become warriors.

  Muhammad was not only a warrior himself, who had been wounded in a battle at Medina against the men of Mecca in 625. He preached as well as practised war. In his last visit to Mecca in 632 he laid down that, though all Muslims were brethren and should not fight each other, they should fight all other men until they said ‘There is no god but God’.57 The Koran, which Muslims believe to have been taken down from his words by disciples, elaborates this command extensively. Even more specifically than Christ had done, Muhammad insisted that those who accepted the word of God thereby formed a community (umma) whose members owed each other responsibilities; thus it was not enough simply to avoid fratricide: Muslims were under an obligation to do positive good to less fortunate Muslims by assigning a certain portion of their income to charity; they also had a duty to care for each other’s consciences. Beyond the umma, however, the obligation was reversed: ‘O you who believe, fight the unbelievers who are near to you.’58 This was not a call to forcible conversion. Non-believers who were prepared to live under Koranic authority were positively entitled to protection and, in strict theory, those outside the umma who kept the peace ought not to be attacked. In practice, however, the bounds of the umma came to coincide with the House of Submission (Dar el-Islam), while outside inevitably lay the House of War (Dar el-Harb). Against the House of War, Islam fell into conflict from the moment of the Prophet Muhammad’s death in 632.

  Conflict with the Dar el-Harb shortly became jihad, ‘holy war’. It was not simply the command of the Prophet that made Muslims wage it so successfully, though wildly successful they were as warriors. There are at least two other reasons to explain the ease of their early victories. First, there is no conflict in Islam between devotion and material well-being. Christ, to the great moral perturbation of his followers ever afterward, held up poverty as a holy ideal. Muhammad, by contrast, had been a merchant, had a keen understanding of the value of wealth, properly used, expected the umma to accumulate it and saw it as a means of doing good, both collectively and individually. He himself raided the caravans of the rich, unbelieving merchants of Mecca, and spent the loot to further his cause. That was an example that his holy warriors followed in their assault on the rich kingdoms of Byzantium and Persia.

  Secondly, Islam dissolved the two principles on which war had so often been fought before: territoriality and kinship. There could be no territoriality in Islam, because its destiny was to bring the whole world to submission to the will of God. Islam means submission and ‘Muslim’, formed from the same word, someone who is under it. Only when the whole of the House of War had been brought within the House of Submission would Islam’s destiny be complete. Then all men would be Muslims and therefore also brothers. In practice, the first Arab Muslims, still enmeshed by the powerful clan kinships of the desert world, resisted the principle of brotherhood, so that converts from beyond their tribes had to accept the status of clients (mawali) for a time.59 Eventually, however, it was to prove one of the glories of Islam that it dissolved barriers of race and language to an extent that no religion or empire — and Islam comprehended both concepts — had succeeded in doing before.

  Another factor greatly assisted the Arabs who, in the last years of Muhammad’s life, set out to extend the boundaries of Islam: the kingdoms on which their force fell were in decline. Byzantium had expended much of its strength in resisting the Avars on its northern boundaries; more wearingly, it had been engaged since the beginning of the seventh century in the last of its great wars with Persia (603–28), a war that exhausted both empires. As for Persia, historically a great power, it had also suffered historically from the weakness of its geographical position, poised between the steppe on the one hand and the rich lands of the Middle East on the other. Before the rise of the horse peoples, it had frequently been able to profit from decline or collapse on its western borders to extend its imperial frontiers. A millennium before, in Alexander the Great, it had met an opponent of such towering skill and determination that its native dynasty had been supplanted and its imperial possessions divided among his generals. Seleucus, the Alexandrian general to whom the Persian heartland fell, succeeded in sustaining Hellenistic power but not in Hellenising Persian society. His empire eventually passed to the Parthians, another Iranian people who had begun in Central Asia. Though a horse people — it was their cavalry prowess that overthrew the Seleucid infantry — they readily assimilated to civilisation, founded a great empire, and, between the first century BC and the beginning of the third AD, were Rome’s chief enemies in the east. The wars between Persia and Rome were often marked by Persian victories: the campaign of 363, in which the emperor Julian the Apostate was killed in Mesopotamia, was almost as great a disaster for Rome as the Gothic victory at Adrianople fifteen years later. But the strain of constant warfare depleted Persia’s wealth, manpower and resilience, and the empire was thereafter increasingly harried by nomads on their steppe frontier.

  When, therefore, in 633 an Arab army invaded northern Mesopotamia, the Persian army was not what it had been; neither was the Byzantine. Audaciously, the Arabs chose to operate against both simultaneously and, though compelled to transfer forces between the two fronts, they succeeded in holding their own; in 637 at Qadisiyah, near modern Baghdad, they won a victory that ensured the triumph of Islam in Persia; the significance of that victory remains so great in the Arab world that in the 1980s it was constantly evoked by Saddam Hussein during his war of attrition with Iran. Meanwhile other Arab armies were conquering Syria (636), Egypt (642) and pressing westward along the Mediterranean coast toward the Byzantine provinces in North Africa. In 674 Mu’awiya, the fifth caliph or ‘successor’ to Muhammad, decided to lay siege directly to Constantinople, and though the Arabs gave up the effort in 677, they returned in 717. By that year, they had taken the whole of North Africa (705), crossed to Spain (711) and reached the Pyrenees, over which they shortly invaded France. In the east they conquered Afghanistan, raided into north-
west India, annexed part of Anatolia (modern Turkey), pushed their northernmost boundary to the line of the Caucasus mountains and crossed the Oxus into Transoxiana where, at the Talas River in 751, they fought a decisive battle with the Chinese for dominance over the great caravan cities of Bokhara and Samarkand, on the Silk Road which led to the Great Wall.

  What made the Arab victories all the more astonishing was the relatively poor quality of their armies. The Arabs, despite centuries of desert feuding, had no real experience of intensive warfare; they were indeed ‘primitive warriors’ whose preferred form of operations was the raid (ghazwa).60 Nor does their generalship seem to have been particularly cunning. They certainly enjoyed no advantage in equipment or military technique. The Arab horse was already a fast, spirited and elegant beast, pampered and often hand-fed, in appearance almost a different animal from the shaggy steppe pony, but there were few of them. The camel, domesticated during the first millennium both in its single-humped (Arabian) and double-humped (Bactrian) species, was available in larger numbers but, though its powers of endurance were high, it was relatively slow and decidedly unwieldy.61 Strategically, the camel enabled Arab armies to cross terrain civilised armies thought impenetrable, and to appear on the field of battle quite unexpectedly; tactically, it was of limited use at close quarters. Arab tactics, therefore, were to make the approach march on camelback and to transfer to the led horses — there may have been as few as 600 at Qadisiyah — only at the moment of contact.62 These were the methods by which Khalid, one of the leading generals of the conquests, brought his army from Mesopotamia to deliver a crushing blow at the side of his comrade-in-arms, Amr, at the major victory of Ajnadain over the Byzantines in Palestine in July 634. On the battlefield itself, Arab armies chose positions defended by natural obstacles where their dismounted soldiers, armed with the composite bow, could fight behind some protection; they also preferred ground from which they had an easy escape route into the desert.63