Studies on Mounted Warfare in Asia IV: The Turanian Tradition – The Horse Archers of Inner Asia,c . ce 550–1350

I. Introduction

In a previous article two terms were introduced referring to two specific types of warfare, the Turanian tradition of the pastoral nomads on the steppes of Inner Asia and the Iranian tradition of the sedentary populations under their influence. ‘Iranian’ was not used in an ethnic or geographical sense, but to designate the military adaptations of sedentary civilizations to the direct influence of the pastoral nomads from Inner Asia, and ‘tradition’ was used to show that these adaptations remained essentially unchanged during the period under discussion. It was then argued that the Iranian tradition was characterized by a specific kind of warrior, the armoured horse archer. Two subsequent articles have given a description of the tactics, equipment, infantry, and servants of the armoured horse archers of the Middle East. In the present article the tactics of the armies of Inner Asia will be discussed, showing the similarities and the differences between the Turanian tradition and the Iranian tradition and how each one could transform into the other. ¹

The greatest military threat to the Middle East came over the road to Khorasan. In antiquity it had been invaded by Scythians and Parthians, and in the Parthian and Sasanian period eastern Iran had periodically been occupied by nomadic invaders, Sakas, Kushans, and Huns, but they chose to move on into South Asia instead of further penetrating the Middle East. After the Arabs had conquered Iran, a bitter struggle ensued for control of Transoxania. However, in the second half of the period under discussion, this ‘eastern front’ collapsed and nomadic peoples, first the Oghuz Turks and then the Mongols, again invaded the Middle East. ²

The fact that nomad warfare is the only alien type of warfare extensively discussed in the sources emphasizes how acutely the threat from the north-east was perceived. The topoi concerning the pastoral nomads of Inner Asia are often in an explicit context of a comparison or confrontation between Turan and Iran. The picture painted by the sources from the Middle East corresponds remarkably well with the descriptions of nomadic warfare coming from Europe and China. At the end of the period under discussion the Mongols were the subject of many of these descriptions, and during their reign the first extensive source appeared that was written from the perspective of the nomads themselves, the Mongol Secret History. We can now create a model of this Inner Asian warfare, the Turanian tradition, by comparing all these sources and keeping in mind what we have already established about its descendant, the Iranian tradition.

II. The Nomad Army

This article will be an attempt at a ‘Turanian tactica’, but it should first be emphasized that the pastoral nomads of Inner Asia mostly harassed their enemies with ‘small war’, ambushing adult males, rustling herds, and abducting women and children, rather than engaging in pitched battle. This mutual raiding seems to have been endemic among most nomadic tribes. It is often assumed the nomads developed their military tradition first of all to attack the sedentary civilizations. This bias has led to a distorted image of Turanian warfare, in which the nomad army is exclusively described as a weapon against the armies of an external enemy, for instance Persia or China, with a different military tradition. But while those armies could usually be evaded, there was no enemy more ineluctable to a nomadic tribe than another nomadic tribe. This article will therefore first discuss battle within the Turanian tradition, because it is only possible to understand nomad warfare by looking at it within its own context. ³

Most nomads were poor herders, subsisting by raising livestock, gathering, fishing, and hunting. Because of their hard nomadic lifestyle, they were tough warriors and unequalled as horsemen and archers, but they did not practise moving in formation or use armour, lance, and sword. They only had a knife to protect themselves in close combat, a weapon rather ill-suited to use on horseback, so practically speaking the bow was their only weapon. ⁴

Since the greatest threat to any nomadic tribe came from other nomadic tribes, next to these unarmoured horse archers the nomads also fielded armoured horse archers. ⁵ Such warriors were an excellent weapon against herders, since they matched their firepower and mobility while being almost impervious to their arrows. The armoured horse archer is so well adapted to nomad warfare that it is very likely that this type of warrior was first developed specifically for defence against a nomad army. ⁶ Although it may seem an obvious choice, this article will not differentiate between the armoured horse archers and the unarmoured horse archers with the designations of ‘heavy’ and ‘light’ cavalry respectively. Seen from a European - tactical - perspective, armoured horse archers fought as ‘light cavalry’, scouting, skirmishing, raiding, ambushing, and fighting with a missile weapon, the bow. It will also be shown that, seen from an Asian - strategic - perspective, the armoured horse archers of the nomads were yaridah, cavalry travelling light, without foot-servants or heavy baggage. So distinguishing the armoured horse archers of the nomad armies from the nomads without armour and close-combat weapons by describing them as heavy cavalry would be inaccurate both in a European and in an Asian context. In addition, the term “heavy cavalry” has such strong connotations of European origin that it would be more confusing than helpful in an Asian context. To avoid confusion the ordinary nomads will simply be referred to as herders and their well-equipped elite as armoured horse archers. ⁷

The horses of the herders were tough creatures. Though not as strong and as fast as the costly warhorses of the armoured horse archers, they could subsist on grass alone and were capable of travelling long distances and enduring difficult conditions. In the Middle East the horses were fed during a campaign and kept in their stables in winter. The nomads, however, pastured their horses on the steppes all year round, without additional feeding. As long as they had enough pasture for their horses, the herders did not need to carry supplies, send out foragers to cut grass, or establish depots. ⁸

Some scholars refer to the nomads’ horses as ‘ponies’. This modern category is used in various contexts with very different meanings, but the word is usually associated with primitive breeds – meaning only moderately shaped by human selective breeding – of rather small, hardy animals of low maintenance. This description is only partly appropriate for the horse of the Inner Asian nomad and offers little help in differentiating this animal from the warhorse of the armoured horse archers. First of all, compared with modern horse breeds, the Asian warhorse of the armoured horse archer was also a small animal, rarely exceeding a height of 1.45 m at the withers. More to the point, because horse-breeding first started on the Pontic-Caspian steppe, the various breeds of horse used by the pastoral nomads of Inner Asia were not primitive breeds at all. They were the oldest breeds of domesticated horse in existence, with the longest history of human artificial selection. Only their toughness and their ability to perform on a meagre diet of grass would justify the use of the modern category ‘pony’. Therefore, to differentiate the horse of the herder from the warhorse of the armoured horse archer, the former will be called the riding horse. ⁹

The nomads did not only use their riding horse for combat. In Inner Asia the use of the horse was much more varied than in the Middle East. As a food source its role was not as important as that of sheep and cattle, ¹⁰ but the horse was the only riding animal. Because of their considerable physical strength, the ox and the two-humped Bactrian camel were used as pack and draught animals, but they were too slow – and the camel too rare and costly – to be practical for military transportation. The dromedary camel could not survive on the cold steppes, while the domesticated ass was too rare in Inner Asia to offer an alternative. ¹¹ The nomads therefore used the riding horse both as a tactical and as a strategic mount. It was used for transportation as a riding, draught, and pack animal, for herding, and as a source of milk, blood, and meat. ¹² However, although the horse was unsurpassed as a military mount because of its speed, it was far less suitable for hard labour than our other domesticated mounts. Every nomad needed countless spare horses, because if he did not want to spoil his horse he could only ride it once every three days. ¹³ For their mobility the nomads relied on large herds of horses. Not all the nomads owned more than one horse, but their elites owned horses by their thousands, tens of thousands, and hundreds of thousands. ¹⁴

As soon as war threatened, the nomads sent their families, the heavy baggage, and the less mobile livestock away to inaccessible wastes, out of danger. An army of Inner Asian nomads consisted exclusively of horsemen. Unlike the huge and cumbersome Middle Eastern baggage-train of porters, walking rations, and pack animals – asses, mules, oxen, camels, and sometimes even elephants – the baggage of the nomads was all stowed on packhorses, making their baggage-train (Mongolian qoïtu’ul) very light and mobile. During a campaign the nomads fed on game and on the milk, blood, and meat of their horses, while their horses themselves were left to graze when not needed, so they did not need to carry supplies, only taking with them their tents and combat equipment. Therefore, a nomadic army pursued or fled as fast as the most heavily burdened horse could run. In mentioning the light equipment of the nomads, the sources are referring to their light baggage, not to an absence of armour and close-combat weapons. ¹⁵

III. The Turanian Battle Order

Just like a marching army in the Iranian tradition, a marching nomad army was ordered in five divisions, front (Mongolian manglaï), rear (chaqda’ul), left (je’unqar), right (bara’unqar), and centre (qol). ¹⁶ However, a Turanian battle order looked rather different from what was usual in the Iranian tradition. Even though the Strategikon claims the battle order in three lines had been copied from the nomads, ¹⁷ the nomads did not know the line formation. Their battle order was in fact composed of large hordes – Persian guroh, Arabic kurdūs (from Greek koortēs, from Latin cohort), Greek droungos (Latin drungus, or globus, ‘ball’), Mongolian na’ur (‘lake’) – deep, roughly circular, mob-like formations that comprised, in order of appearance, a unit of armoured horse archers arranged in ranks and files, a large, irregular crowd of herders, and a light baggage-train, including a large herd of spare horses. The sources often emphasize the contrast between the three divisions in lines of the Iranian battle array and the numerous hordes of the Turanian battle array. ¹⁸

The Strategikon contains a clue as to what the Turanian battle order would have looked like. The author advocates a new battle array, the squadron battle order, inspired by the example of the Avars and Turks, in which spacing was maintained between each of the cavalry squadrons of the first line, behind which the squadrons of the second line would form up, rather like the checkerboard formation of the ancient Roman maniple system. The advance guard became the most powerful line. The second line changed into a more modest reserve or support line (Persian pusht, Arabic muqāwamat, Greek boētos, Mongolian gejigä) and the third line was reduced to a right rear guard and a left rear guard only, without the heavy baggage in the middle. Instead, only a light baggage-train accompanied the soldiers, the families and possessions of the soldiers being left behind after the nomad example. ¹⁹ Some soldiers and pages were designated to form up in a line just behind the squadrons of the second line, to make the reserve look more impressive. In the event of a collapse of the first line, this thin line of men collected the fugitives. They probably dismounted to stop them, as on horseback their mounts would have been swept along by a rash retreat of the collapsed first line, but no actual infantry are mentioned in this battle array. They too were left behind with the baggage. With the encouragement of the men filling up the openings in the second line, the fugitives of the first line could again form up between the squadrons of the second line and then join them as they mounted a counter-attack. ²⁰

With this battle array, an army consisting of only a limited amount of units could still form up in three lines, while retaining the impression of a closed front. The amount of soldiers available to its commander was employed much more economically, since the first two lines would have been composed of the same amount of soldiers as a traditional single line of the same width. The new Roman battle order (in this example of 11 squadrons) is shown schematically in Figure 1. ²¹

OPEN IN VIEWER

Figure 1.

Squadron battle order.

The author of the Strategikon explicitly claims this squadron battle order had been copied from the Turks and Avars. This suggests that the nomads placed their units in a checkerboard battle array as well, with a powerful advance guard, a less powerful support and a still smaller rear guard. However, these units were not squadrons in line formation, ordered in ranks and files, but mobs of tribesmen assembled in irregular hordes. All these hordes are said to have added up to 12 or even 15 in number, depending on the size of their forces. Far in front of the battle array, in fact much further away from the main force than was considered prudent in the Iranian tradition, the skirmishers of the nomad army (Mongolian irä’ul) discharged their projectiles at the enemy. And finally we know that a light baggage train of spare and pack horses was formed up in the rear of each horde. The nomadic battle array could therefore be schematically represented as shown in Figure 2, with each circle symbolizing such a horde (the names for the lines are in Mongolian). ²²

OPEN IN VIEWER

Figure 2.

Turanian battle order.

IV. The Turanian Way of War

This section will reconstruct how the nomads actually fought. As we have discussed, the majority of the warriors in a nomad army did not own armour or close-combat weapons, so they preferred to pepper their enemies with arrows from a safe distance. ²³ Their style of skirmishing was markedly different from the combat style of the attackers in the Iranian tradition. While the elite of armoured horse archers at the front of the nomad horde formation stayed behind in closed order, ²⁴ the herders, in batches of three to five men, came from behind their elite, left the horde and galloped pell-mell towards the enemy, turned as they came within range while rapidly discharging a handful of arrows and then, without any regard to decorum, immediately fled back again towards the horde from which they had come, and instead of fighting in professional silence, they attacked screaming and howling (Mongolian sürä). ²⁵

John Masson Smith, Jr., ignoring the opinion of the sources that fighting from a distance was characteristic for the nomad, lets him approach his opponent up to a distance of about 40 m, in order to pierce his armour. ²⁶ This does not seem to be a very sensible course of action, since it allowed that opponent to turn the approaching herder into a pincushion well before this arrow could have been released. And it is not very likely that his arrow would have pierced armour, because, as discussed in a previous article, if a horseman wanted to be sure he would pierce his opponent’s armour, he either had to reduce the distance separating him from his opponent to less than 5 m (for the powerful nīkan) or dismount (for the accurate bārīkan). However, herders could not roam or stand about in the zone of death, as without armour they were extremely vulnerable to a hail of arrows, so they appeared only for an instant within range before disappearing again, forming a continuously renewing screen between the enemy and their battle array of hordes. They used their speed to reduce their vulnerability, not to hurl themselves against their opponents in a suicidal attack. Just like in the Iranian tradition, the continuity of the hail of arrows discharged into the zone of death was effected by continuous attacks of individual skirmishers. Because of their unequalled skill in archery and being unencumbered by armour, shield, or close-combat weapons, the herders discharged a heavy rain of arrows. ²⁷

However, according to Smith the nomads were unable to discharge such a hail of arrows. Even though the sources unanimously respect nomad archery, he claims the nomad herder was an inferior archer compared with the Mamluk armoured horse archer, rapid archery especially being a technique that could only be taught to these slaves. Even though the panjīkan, the rapid discharge of a handful of arrows, was already present in Tabari’s Sasanian sources, dating to the beginning of the period under discussion when no slave soldiers existed, Smith seems to believe the survival of Arab archery manuals from the Mamluk period, dating to the end of the period under discussion, must of necessity connect rapid archery exclusively with slave soldiers. And even though the sources make it quite clear that the nomads of Inner Asia, in fact all peoples known for their archery, learned their skill from a very young age, Smith sees this as characteristic for slave soldiers only. He claims that no nomad herder ‘would have access to any large-scale or intensive training program’. Perhaps not, but it would seem reasonable to assume that, through the daily use of his bow in hunting for his food, protecting his herds from predators, and the incessant ‘small war’ between nomadic tribes, the nomad became the accomplished archer described by the sources. ²⁸

In order to keep the enemy in the dark about their army’s actual size and location, and to allow their herders as much freedom of movement as possible, the nomads formed up in battle array at a considerable distance from their enemy, if possible concealed behind folds and ridges. The wide and deep territory for manoeuvre between the two armies was indispensable for a horseman without armour. Naturally, the long distance the herders had to cover while attacking and fleeing, added to the speed of their attack and subsequent flight, exhausted their riding horses, but as we have seen the nomads brought an abundant supply of spare horses. As he returned, the herder jumped straight from the horse he was riding onto one of the spares in the rear of the horde. ²⁹ If he had emptied his quiver, he could pick up new arrows carried on the packhorses. Since the nomadic battle array often did not show itself, the opponent simply being bled white with hails of arrows, some sources unfamiliar with the Turanian tradition claim that the nomads were just a noisy and disorderly mob. ³⁰

When encountering armoured horse archers, unsupported herders suffered severe casualties. Smith believes that, when facing herders, the armoured horse archer rapidly discharged his arrows from a standing horse. However, this procedure would have discarded the advantage of mobility, turning him into a large, stationary target instead, without much improving his aim (no hot-blooded oriental horse stands still in battle; it will stamp and paw the ground with excitement). Just like the herder, the armoured horse archer used the speed of his horse to increase his safety. To pick off individual opponents, he would either use the charging nīkan or the dismounted bārīkan, but rapid archery – the galloping panjīkan – was used to create a zone of death that would almost certainly wound any unprotected herder, while discouraging even a well-protected opponent from entering it. ³¹

A nomad chief would therefore select a unit of armoured horse archers to protect the herders. Unlike the herders, the armoured horse archers were supposed to remain close to the enemy and fight him continuously. When eventually weakened by fatigue, losses, or lack of success, they would be signalled to rally to the rear, to be relieved by the next unit. So instead of each unit of armoured horse archers in a nomad army dividing itself into attackers who assaulted the enemy in skirmishing order and defenders who remained behind in closed order, as in the Iranian tradition, an entire unit of armoured horse archers was selected to be sent to the front to skirmish. In the Turanian tradition these attack groups of armoured horse archers (Mongolian qoshi’un), supported by the skirmishing herders, took charge of all the tasks performed by the attackers, flanks, and ambushers of the Iranian tradition. ³²

Rather than the elite of his own tribe, a nomad chief would first select allied units and penal units of recruited prisoners of war for these attack groups, since their subservient status and ‘expendability’ were closer to that of the herders they would fight alongside with. ³³ It seems to have mattered little to the nomads if these allies and POWs were somehow related to their opponents. Unlike an ‘Iranian’ commander, who would try to isolate from the enemy any unit he did not trust, by placing it in a denied left wing or in the rearguard, or better still by sending it away to another front, it seems a ‘Turanian’ chief would without compunction throw any foreign units into the fray. However, just like an ‘Iranian’ commander would send a reconnaissance unit of his bodyguard to the front that was also designated to intercept any deserters, traitors, or spies, a nomad chief would send another attack group of his own elite retinue after the foreigners to supervise and, if need be, ‘encourage’ them. ³⁴

The attack group would split up into two sections, the one attacking, and the other staying in reserve in order to immediately plug any gap, exploit any success, or pounce on any pursuer fooled by a feigned retreat of the other section. With their continuous attacks, the armoured horse archers of the attack group, supported by the skirmishing herders, probed the battle array of the enemy for any weaknesses. ³⁵ Some tried to move around the flanks of the enemy in order to surround him and attack him from all sides. If the enemy became nervous because of this encirclement manoeuvre and fell back, the attack group would increase the pressure to encourage the enemy into full flight. ³⁶

As was shown in a previous article, in the Middle East the screen of attackers in front of the army consisted of a skirmishing line out of which the armoured horse archers would charge into the zone of death, shoot, and retire, while in front of them a handful of heroes duelled with their antagonists. Analogously, the screen of skirmishers in front of a nomad army also consisted of two parts. At a safe distance a ghostly screen of continuously appearing, shooting, and disappearing herders bombarded the enemy, while in front of them the duelling armoured horse archers of the attack group protected the herders from the armoured horse archers of the enemy. ³⁷ So in the Turanian tradition, the elite of armoured horse archers performed the same task as did the heroes in the Iranian tradition. The Mongolian word for ‘hero’ (bahadur) originally meant someone who had distinguished himself in close combat. ³⁸ Although bow, arrow, and quiver are mentioned in the very few Turanian sources, Turkish and Mongolian heroes are most often portrayed wielding lance and sword. The nomad heroes were armoured horse archers, set apart from the ordinary herders by their use of armour and close-combat weapons. Among the sedentary states, this nomad elite was highly esteemed and sought after because of their bravery and skill in close combat. ³⁹

In order to keep on fighting as an attack group, the armoured horse archers needed a constant supply of ammunition and spare horses, so the herders must also have supported their heroes as mounted servants, bringing them full quivers, fresh warhorses, water, and spare equipment, and mounting the wounded behind them on their riding horses to take them back. The page of the Iranian tradition and the herder of the Turanian tradition were equal to each other, since both were horsemen with the status of a warrior, but without armour or close-combat weapons. ⁴⁰

While the fighting of the armoured horse archers of the Middle East is called ‘attacking and retreating’ in the sources, the fighting of the nomads should more appropriately be called ‘attacking and fleeing’. In the Iranian tradition the armoured horse archer retreated calmly, avoiding the impression of a panicked flight, as that would only encourage the enemy and discourage his comrades. ⁴¹ The nomads, however, showed no such inhibitions, consciously encouraging their enemies to believe that they were fleeing in panic, in order to tempt them into a pursuit. ⁴² Feigned flight was part of the standard repertoire of the Turanian tradition. Opponents foolish enough to be lured into a reckless pursuit soon found themselves cut off and surrounded. ⁴³

It might occasion surprise or even disbelief that a stratagem that was continuously used could still be successful. The cause of its enduring success did not lie in the negligence of those in command, but in the psychology of the individual combatant. As soon as his enemies showed their backs, it was very difficult to restrain him from running after them, because if he would let them run away they would eventually turn around and attack again. Rather than waiting for their next attack in anxiety and frustration, a warrior would prefer to kill them when they seemed most vulnerable. Even if he knew he stood little chance of catching up with his enemies, he felt he obstructed their attacks as long as he ran after them. A pursuer will not easily give up enjoying this feeling of relative security, even if it is only a dangerous illusion. ⁴⁴

Both on a tactical and on a strategic scale, flight was a characteristic component of nomad warfare. Since the nomads were extremely skilful in extricating themselves from the fighting, a nomad chief could not be satisfied with the flight of his enemy. In the Iranian tradition a fleeing enemy was often left unmolested, since the capture of the opposing army’s camp and baggage was enough of a disaster to let them acknowledge their defeat. ⁴⁵ Once a nomad army fled, however, it remained intact to fight another day. As the entire army moved on horseback, it could quickly vanish, only leaving behind the worn-out horses that could no longer keep up. So although the nomads would leave their wavering enemy some means of escape in order to seduce him into flight, in the Turanian tradition this escape route was an illusion. As soon as they suspected their opponents to be on the brink of collapse, the nomads would take care to completely surround them, to prevent anyone from escaping. ⁴⁶ If this failed, fleeing enemies were pursued for days, the nomads not being satisfied until they had killed or captured all of them. ⁴⁷ Their countless spare horses allowed the nomads this fast pursuit, since after hours or even days of fighting, they would still have enough fresh horses left to them to pursue their enemy with lightning speed. They accepted the inherent dangers of such a fast pursuit – disorder and vulnerability to ambush and counter-attack – because it was the only way to decisively defeat a nomad army. ⁴⁸

Just like the Iranian tradition, the Turanian tradition preferred stratagem and ambush to the risks of pitched battle. ⁴⁹ For this reason the nomads surrounded their armies with scouts (Mongolian qara’ul) to timely detect the presence and intentions of the enemy. ⁵⁰ As in the Iranian tradition, these scouts were not specialized ‘light cavalry’ but belonged to the personal retinue of armoured horse archers of the nomad chief. ⁵¹

Many nomad stratagems were intended to make their army seem larger. By hiding his battle array and surrounding the enemy with skirmishers, a nomad chief could suggest a huge, omnipresent army, while in fact fielding an inferior force. If showing one’s host could not be avoided, the large herds of spare horses at the back of the hordes also served to make these formations appear more formidable. ⁵² Sometimes a few horsemen would tie branches to their horses’ tails that, when dragged along, would raise huge dust-clouds, suggesting the approach of large cavalry units. In one stratagem repeatedly ascribed to the Mongols, a chief would surround himself with puppets mounted on spare horses, in order to make it seem as if he had large reserves to draw on. ⁵³ They also used the stratagem of lighting many campfires at night to make their army seem larger (or elsewhere). ⁵⁴ To suggest additional military units, captured civilians, driven along in their armies to serve them, were forced to carry banners. ⁵⁵

If night fell while the nomad battle order was in close proximity to the enemy, the wings of their army would curve backwards until they met each other in the rear, so that the army would curl up in a defensive circular battle array (Persian parwaz, Arabic kurat, Greek kyklos, Mongolian küre’en). In this way they could meet an attack from any direction, sleeping in shifts in their boots and armour, with the reins in their hands. ⁵⁶ Usually, however, the nomads seem to have been rather careless about security when they rested. They would pitch their tents over a wide area to accommodate their horses, encircling the tent of their chief. As they lacked the throng of labourers from the civilian levy of the Middle East, they omitted to surround their camps with trenches and palisades. ⁵⁷ The nomads were more inclined to rely on their evasiveness than on fortifications and fled as soon as danger threatened, abandoning their meagre possessions. ⁵⁸ The families, herds, and heavy baggage of the nomads had usually been sent away to be hidden, far away from the enemy, but if that had not been possible or if the tribe was surprised before this precaution could have been taken, the carts and waggons were tied together in a circular laager (Mongolian kürän), while branches and stakes could be driven into the ground as obstacles. The nomads would then form up between the approaching enemy and their families. In the eventuality of a defeat, they could retreat into their laager, dismount, and discharge their arrows from behind these improvised field fortifications. ⁵⁹

V. Iran contra Turan

This section will discuss the confrontation between the Iranian tradition and the Turanian tradition. In the sources it is not the Persians, Romans, Arabs, or Indians who are depicted as the most formidable aliens. The nomads are feared above all other enemies, and waging war with them is covered fairly extensively. ⁶⁰

Avoiding war altogether seems to have been the preferred method to deal with the nomads. War against other sedentary states could provide rich spoils and territorial gain with associated tax-payers and mineral resources, but war against nomads possessed none of these attractions. Nomads were relatively poor and there was little profit in occupying the wilderness they inhabited. To reward the soldiers it was not enough to let them share in the spoils; the state had to foot the bill. A diplomatic approach consisting of flattery, presents, and promises was therefore the cheapest and safest option. ⁶¹

The unstable nature of nomadic chiefdoms could be exploited in a policy of divide and rule, by fostering rivalry and animosity among the nomads. The fact that they devised so many stratagems to make their armies appear more powerful than they actually were suggests the nomads had something to hide. On the steppes, population density was low. Even with their levée en masse, the nomads could only field armies of rather modest size, and few of those warriors possessed arms and armour. The best way to keep the nomads on the steppes was to impress upon them the huge resources at the disposal of the sedentary state. Since the Achaemenids the Persians had been wont to advertise the enormous size and wealth of their armed forces in order to overawe their enemies. ⁶²

When diplomacy, intrigue and intimidation failed and a powerful nomad chief adopted a menacing stance, the propaganda of superior might had to be turned into reality by mobilizing a large army and garrisoning it along the threatened border, in order to reduce the nomads’ freedom of movement and intercept booty-laden marauders returning to the steppes. This was an extremely difficult and costly logistical operation, because this large army had to be kept supplied with food and forage, while the area threatened by the nomads had often already suffered from their depredations to such an extent that it could no longer independently provide the surplus to feed all the soldiers and their animals. ⁶³

If a military confrontation with the nomads could no longer be avoided, it was essential to neutralize the two greatest advantages the nomads held over their sedentary opponents. These advantages were their superior mobility and their superior firepower, i.e. their abundance of horses and their unsurpassed mounted archery respectively. ⁶⁴

The mobility of the nomads could be sabotaged with prudent strategic choices. The horses of the nomads were pastured on the steppes all year round. During winter they had to dig up their fodder from beneath the snow with their hooves, using up a lot of energy to reach withered vegetation of reduced nutritional value, so by the end of winter the horses of the nomads were emaciated and quite useless. Therefore, late winter or early spring was the best time of year to fight the nomads. ⁶⁵ For instance, when the Chaghadaid Khan Baraq attacked Iran with his army of Central Asian Mongols in the late autumn of 1269, the Ilkhan Abaqa counter-attacked with his army of Iranian Mongols as soon as next year’s spring had arrived. By that time the horses of Baraq’s soldiers were still recuperating from the hardships of winter and unfit for riding (Mongolian qadäq). Baraq had even ordered his men to ride oxen or asses if necessary, while their horses were recuperating on the meadows of Northeastern Iran. Even though Abaqa was a Mongol ruler, his soldiers no longer relied on riding horses for their strategic mobility, but instead used asses, mules, and dromedary camels for strategic transportation. Their warhorses, instead of having to break the snow to reach the vegetation underneath during the winter, had stayed indoors while being fed on agricultural produce. Consequently, Abaqa kept the initiative during the entire campaign, Baraq having little choice but to slowly retreat to Herat and there to wait for events to unfold. ⁶⁶

If the nomads took the initiative to attack, they would mobilize in autumn, as after a summer of grazing their horses had been fattened and were in prime condition. ⁶⁷ In the event of such an invasion, fields and grassland along their expected routes had to be made useless for grazing by fire or poison, so the nomads would not be able to find enough pasture for their huge herds of horses. The harvest was either stored in well-fortified granaries or destroyed, to prevent the nomads from feeding their horses on locally captured grain. A lack of pasture was the most serious obstacle the nomads could face. As soon as they entered another ecological zone or could not find enough pasture because of a scorched earth strategy, they either had to limit themselves to a short raid before their horses died of starvation, or they had to leave the larger part of their spare horses behind, causing the remainder of their horses to become exhausted and their army to lose its superior mobility. ⁶⁸

There were also tactical precautions that had to be taken against the superior mobility of the nomads. As their abundance of horses allowed the nomads to quickly appear and disappear, their raids and stratagems were a matter of great concern. ⁶⁹ The judicious commander followed the nomads’ example by not only sending reconnaissance units ahead of the army, but doing so in all four directions. ⁷⁰ The sources emphasize the need for vigilance, letting the army march in a defensive formation and surrounding the camp in which the army would spend the night with trenches. One also had to take care the army was well supplied with water, rations, and forage, so the nomads could not starve the army by impeding the work of its foragers. ⁷¹

Since the nomads were in the habit of outflanking and surrounding their enemies, the commander had to provide for sufficient protection for the baggage-train by moving it closer to the centre during the march and giving it an escort of strong detachments. At the end of each day’s march and before the beginning of battle, the baggage had to be left in a fortified camp or another kind of safe position, because as soon as the soldiers believed their families and possessions to be threatened, they would desert to protect them. By forming up in front of a deep river, a steep mountain, or an impenetrable swamp, or by covering his flanks with such obstacles, the commander could prevent an encirclement. ⁷²

Most countermeasures against the nomads’ superior firepower were of a tactical nature. For instance, on hilly terrain, the commander always had to take care his army was in a higher position than the enemy before battle commenced, so the arrows discharged by the nomads would have inferior power and range. If that was not feasible, he should look for a flat plain so that his soldiers were at least not shot at from above. ⁷³

Another precaution against nomad archery was to fight in humid weather, because exposure to water lessens the effectiveness of the composite reflex bow. The sources mention the sudden rainstorms that for instance allowed the Arabs to defeat their opponents from eastern Iran and Central Asia at the battle of the Oxus River in 652 CE, or allowed Emperor Theophilos to escape annihilation by the Turkish horse archers of Caliph al-Maʾmun’s general al-Afshin after the battle of Dazimon in 836. The effects of rainstorms can be compared to those of dust storms. They interfere with the flight of the arrow and reduce visibility. Humidity will also influence the performance of the weapon itself. Bowstrings made of sinew are said to have become softer and more elastic when wet. Although a short rainstorm will not immediately damage a composite reflex bow, long exposure to humid conditions will weaken its reflex. Such a weakened bow needs to dry for a few days before returning to its former effectiveness. ⁷⁴

When the cavalry could not possibly match the firepower of the nomad archers, it was advisable to form up in a defensive battle order, infantry in front of cavalry, in a carefully chosen position. If the infantry was of insufficient quantity or quality it had to be reinforced with dismounted horsemen. The army had to be provided with enough food and water to prevent the nomads from starving the soldiers in their defensive position. When confronted with infantry formed up behind their large standing shields, the nomads quickly retired, since the herders, lacking armour, were very vulnerable to the far-reaching, powerful, and well-aimed arrows of archers on foot. ⁷⁵

This brings us to the advantages the sedentary Iranian tradition held over the nomadic Turanian tradition. Their shortage of arms and armour was a serious weakness of the nomads as it meant many among them were vulnerable at close range and virtually defenceless in close combat. To exploit this weakness the commander of an ‘Iranian’ army would try to get his battle array closer to the battle array of the ‘Turanian’ army. The cavalry was formed up without infantry, who were left behind with the baggage. By slowly but steadily moving his line formations towards the nomad horde formations, the distance between the two armies was shortened, which reduced the herders’ freedom of manoeuvre. The flanks needed to be reinforced, to prevent the nomads from surrounding the army. Such a tactic was preferably employed on a wide open plain with unimpeded view all round, so the commander knew the position of the enemy, and his army could not be surprised by an outflanking or encirclement manoeuvre by the nomads. ⁷⁶

As the herders immediately fled after discharging their arrows, and the nomads invariably employed the ruse of feigned flight, the author of the Strategikon warns the attackers not to stray too far from their defenders as the army carefully advanced. The soldiers were enjoined to fight with the bow and not to be seduced into prematurely pursuing the nomads. It was the slow and relentless advance of the defenders that was supposed to corner the nomads. The attackers were initially only used to screen their closed formations from the arrows of the nomads by skirmishing in front of them. Only after hemming in the nomad skirmishers between the advancing lines of the defenders and their own hordes could the attackers charge with a realistic hope of catching up with them. ⁷⁷

Wise and knowledgeable commanders cited by al-Tabari saw this circumspect, relentless, and orderly advance in line formation as the most noticeable difference between the Iranian tradition and the Turanian tradition, and according to the Strategikon, this was the tactic the nomads feared most. If an ‘Iranian’ army would keep its advancing formations in order without letting its flanks be surrounded, the herders would eventually run out of sufficient room for manoeuvre, because their dense hordes themselves would come within the range of a charge. Ultimately, there was no alternative left to them but to scatter and flee. ⁷⁸

This shows an important difference in quality between the Iranian and the Turanian battle array. The hordes of the nomads functioned as reservoirs from which the skirmishers would be sent forward to attack, but as a psychological weapon the Turanian horde formation could not compete with the Iranian line formation. The nomad hordes were huge, impressive-looking mobs, but deep formations suffer from the weakness that the men in the rear are unable to see clearly what is happening in front, which causes them to panic easily. An advance with a deep formation could only be successful if the enemy had been sufficiently weakened, so that he would already panic and flee as soon as the great mass of men and horses approached. If the enemy refused to be intimidated and those men at the front of the deep formation hesitated, those in its rear bumping into them sensed that something was wrong, which led to the formation’s rapid disintegration as the men panicked and fled. If the Iranian line formations succeeded in reaching the Turanian horde formations in good order, there was little else the nomads could do but run. ⁷⁹

However, a general flight usually was an effective means of extricating a nomad army from the fighting without much loss. Pitched battle between an army fighting in the Iranian tradition and an army fighting in the Turanian tradition therefore often led to a stalemate. Using the safest method, the defensive battle order, a victory only meant successfully blocking the path of the enemy. When using the offensive battle order, a victory meant that the enemy was put to flight but the vanquished rarely seemed much impressed by their defeat. ⁸⁰

It was precisely this attitude that was so frustrating for the nomads’ sedentary opponents. While in the Iranian tradition the party that had fled the field usually acknowledged its defeat by sending negotiators, the morale of an expelled nomad army did not suffer much because of its flight, as its forces had remained intact. ⁸¹ The nomads were not in the habit of taking their families, livestock, and heavy baggage along, so when they fled, neither a fortified camp, filled with military equipment and treasury, nor the soldier’s families and possessions were lost to the enemy. ⁸² Flight was an important part of the nomads’ strategy and tactics. They did not see flight as a defeat and they shamelessly simulated panic and confusion to delude their enemies into thinking they had defeated them. The Strategikon warns the commander not to become complacent after a ‘victory’ over nomad forces and to prepare his army for fighting again. ⁸³

If the conflict could only be resolved on the battlefield, an Iranian army needed to transform itself into a Turanian army. To achieve this, the families, foot-servants, and cumbersome baggage-train had to be left behind. Riding horses were procured by enlisting the aid of other nomads, so the army could match the strategic speed of the nomads by using the horse for the strategic transportation of its soldiers and a light baggage. Mounted on riding horses, armoured horse archers became yaridah, cavalry making forced marches, unencumbered by foot-servants and heavy baggage. The commander too had to leave his mule, elephant, or throne behind, to direct his army on horseback like a nomad chief. ⁸⁴ However, even when the commander and his soldiers were willing to sacrifice their comfort, obtained the cooperation and horses of other nomad tribes, and successfully adopted the Turanian tradition, the nomads could only be decisively defeated by stratagem. If a flying column succeeded in surreptitiously reaching the families and herds of the nomads, following their trail of cropped grass, manure, broken cartwheels, and extinguished campfires, a night attack could be extremely successful. With their families captured, the nomads could no longer use their favourite flight tactic, being forced either to fight or to surrender. ⁸⁵

VI. Turan in Iran

From an ‘Iranian’ point of view, the blood-curdling screaming and howling of the nomads might be considered a characteristically unprofessional trait. The nomads’ war cries initially distinguished them from the professional soldiers adhering to the Iranian tradition, in which noisy soldiers were seen as weak, undignified, and unprofessional. Admittedly, raw recruits have often betrayed themselves by the shouting with which they hoped to overcome their own fear and instil fear in their enemies, while veteran soldiers often went into battle silently, but battle cries are also a matter of habit. The Turkish armoured horse archers fighting the crusaders in the twelfth century, by then professional soldiers of a sedentary state, did not disown their descent and accompanied their attacks with characteristic howling and yelling, so the coming of the Oghuz Turks to the Middle East had changed the perception of how a professional soldier ought to behave. ⁸⁶

Such changes are often difficult to detect, since the Iranian tradition and the Turanian tradition, being so closely related, are in many respects very similar. A marching army in the Iranian tradition, soldiers followed by their families, servants, baggage, and walking rations, was an unmistakeable descendant of a wandering nomadic tribe, pastoralists with their families and herds. The most conspicuous new developments in the equipment of the Middle Eastern horse archer usually came from the steppes, from the ‘Scythian’ akinakes and pad saddle to the ‘Turkish’ sabre and wood treed saddle with stirrups. However, the composition of an Inner Asian army was very different from the composition of an army in the Middle East. While the armoured horse archers of the nomads were supported by another type of cavalry, the herders, the armoured horse archers of the Middle East were supported by men on foot. This transformation from the Turanian to the Iranian tradition was enforced by the migration from the biogeographic provinces of Inner Asia to those of the Middle East. ⁸⁷

On the other side of the Inner Asian steppes, the nomads migrating into northern China developed another variant of the Iranian tradition. ⁸⁸ The literature of yet another variant has already been used for the previous articles. The specific Iranian military tradition of the Middle East has been demonstrated with examples from Roman sources from the early Byzantine period, such as Procopius, Agathias, and the Strategikon, but many of the Roman soldiers came from the Balkan peninsula, not from the Middle East. As the Huns, Bulgars, and Avars settling on the Balkans subjugated local Romans, Germans, and Slavs, mixing with their elites, adopting their languages, and employing the rest of the indigenous warriors as their infantry, these former nomads developed another variant of the Iranian tradition that had kept some of its Turanian traits. It was this particular local variant of the Iranian tradition that is advertised for Roman use in the Strategikon. For instance, the Romans deviated from the Middle Eastern habit of feeding the horses during a campaign, instead pasturing them in the fields, just like the nomads. Roman professional footsoldiers were much more numerous than their Middle Eastern counterparts, and many of the Germanic and Slav soldiers in Roman or ex-nomad service were equipped with shield and spear instead of bow and arrow. ⁸⁹

Just like in the Middle East, the grasslands of the Balkan peninsula were not nearly extensive enough to feed the enormous herds of the nomads of Inner Asia. As a consequence, the Balkan cavalry was relatively modest in size, even the cavalry of the nomad tribes that had settled on the peninsula and had lost their steppe territory. Since neither the large herds of spare horses used on the steppes were at their disposal, nor the dromedary camels used in the Middle East, the Hun, Bulgar, Avar, and Roman cavalry were dependant on a few horses and mules for marching, which reduced their strategic mobility to that of men marching on foot. ⁹⁰

The settling nomads were also forced to follow the Roman example by using slow ox-carts for their baggage-train. Under nomad influence, a few pack horses had transplanted the mule-trains that had been used by the Romans in the previous centuries, but the most common beast of burden must have been the ass. Another, less common pack-animal was the powerful two-humped camel, bred both to the north and to the south of the Black Sea. Just like the nomads, the Romans would also use their carts and waggons to build a laager, as an alternative to the trenches common in the Middle East. In addition, they employed them to improve the battlefield. ⁹¹

Not only the Roman army of the early Byzantine period had adopted certain traits from the armies of the nomads; the Middle East too felt the Turanian influence. An example of such an influence seems to have been the battle order in separate squadrons (Arabic kurdūs, pl. karādīs, which also designated the horde formation) explicitly ascribed to the nomads in the Strategikon. In the third section of this article we discussed this particular kind of battle order, in which the units were not formed up in continuous lines but rather in a chequerboard array of separate squadrons. In the ninth and tenth centuries CE, deploying in squadron battle order seems to have become the standard practice in the Near East, the traditional Persian line formations having become an unintelligible phenomenon of the past. ⁹² Al-Tabari and al-Baladhuri use the battle array in lines as a purely moralizing topos, by taking its description silsilah – ‘chained’ – literally and suggesting the infidels had actually chained themselves together to compensate for the irresolution caused by their lack of the true faith. ⁹³ An account of one of al-Tabari’s sources contains an advice, seemingly cribbed from a military manual, to the avengers of Husayn, in which the squadron battle order is related to cavalry fighting against superior numbers. Such a situation requires the economical use of the available forces, the main advantage of the squadron battle order. The dismounting of a part of the reserve, in case the first line is driven back, is also mentioned. ⁹⁴

The introduction of the squadron battle order to the Middle East may perhaps be attributed to a Roman-Turanian influence from former Roman Syria that began to make itself felt after 644, when the Syrian Arabs became the pillars of the Caliphate in the Umayyad period. Even after its conquest by the Muslims, Syria was still affected by the influence of the Eastern Roman Empire through the influx of slaves and POWs. ⁹⁵ Al-Tabari (tenth century) ascribes the typical Turanian stratagem of the feigned flight to the Syrians, and he ascribes the battle order in squadrons to the last Umayyad caliph, which would confirm such a western origin. However, the historical veracity of such claims is highly questionable. ⁹⁶

The squadron battle order is in fact more likely to have been the result of a Khorasani-Turanian influence on the Middle East after the Abbasid revolution in 750, when warriors from Khorasan became the pillars of the Caliphate. Nomadic tribes had wandered into Khorasan long before the coming of the Arabs, exerting a strong military influence at the eastern front. ⁹⁷ In the ninth century one of al-Tabari’s sources mentions the ‘Tibetan’ armour of the Khurasaniyya, and al-Jahiz notes the curved scabbards containing their Turkish sabres. These soldiers came to Iraq with their own arms and armour, so it is very likely they also introduced some of their own tactics. ⁹⁸

It could also be that this battle order had always been part of the Iranian tradition, but became more common when armies diminished in size, forcing their commanders to economize. If this were true, the squadron battle order in the Strategikon could be a symptom of a decline in manpower in the Eastern Roman Empire in the sixth century that ultimately led to the loss of almost the entire Balkan peninsula to the Avars and the loss of Egypt and almost the entire Asian part of the Eastern Roman Empire to the Persians; the squadron battle order mentioned by al-Tabari could be a symptom of the fragmentation and bankruptcy of the Abbasid caliphate in the ninth century.

Since the horse archer is traditionally seen as an ethnically Turkish phenomenon, some might ascribe the Turanian influences on the caliphal armies first of all to their Turkish slave-soldiers. However, the Turkish slaves in the Middle East were not racially predisposed to fight according to the Turanian tradition; they were biogeographically constrained to fight according to the Iranian tradition, so it is not surprising that they were trained in this tradition at the Muslim courts. For example, in his description of the battle of Dandanqan in 1040, Gardizi explicitly contrasts the traditional Iranian deployment in continuous lines (s.aff-hā) of the army of the Ghaznavids, a dynasty of Turkish slave descent, with the Turanian deployment in separate hordes (kurdūs kurdūs) of the army of the Seljuks, a dynasty of free, nomadic Turkish descent. It was the victory of the latter that allowed the Oghuz Turkish nomads to migrate to the Middle East and introduce more obvious aspects of their military culture. ⁹⁹

But although such an invasion brought some changes to the Iranian tradition of the Middle East, the armies of the nomads were far more radically transformed by this geographical shift. Just as an ‘Iranian’ army needed to adapt itself to the Inner Asian circumstances to beat the nomads on their own turf, the nomads needed to adapt themselves as they left their own ecological zone. When the Oghuz and Mongols were no longer satisfied with raiding and moved into Iran en masse, they reserved pastures, built up depots, and repaired bridges. To serve them – and especially to help them besiege fortifications – they rounded up the civilian population. They also enlisted local professional soldiers, both cavalry and infantry. The wealth acquired by plunder and conquest allowed them to procure more arms and armour in order to increase the size of their units of armoured horse archers, formed up in ranks and files at the front of their formations, while reducing the size of the irregular mob of herders at the back of their formations to a small group of pages. They replaced their riding horses with local dromedary camels, mules, and asses. Their warhorses were fed during a campaign, instead of having to graze on the fields and meadows. In other words, the immigrants transformed their type of warfare from Turanian to Iranian. ¹⁰⁰

However, their style of fighting still betrayed some Turanian influences, for instance the above-mentioned screaming and howling. Turanian influence can even be detected in the literature itself. For example, in his military treatise, al-Aqsara’i (fourteenth century) lumps the traditional Iranian attackers, flanks, ambushers, and scouts together under one single designation, the Arabic kamīn, ‘ambushing party’. Since in the Turanian tradition their tasks had indeed been performed by one single tactical category, the attack group of heroes, this suggests that kamīn was in fact his rather awkward translation of a Turkish word – the equivalent of the Mongolian word qoshi’un mentioned in the fourth section of this article – introduced to the Middle East by the Oghuz Turks and still used by the Mameluke military. Even though it was less specific than the existing ‘Iranian’ terminology of the Middle East and derived from a different military tradition, it seems such a word was still in use with its original meaning. ¹⁰¹

Nevertheless, in literature the moral precepts of the Iranian tradition remained paramount. Ibn Khaldun (also fourteenth century) saw the popularity of the ‘Turanian’ custom of leaving the families of the soldiers safely behind in order to campaign as yaridah - flying column - as a moral decline. He believed it showed the soldiers were no longer prepared to fight to the death to defend their families and possessions. The nomads of Inner Asia would probably have disagreed with this ‘Iranian’ interpretation of their warfare, but one cannot help wondering what the Turcophone professional soldiers in the Middle East would have made of Ibn Khaldun’s verdict. ¹⁰²