The Imperial Roman High Command

The aspirations of soldiers who wished to enter into the militiae equestres highlight the often strange and convoluted path to advancement in the Roman army and administration. The usual pattern of promotion from the ranks of the army (via the primipilate and the Rome tribunates) bypassed the equestrian officer commands in the militiae and instead led to the procuratorial career. The opportunities for a former soldier to be placed in direct command of troops at a more senior level included the posts of praefectus classis, praesidial procurator, or the prefectures of the vigiles and praetorian guard. However, there are few indications that the Roman administration actively preferred former soldiers for these posts, and many a primipilaris is later found in financial procuratorships. The senior legionary and provincial commands were restricted to senators; experienced primipilares, as middle-aged men, were not normally suitable for entrance into the senate. This meant that there was no coherent career path from soldier to general in the principate. The promotion of former soldiers into the militiae equestres represented one challenge to this system, but it was not enough in and of itself to prompt the overhaul of the military career structure. This only happened gradually over the course of the late second and third centuries AD.

The emperors traditionally invested military authority in their senatorial legates, both the governors of consular and praetorian provinces, as well as any senators appointed to ad hoc supra-provincial commands, as in the case of Cn. Domitius Corbulo or C. Avidius Cassius. Important campaigns requiring significant forces, such as Trajan’s Dacian and Parthian Wars, saw the emperor and his senatorial generals assume primary command of the legions. Equestrian officers, usually in the militiae equestres, were placed in control of auxiliary troops or smaller detachments. For example, in the Parthian War of Lucius Verus, M. Valerius Lollianus, prefect of the ala II Flavia Agrippiana, was appointed praepositus of vexillations of auxiliary units in Syria. During this campaign Lollianus answered to the senior senatorial commanders: the governor of Cappadocia, M. Statius Priscus Licinius Italicus, and M. Claudius Fronto, who was legatus Augusti in charge of an expeditionary army of legions and auxiliaries. The majority of Marcus Aurelius’ senior commanders during his German wars, which occupied most of the 170s, were likewise senatorial generals. The praetorian prefects, who commanded the cohortes praetoriae and the imperial horse guard (equites singulares Augusti), were the exception to this roster of senatorial commanders. The praetorian prefect was occasionally entrusted with more senior authority, as when Domitian gave Cornelius Fuscus control over the conduct of his First Dacian War after the senatorial governor of Moesia, Oppius Sabinus, was killed in battle. Marcus Aurelius likewise invested his prefect Taruttienus Paternus with command of an expeditionary force at the beginning of his Second German War in AD 177. These shortterm appointments did not in and of themselves bring about a change in senatorial military authority.

There was a clear military hierarchy for senators: they could serve as military tribunes, then as legionary legates, then govern a two- or three legion province. There was no such well-defined path for equites, and no opportunity for talented equestrians to lead large expeditionary forces at a high rank. This meant that ad hoc solutions had to be devised, as happened in the 160s-170s AD during the reign of Marcus Aurelius. M. Valerius Maximianus, who began his career in the militiae equestres, was placed in charge of cavalry units sent to the eastern provinces to assist in suppressing the revolt of Avidius Cassius. Since he had advanced beyond the militiae, Maximianus’ higher standing was recognised by giving him the status of centenarius, the equivalent of a procurator. The same type of promotion was employed for his contemporary, L. Iulius Vehilius Gallus Iulianus, who had also advanced beyond the militia quarta. Iulianus was granted the exceptional title of `procurator Augusti and praepositus of vexillations’, as a way of recognising his seniority in several campaigns during this period. These commissions at procuratorial rank represented an attempt to create an equestrian equivalent to the senatorial legionary legate. The only alternative would have been to promote these equestrians into the senate at the rank of expraetor. This did eventually occur in the case of M. Valerius Maximianus and two of his Antonine contemporaries, P. Helvius Pertinax and M. Macrinius Avitus Catonius Vindex. But Iulianus remained an eques, eventually ascending to the praetorian prefecture under Commodus.

It must be emphasised that these promotions did not represent any attempt to advance hardened soldiers from the ranks to senior commands. Maximianus was from the curial class of Poetovio in Pannonia, while Vindex was the son of the praetorian prefect M. Macrinius Vindex. Pertinax was the son of a freedman, but had obtained equestrian rank and a commission in the militiae thanks to prominent senatorial patrons. The origins of Iulianus are unknown, but he certainly began his career in the militiae. There was only one seasoned solider on Marcus Aurelius’ staff: the praetorian prefect M. Bassaeus Rufus, who was from a poor and humble background, and had risen via the primipilate and a procuratorial career. The wars of Marcus Aurelius therefore introduced some important innovations, which highlighted notable problems with the developing equestrian cursus. The second century AD had witnessed the consolidation of the equestrian aristocracy of service, men who were prepared to serve the state domi militiaeque in the same manner as senators. Yet there was no clear way for these men to assume high military commands as equites, resulting in the creation of ad hoc procuratorial appointments.

The reign of Septimius Severus witnessed important developments for the Roman military establishment, and the place of the equestrian order within it. Severus created three new legions, the I, II and III Parthica, each of which was placed under the command of an equestrian praefectus legionis, not a senatorial legate. The first and third Parthian legions were stationed in the new province of Mesopotamia, which was entrusted to an equestrian prefect on the model of the province of Egypt. The commanders of the legions therefore had to be equites in order to avoid having a senator answer to an equestrian governor. This had been the practice of Augustus when he installed the legio XXII Deiotariana and the legio III Cyrenaica in Egypt under equestrian prefects. The same command structure was maintained in the legio II Traiana, which was the sole legion stationed in Egypt in the Severan age. The third new legion founded by Severus, the legio II Parthica, was quartered at Albanum just outside Rome, and thus became the first legion to be permanently stationed in Italy. One prefect of the II Parthica, T. Licinius Hierocles, is recorded with the exceptional title of praefectus vice legati (`the prefect acting in place of the legate’), though this was probably only a formality, since no senatorial legates are on record.

The career paths for the officers of the Parthian legions followed the pattern of the legions stationed in Egypt. Their tribunates were integrated into the militiae equestres, with some tribunes of the Parthian legions going on to procuratorial careers in the usual manner. The traditional route to the prefecture of the legio II Traiana in Egypt was via the primipilate and the Rome tribunates. The command of this legion ranked as a ducenarian procuratorship by the Antonine period, and the same status was given to the prefects of the new legiones Parthicae. The first prefect of a Parthian legion, C. Iulius Pacatianus, was promoted from the militiae equestres, but thereafter the commands appear to have been given to primipilares, following the Egyptian precedent. This suggests that Septimius Severus was following traditional status hierarchies when establishing his new Parthian legions. There was certainly no move to replace senatorial legates with equestrian prefects elsewhere in the empire. This had been attempted by Sex. Tigidius Perennis, Commodus’ praetorian prefect, after the British legions acclaimed the senatorial legionary legate Priscus as emperor. When Perennis tried to place equestrians in command of the legions, this punitive measure provoked a military revolt that eventually led to his downfall. Severus was not about to repeat this mistake, and therefore his new legions fitted with existing equestrian paradigms and career paths.

The foundation of the Parthian legions did, however, lead to changes in the expeditionary forces, particularly their overall command structure. The legio II Parthica was designed to accompany the emperor on campaign, a role it performed during Septimius Severus’ two Parthian wars and his British expedition. The question of whether the legion came under the direct command of the praefectus praetorio is a vexed one. In Cassius Dio’s Roman History the character Maecenas advises Octavian that the praetorian prefect should control all the forces stationed in Italy, a statement that could be taken refer to the situation in Dio’s own lifetime. As an official imperial comes during Severus’ Parthian campaigns, the prefect Fulvius Plautianus certainly joined the emperor in the east, but he is not mentioned in any specifically military capacity, in contrast with the abundant evidence for Severus’ senatorial generals leading troops in battle. It seems likely, therefore, that the authority of the praetorian prefect over the legio II Parthica evolved gradually. During Caracalla’s campaign against the Parthians his expeditionary force was composed of the legio II Parthica, the cohortes praetoriae, and the equites singulares Augusti, as well as vexillations of legions based on the German, Danubian and Syrian frontiers, totalling some 80-90,000 soldiers. This is what scholars call a `field army’, a modern term of convenience used to describe a large force composed of vexillations from a range of legions and auxiliary forces, which accompanied emperors or their leading generals on campaigns. Apart from the legio II Parthica, the only other legion that may have participated in Caracalla’s campaign as a complete unit was the legio II Adiutrix of Pannonia. This meant that the legio II Parthica was effectively the central core of the force and – although no ancient source explicitly attests this – the logical commander of the field army would be the praetorian prefect. Both of Caracalla’s prefects, M. Opellius Macrinus and M. Oclatinius Adventus, are known to have accompanied him to the east. This necessitated the appointment of a substitute prefect in Rome to handle the judicial responsibilities of the position (a problem we examined earlier in the chapter).

The legio II Parthica later formed the core of the forces marshalled by Severus Alexander and Gordian III for their eastern campaigns against the revived Persian empire. Indeed, it is during Gordian III’s reign that the connection between the legion and the praetorian prefect is shown clearly for the first time. Both the emperor’s praetorian prefects, C. Furius Sabinius Aquila Timesitheus and C. Iulius Priscus, formed part of the retinue that left Rome for the Persian front in AD 242. In the same year, Valerius Valens, praefectus vigilum, is attested in Rome `acting in place of the praetorian prefect’ (vice praef(ecti) praet(orio) agentis). In this capacity he oversaw the discharge of the veteran soldiers of the legio II Parthica. These men had originally enlisted in AD 216, and had been left behind in Rome rather than journeying to the east. The prefects on campaign with their emperor became enormously powerful individuals: C. Iulius Philippus, who succeeded Timesitheus, was able to arrange the downfall of Gordian III in the east, and returned to Rome as emperor. Successianus, an equestrian commander on the Black Sea in the 250s, was summoned by Valerian to serve as his praetorian prefect in the east, where he commanded the field army against the Persians. The composition of Valerian’s army is strikingly demonstrated by the account of the Roman forces in the account of the Persian king Shapur, known as the Res Gestae divi Saporis. This includes the detail that the praetorian prefect was captured by the Persians in AD 260 alongside the emperor and members of the senate. The employment of the legio II Parthica as a permanent core of the emperor’s own field army enhanced and consolidated the position of the praetorian prefect as a senior military commander in addition to the senatorial generals.

The rise of the field armies attached to the emperor and the praetorian prefect sometimes offered new opportunities to soldiers of other ranks. In the previous section we observed the marked correspondence between soldiers who served in the praetorian guard, the equites singulares, and the legio II Parthica, and those who obtained advancement into the militiae equestres or the promotion of their sons to equestrian rank. Proximity to the emperor and his senior staff on campaign evidently had its advantages. The same phenomenon can be observed in the careers of prefects of the legio II Parthica, which, since it accompanied Caracalla to the east, was intimately bound up with the political machinations of the years AD 217-18. In this period the empire passed from Caracalla to his prefect Macrinus and then to the boy emperor Elagabalus, with the crucial battles all happening in Syria. The commanders of the legio II Parthica included Aelius Triccianus, who had begun his career as a rank-and-file soldier in Pannonia and ostiarius (`door-keeper’) to the governor. Other ostiarii are attested as being promoted to centurion, so it is likely that Triccianus himself became a centurion and primus pilus, a career path attested for comparable equestrian legionary prefects. This was a spectacular career, but not unprecedented or improper. The same can be said for P. Valerius Comazon, who served as a soldier in Thrace early in his career, before rising to become praefectus of the legio II Parthica. Again, there is nothing truly exceptional in and of itself about soldiers who ascended to the Rome tribunates or camp prefecture via the primipilate. But the command of the legio II Parthica offered connections to the imperial court, and the favour of Macrinus and Elagabalus, respectively, enabled Aelius Triccianus and Valerius Comazon to enter the ranks of the senate. Their promotion earned the ire of the senatorial historian Cassius Dio, who disliked the progression of soldiers into the amplissimus ordo. Dio did not resent the advancement of equestrians per se, but the elevation of soldiers who were able to enter the equestrian order and then into the curia. Triccianus and Comazon were quite different from M. Valerius Maximianus, who originated from the curial classes of Pannonia. Such opportunities would only become more common as emperors spent more time on campaign with their field armies.

In addition to the creation of the Parthian legions and the growing importance of the field army, the first half of the third century AD witnessed equites appointed to ad hoc procuratorial military commands. We have already noted this phenomenon in the wars of Marcus Aurelius, when M. Valerius Maximianus and L. Iulius Vehilius Gallus Iulianus commanded army detachments with the rank of a procurator, as a way of compensating for the lack of any defined military pathway for equestrians after the militiae. In the reign of Severus Alexander, P. Sallustius Sempronius Victor was granted the ius gladii with a special commission to clear the sea of pirates, a command that was probably associated with his existing procuratorship in Bithynia and Pontus. This creation of new military commands within the procuratorial hierarchy can also be seen vividly in the case of Ae[l]ius Fir[mus]. Following a series of financial procuratorships in Pontus and Bithynia and Hispania Citerior (high-ranking posts in and of themselves), Fir[mus] was placed in charge of vexillations of the praetorian fleet, detachments of a legio I (possibly Parthica or Adiutrix), and another group of vexillations, in the Parthian War of Gordian III. In this capacity he ranked as an army commander and procurator at the ducenarian level, without actually holding a standing military post (such as fleet prefect, praesidial procurator or praetorian prefect). The adaptability of the equestrian careers to meet the new demands is demonstrated by the case of a certain Ulpius [-].227After series of administrative procuratorial positions, Ulpius was praepositus of the legio VII Gemina. Since this legion was normally stationed in northern Spain, Ulpius probably commanded vexillations of the legion in a war conducted in the reign of Philip. He then returned to the usual procuratorial cursus, serving as sub- praefectus annonae in Rome.

Some equestrians were given special appointments as dux with responsibility for a specific province or series of provinces. This can be observed in Egypt, where generals with the title of dux or σρατηλάτης [commander] appear in the 230s-240s. The archaic Greek word σρατηλάτης is rarely used in the imperial period before the third century AD; the only exception is inscribed account of the career of the Trajanic senator and general C. Iulius Quadratus Bassus at Ephesus. But it makes a reappearance in the third century AD to describe senior equestrian military commanders. The first Egyptian example is M. Aurelius Zeno Ianuarius, who replaced the prefect in some, or probably all, of his functions in AD 231. His military responsibilities should be connected with the beginning of Severus Alexander’s Persian War. The second dux/σρατηλάτης mis attested ten years later, in AD 241/2, which is precisely when war broke out between Romans and Persians again under Gordian III. This time, the dux was Cn. Domitius Philippus, the praefectus vigilum, who appears to have been sent directly to Egypt while retaining his post as commander of the vigiles. In both cases the new military command was an ad hoc addition to their usual equestrian cursus. The final example occurs in the 250s, when M. Cornelius Octavianus, vir perfectissimus, is attested as `general across Africa, Numidia and Mauretania’ (duci per Africam  Numidiam Mauretaniamque), with a commission to campaign against the Bavares. This substantial command was in succession to his appointment as governor of Mauretania Caesariensis. Octavianus then departed to become prefect of the fleet at Misenum, working his way to a senior post in the equestrian procuratorial cursus. All these cases show the essential adaptability of the imperial system, which allowed third-century emperors to appoint equestrians to senior military commands when it suited them. This may have been because an equestrian was the person the emperor trusted most in the circumstances; for example, Cn. Domitius Philippus, as praefectus vigilum, was one of the most senior officials in the empire. This represents the same pragmatic approach we saw in the appointment of equestrians as acting governors. On a practical level, it did not matter whether an army commander was an eques Romanus or a senator, because the military tasks that he was capable of performing, and was entrusted with by the emperor, were essentially the same. The new ad hoc army commands gave members of the equestris nobilitas further opportunities to serve the state domi militiaeque alongside the senatorial service elite.

At the same time, it is necessary to point out that these changes did not lead to senators being ousted from military commands prior to the reign of Gallienus. Rich epigraphic evidence, combined with the testimony of Dio and Herodian, preserves a long list of Septimius Severus’ senatorial generals. P. Cornelius Anullinus, L. Fabius Cilo, L. Marius Maximus, Ti. Claudius Candidus and L. Virius Lupus commanded Severus’ troops as duces or praepositi in one, or both, of his civil wars against Pescennius Niger and Clodius Albinus. Candidus also participated in the emperor’s Parthian campaigns, alongside Ti. Claudius Claudianus, T. Sextius Lateranus, Claudius Gallus, Iulius Laetus and a certain Probus. These senators were rewarded with a range of honours, from consulships and governorships to wealth and property (the sole exception was Laetus, who was executed for being too popular with the troops). In the face of such overwhelming testimony, it proves difficult to marshal support for the still-popular scholarly argument that Severus prioritised equestrian officers over senators. Equestrian commanders continued to participate in campaigns as subordinates to the senatorial generals, as we see in the case of L. Valerius Valerianus, who commanded the cavalry at the Battle of Issus under the authority of the consular legate, P. Cornelius Anullinus.

The same pattern can be found in Severus Alexander’s Persian War of AD 231-3. Herodian’s History, our major historical account of this conflict, is notoriously deficient in prosopographical detail. Yet senators are attested in inscriptions, as in the case of the senior consular comes, T. Clodius Aurelius Saturninus, who accompanied Alexander to the east. The senator L. Rutilius Pudens Crispinus, praetorian governor of Syria Phoenice and legate of the legio III Gallica, also served as a commander of vexillations during this conflict. But we only know about Crispinus’ command from an inscription from Palmyra, which recounts the assistance rendered by the local dignitary Iulius Aurelius Zenobius to Alexander, Crispinus and the Roman forces. The inscribed account of Crispinus’ career from Rome merely states that he was legatus Augusti pro praetore of Syria Phoenice. It is probable that senatorial governors, such as D. Simonius Proculus Iulianus, consular legate of Syria Coele, continued to play important roles in eastern conflicts under Gordian III. Indeed, the evidence for equestrian procurators acting vice praesidis in Syria Coele, discussed above, suggests that the procurator assumed judicial responsibilities while the consular governor was preoccupied with warfare. This indicates that senatorial governors continued to play a major part in military campaigns, even if it was not specifically noted in inscriptions recording their cursus.

This argument is supported by the literary sources that show senators assuming military commands through to the middle decades of the third century AD. We can observe this in particular in the Danubian and Balkan region, which was a near-continuous conflict zone. Tullius Menophilus fought against the Goths as legatus Augusti pro praetore of Moesia Inferior in the reign of Gordian III. During the incursion of the Goths under Cniva in AD 250/1, the Moesian governor C. Vibius Trebonianus Gallus successfully defended the town of Nova. In AD 253 M. Aemilius Aemilianus, governor of one of the Moesian provinces, pursued the fight against the Goths, before being acclaimed emperor. Senators also continued to receive special commands, as in the case of C. Messius Quintus Decius Valerinus and P. Licinius Valerianus, both future emperors, who were placed in charge of expeditionary forces by the emperors Philip and Aemilius Aemilianus, respectively. In Numidia, the governor C. Macrinius Decianus conducted a major campaign against several barbarian tribes in the middle of the 250s. In fact, if we examine the backgrounds of the generals who claimed the purple up to and including the reign of Gallienus, the majority of them were actually senators, a fact obscured by the common use of the term `soldier emperor’ for rulers of this period. Decius, one of the few known senators from Pannonia, successfully allied himself with an Etruscan senatorial family when he married the eminently suitable Herennia Cupressenia Etruscilla. His successor, Trebonianus Gallus, was of remarkably similar background to Etruscilla, coming from Perusia in central Italy. The emperor Valerian likewise had close links with the Italian senatorial aristocracy, marrying into the family of the Egnatii. Some of the more ephemeral emperors deserve notice too, such as Ti. Claudius Marinus Pacatianus, the descendant of a Severan senatorial governor, who rebelled in the reign of Philip. P. Cassius Regalianus, who was probably consular legate of Pannonia Superior when he began an insurrection against Gallienus in 260, was himself descended from a Severan suffect consul. These men were not soldiers promoted from the ranks, but senatorial generals who used their positions to make a play for the imperial purple.

The Roman military hierarchy in the first half of the third century AD was therefore characterised by a mixture of continuity and change. The creation of the legio II Parthica, and the necessity for the emperor and his praetorian prefects to campaign on a regular basis, meant that emperor was in close contact with members of the expeditionary forces. Officers in the field army could receive imperial favour and embark on spectacular careers, like Aelius Triccianus or Valerius Comazon, or even Iulius Philippus, the praetorian prefect who snatched the purple from Gordian III while in the east. It is no coincidence that many of the soldiers’ sons attested with equestrian rank belonged to the praetorian guard, the equites singulares and the legio II Parthica. At the same time, the imperial state tried to create senior army roles for promising equites in a manner analogous to senatorial legates by instituting ad hoc procuratorial commands (as seen in the case of Valerius Maximianus and Vehilius Gallus Iulianus). This gave members of the equestris nobilitas, the equestrian aristocracy of service, access to army officer commands beyond the militiae equestres. It should be noted that for the most part these men were not lowborn ingénues from the ranks, but members of the municipal aristocracy who served the res publica in a comparable manner to senators, as their predecessors had before them. It is also imperative to point out the endurance of tradition within the high command. Senatorial legates and generals still commanded armies in the emperor’s foreign wars on the Rhine, Danube and Euphrates frontiers. Their military authority continued to make them viable and desirable candidates for the purple in the first half of the third century AD. There was as yet no attempt to undermine the positions of senatorial tribunes or legionary legates. It was the dramatic developments in the 250s-260s that provided the catalyst to set the empire on a radically different path.



A reconstruction of the battle of Pydna, by Peter Connolly, demonstrating how the broken ground disrupted the Macedonian phalanx, allowing the Romans to close with the phalangites and use their superior swordsmanship to good effect.


Troop movements before the battle.


Dispositions prior to the battle.


We do not know why the battle of Pydna finally began. In one tradition Aemilius Paullus sat in his tent waiting for the sun to decline into the west, so that it would shine in the eyes of the Macedonians. When the time was right he brought on the battle with the ruse of having the Romans pursue a runaway horse toward the Macedonian lines. In another version, Paullus did not wish to fight at all that day because he was waiting for his foragers to return, and the battle began accidentally as the result of a skirmish between drawers of water. But this we know: the Roman army deployed in the conventional manipular formation and fought Perseus’s phalanx nose to nose on the plain. And, predictably, the Romans began to lose. When the Romans met the Macedonian phalanx the spears of the Macedonians pushed them steadily back: “At the onset, Aemilius arrived and discovered that the Macedonian units had already planted the tips of their sarissas in the shields of the Romans, who could therefore not get forward and reach them with their swords. And when he saw the rest of the Macedonians drawing their shields around front of their shoulders [in preparation for combat], and that the sarissas leveled at a single signal were withstanding his shield-armed soldiers, and when he saw the strength of the locked-shield formation and the harshness of the attack, astonishment and terror seized him, because he had never seen a sight more fearful. And later he often used to recall his emotion and what he saw.” The Romans were driven to extremities to break the Macedonian formation and stop the relentless pressure. The commander of a contingent of Rome’s Pelignian allies cast his soldiers’ standard into the midst of the phalanx, and the Pelignians tried heroically to thrust the Macedonian spears aside—with swords, shields, even their bare hands —to recover it. But to no avail. The Romans were soon in full retreat. Aemilius Paullus rent his garments in despair.

Why Aemilius Paullus was willing to fight a frontal battle with Perseus’s phalanx is a mystery. Plutarch attributes to Paullus awareness of the special dangers presented by fighting the Macedonian phalanx: and well might he have been. The Romans had been fighting the Macedonian phalanx for more than a century. Pyrrhus defeated the Romans with it in the early third century, the Carthaginians in Africa in the middle of the century did as well, and Hannibal did the same later. In 197 bc the Romans had won a terrifying victory against Perseus’s father at Cynoscephalae, a battle that vividly illustrated the terrible power of the phalanx’s charge, even on unsuitable ground. In the year before Cynoscephalae, the Roman siege of Atrax had failed when a Macedonian phalanx drawn up in a breach in the wall had proved quite impervious to Roman attack. Polybius’s judgment that “when the phalanx has its characteristic virtue and strength nothing can sustain its frontal attack or withstand the charge” will have been no news to Roman commanders. The phalanx’s fatal flaw, Polybius says, is that it requires flat terrain so that it can preserve its close order. Perseus’s father’s unwise decision to fight on broken ground allowed the Romans to defeat him at Cynoscephalae. But Aemilius Paullus consented to fight the Macedonian phalanx on a plain, ideally suited to it, on ground that Perseus had chosen for exactly that reason.

There was nothing in Aemilius Paullus’s record to suggest he was rash or inept: quite the opposite. When campaigning in Spain, he had chosen his battlefields with a careful eye toward advantage. Later he had defeated the Ligurians with a complicated stratagem, issuing forth from a besieged camp from all gates at once to catch the enemy by surprise, here carefully having the hastati of one legion lead the charge, there the principes of another. After his victory over Perseus men remarked on the careful preparations Paullus made for throwing dinners in Greece, and the Roman general joked that he took as much care over ordering dinners as he did in marshaling a line of battle: a remark that loses its force unless he was thought an especially careful tactician. “A really good general,” Paullus was apt to tell his own son, “does not commit to a pitched battle at all, except in cases of the greatest necessity or the greatest advantage.” Neither was the case at Pydna.

The outlook, past conduct, and behavior of Aemilius Paullus in the days before Pydna—his feints and ruses on the Elpeus—place him squarely in the Roman tradition of crafty, strategy- and tactics-minded generals, in the tradition of Fabius Maximus and the great Scipio Africanus. Rome was old in the use of tactics: there is no reason to suppose there was ever a generation of Roman fighting without tactics, any more than there was in Greece. Guile in war was as ancient as Roman legend: in the fabled duel of the brothers Horatii, the surviving brother overcame the Curatii by the trick of false flight. At the very first Roman land battle of which there is a reliable, detailed account, Bagradas River in 255 bc in the First Punic War, the Romans are already found varying the standard manipular deployment in the hope of gaining a tactical advantage. By the Second Punic War and the early second century, commanders like Scipio Africanus managed to adapt the many moving parts of the manipular legion to sophisticated tactical battle plans: varying the checkerboard disposition of the maniples, deepening the array, wheeling the array, using the lines of hastati, principes, and triarii independently, detaching maniples to flank the enemy, deploying a second line of legions as a reserve, and approaching the enemy in a battle-ready marching formation. Other sections too of the Roman army were creatively exploited by tactical Roman commanders. Cavalry made flank attacks, and the velites were detached from their legions and used as independent forces of light infantry. Such tactics played a crucial role in defeating Carthage and featured largely in the great victories of Scipio Africanus at Ilipa in Spain in 206 bc and Zama in North Africa in 202. Cato’s battle of Emporiae in Spain in 195 may display the most complicated Roman tactics of all in this period.

Although tactics were native at Rome, by the late third and second century bc there is clear evidence of Greek influence on Roman generalship and methods of fighting. Scipio, we are told, was a keen reader of Xenophon’s Cyropaedia, a work in which the Greek expert offered in Persian guise his proposals for Greek military reform. By Paullus’s day Roman cavalry equipment had been reformed on the Greek model. The Romans had also begun actively seeking out specialist Greek troops, like the archers from Crete, rather than merely using them when sent by their allies. Elephants too were part of an integrated Hellenistic army. A force of them had been brought from Africa at the beginning of the war against Perseus, and Aemilius used them at Pydna. Thirty years before, Flamininus had used elephants to fight Philip V at Cynoscephalae. The Romans may have learned more about elephants from fighting Carthaginians than from reading Greek manuals, but this was Greek doctrine at second hand: the army of Carthage was modeled on Greek armies, and Hannibal himself was a Hellenized commander. Now the Romans were using Greek stratagems; now the Romans knew Greek lore like the path around the back of the pass at Thermopylae. Now Romans could be imagined debating questions like, Who is the greatest general of all time?—questions that implied that some of them at least had come to share the Greek conception of generalship as a competition in technical expertise.

Greek influence on Roman command is hardly surprising: it is merely a small instance of the gigantic phenomenon of the Hellenization of the Roman aristocracy in the late third and second centuries bc. This cultural transformation affected nearly every aspect of upper-class Roman existence —from education and language to entertainment, decoration, clothing, and housewares. And many of Rome’s tactically minded commanders were leaders of the philhellenic movement, Scipio Africanus (Hannibal’s foe) and Aemilius Paullus among them. After his victory over Perseus, the Greek-speaking Paullus took a long tour of famous sites in Greece and offered Greek athletic games. He gave the Macedonian royal library to his sons. When Paullus made his grim way home after being appointed to fight the Macedonian king, he found his small daughter in tears. Her little dog Perseus had died. “Good fortune, my daughter! I accept the omen!” crowed the consul (how his weeping daughter greeted his glee is not recorded). But a dog named for a Greek hero is also a small yapping sign of how pervasive Greek influence on Rome had become. When we see Paullus not only employing military tricks and ruses, but also applying natural science to war—geology to dig wells, astronomy to reassure his men—we can be certain that he was up to date with the most advanced trends in Greek generalship.

Yet cerebral generalship was strongly resisted at Rome. Trickery and tactics, as the old senators had complained when the envoys reported the lies they had told to Perseus, could be viewed as opposed to virtus. “Among the Romans, a bit of a trace of the old philosophy of war is left,” wrote Polybius of the Romans of Aemilius Paullus’s day. “They declare war openly, rarely use ambushes, and fight their battles hand-to-hand at close quarters.” Many Romans preferred a battle, as Livy put it, “with standards set against standards, on a clean and open field, where without fear of ambush the affair could be settled by true virtus.” The general, in this view, was to lead the army directly at the enemy, to allow his soldiers to display their virtus, and to display his own. After a loss a Roman general might be prosecuted for personal cowardice, but not for tactical stupidity. Terentius Varro, the aggressive consul who committed the Roman army to the disaster at Cannae, was not punished for his bad planning but thanked when he returned to Rome for “not despairing of the Republic.” It was far more important to the Romans that their generals be plucky and adventuresome than that they be skillful strategists and tacticians.


The Persian [Achaemenid] Army

Persian military forces were drawn from all areas of the Empire, members of the elite corps as well as conscripts levied for local action or for major campaigns. Thus the label “Persian” is not to be understood as describing the ethnic makeup but rather the troops’ allegiance, fighting under Persian officials or commanders. As has been seen, however, the command structure was not thoroughly Persian by any means either, save at the very top of the hierarchy, including most satraps and of course the King himself. The Old Persian word kra may be translated either as “army” or as “people.” This reveals the army’s ultimate origin – among the Persians themselves, many of whom came to form the corps of the standing army – as it results in occasional confusion in modern translation. When kra appears in a text, it is not always evident to us whether the people as a collective group or the specific subset of the army is meant.

Herodotus gives a full and colorful account of the vast and diverse forces of the imperial levy, the full army and navy of Persians and subject peoples, when he tallies the vast forces that Xerxes arrayed against Greece in 480 BCE. Herodotus also names many of the commanders, an elaborate depiction of the peoples of the Empire with descriptions of their clothing and equipment (7.61–100). For example, both Persians and Medes were arrayed in felt caps, colored tunics over scale mail, trousers, wickerwork shields, and a variety of weapons. Ethiopians (Nubians) wore leopard or lion skins and carried large bows. Paphlagonians wore woven helmets and carried small shields and spears. That Herodotus’ entire portrayal better describes a parade than a battle array has long been understood. But it typifies the diversity of peoples and weaponry that the Persian commanders had to weld into an effective fighting force. Persian forces, both infantry and cavalry, were renowned for their use of the bow: a frequent tactic was the unleashing of storms of arrows from behind a shield wall or for horsemen to harry the enemy with volleys of arrows.

Scholars debate the effectiveness of the Persian forces’ armor and tactics especially in the context of Xerxes’ invasion of Greece in 480 and Alexander’s invasion of the Persian Empire in the late 330s. Herodotus (9.62) describes the final crush of the Persians against the Spartans at the Battle of Plataea in 479:

On the one hand the Persians were no less than the Greeks in courage

and strength, but the Persians were without shields and, beyond this,

were unskilled and not the equal of their opponents in experience.

This passage offers just one example of the persistent problems of source evaluation. When Herodotus says that the Persians were “without shields” (Greek anoploi), what does that mean? Were the shields lost in battle? Was this contingent of the Persian army simply not carrying shields? And which group was it, ethnic Persians or some other? Some translate anoploi as “without armor,” which adds another layer to the problem. The Spartans were the most (by far) professionalized Greek soldiers of their day, so even the elite corps of the Persian army would have had their hands full against them. Beyond the elite, levied troops from the provinces of course did not have the same sort of armor, weaponry, or tactics as did, for example, the Persian Immortals and similar contingents. Numerous other passages in Greek sources provide similar perspectives: heavily armed Greek infantry, fighting in tight phalanx formations, trumped the (as generally described in Greek sources) light armed, less experienced, inferior Persian infantry every time – except when they did not. It is difficult to sift the Greek stereotypes from the realities of individual battles. That the Persians were able to conquer and retain so much territory for so long testifies to their army’s effectiveness.

The elite Persian force, numbering 10,000 according to Herodotus (7.83), was called the Immortals. Whenever one of their number died or was wounded or ill, another would take his place so the number of the battalion always remained 10,000. They were the most effective, and feared, Persian infantry force, and clearly comprised elite members of Persian society: men of prominent families or high rank. One thousand of them had gold pomegranates on their spears, some of whom comprised the king’s personal bodyguard, and the other 9,000 had silver. Herodotus’ incidental detail that the Immortals were conspicuous for their gold (bracelets or other marks of status and honor), and that they were accompanied by wagons bearing concubines and many servants, indicates that we are not dealing with the rank and file. Prestige items are frequently mentioned in conjunction with Persian officers and nobles, a phenomenon that also fed Greek stereotypes of Persian effeminacy and weakness. But these items were more symbolic than practical and communicated entirely different messages – honor and status – in a Persian context.

Greek sources often highlight the prominence and skill of Greek mercenaries, and from that perspective it was only thanks to better trained and better equipped Greek professionals that the Empire was able to field any sort of worthwhile fighting force in the fourth century. This trope contributed heavily to the stereotype of the effeminate, decaying Persian Empire before its fall to Alexander the Great. And even though Greek mercenary forces were an increasing phenomenon in the fourth century, and certainly used by Persian commanders, their significance often seems overestimated in Greek sources.

Persian Arms and Equipment

Herodotus describes the arms and equipment of Xerxes’ army in some detail. The Persians themselves wore floppy felt hats, tunics and armour exhibiting a surface of fishlike iron scales, and trousers. They carried wicker shields. Their weapons were large bows, short spears and daggers which were suspended from the belts on the right-hand side. Thus equipped, they might or might not be mounted. Persian armies generally relied upon the large numbers of their horsemen and bowmen.

Apart from the Persians themselves, Herodotus gives particulars of the other national contingents which the Persian kings were able to mobilize, although the statistics on which he based his information may have referred to the potential fighting strength of the entire Persian Empire rather than to Xerxes’ expeditionary force, gigantic though this force unquestionably was. We hear of Assyrians and others with bronze helmets; but in general, the Asiatics were protected only by various kinds of soft headgear and they seem to have worn no substantial body armour. Apart from daggers, bows and arrows, their weapons included iron-spiked clubs, axes and lassoes.

Cavalrymen – especially cavalry officers – may have worn more protective armour. Masistius, the Persian cavalry commander who was killed in the early stages of the Plataea campaign, wore gold scale armour under his scarlet surcoat. When his horse was hit by an arrow, he defended himself vigorously on foot and could not be brought down by body blows. At last, the Athenians who surrounded him guessed the secret and struck at his face.

Persian archers, both mounted and unmounted, carried their arrows in a quiver slung on the hip. This practice differed from that of the Greek archers whose quivers were slung on their backs. The hip position was no doubt more expeditious when there was a requirement for rapid fire.

Herodotus refers to the war chariots of the Indian contingent, but there is no mention of these chariots being used in the fighting. Persian kings normally went to war in chariots, which were also employed by the Persians for hunting. The Greeks of the classical period used chariots only for sporting events. Generally speaking, by the time of the Persian Wars the war chariot had been replaced by the man on horseback. The change had no doubt been brought about by the improved efficiency of horses’ bits, which made it easier for the rider to control his steed.

The Persian High Command

The Persian numbers in the two invasions were so overwhelmingly superior that one tends to blame the Persian commanders for the startling lack of success. The initiative for both enterprises came from the Great Kings themselves and there seems to have been no question of any significant “power behind the throne”. Yet there is nothing particularly blame-worthy in their conduct of the two operations – apart from the undertaking itself. There comes a time in the history of every empire when expansion has gone far enough and stability and consolidation, if not retrenchment, are needed. The handful of Athenian and Eretrian ships that had abetted the Ionian revolt was a poor pretext for such a massive military and naval effort.

If we turn to Aeschylus’ play, we find some contrast between the characters of Darius and Xerxes. The Persae presents the story of Xerxes’ crest-fallen return to Persia after his defeat at Salamis. Darius’ ghost appears and denounces the folly which has led to the recent débâcle. Darius is stern and dignified; in contrast, Xerxes is petulant and ineffective. At first sight, Herodotus’ narrative might seem to confirm this estimate. One recalls the incident when high winds destroyed the first bridge which Xerxes had constructed over the Hellespont, whereupon Xerxes ordered that the rebellious waters should be whipped as a punishment for the outrage. But perhaps this was not mere childishness on his part. In his multinational host there were many simple tribesmen who knew nothing of the enlightened Zoroastrian religion of the Persians; thus, to restore morale, it was no doubt necessary to demonstrate that even the gods of the winds and the waves were subject to the Great Kings of Persia.

Again, we are inclined to regard Xerxes’ return to Susa, his remote capital, after the disaster of Salamis, as weak and cowardly. Mardonius, his general, seems to have been left callously to his fate in Greece. But the matter may be viewed quite differently. The success of the Persian kings lay very largely in their ability to delegate power. Cyrus, when he conquered Lydia, had delegated the completion of his conquest to his general Harpagus, and probably Mardonius was expected to complete the conquest of Greece in the same way. However, when all has been said, the delineation of character in Aeschylus’ play should not be lightly dismissed. Aeschylus was, after all, writing at a time very close to the events which he described and he cannot altogether have overlooked the reputations which Darius and Xerxes had earned for themselves among their contemporaries.

As for Mardonius, he was Darius’ son-in-law, and had commanded the Persian fleet when it met with disaster on the rocks off Mount Athos. Darius’ dissatisfaction with him is clear, for in the subsequent expedition which that monarch launched against Greece, Mardonius was not in command. Datis and Artaphernes were in charge of the fleet which sailed across the central Aegean to Eretria and Marathon. However, Mardonius was a man of no mean ability and his later reinstatement proves that he enjoyed Xerxes’ confidence. After Xerxes’ return to Persia, Mardonius tried by sensible diplomacy to divide the Greek states against one another before deciding to engage in battle with them. His chances of success in this diplomatic initiative were very good and with a little more perseverance he might have succeeded. But, cut off from supplies by sea, he perhaps had difficulty in feeding his large army and was accordingly under pressure to reach a decision with the utmost possible speed.

The Persian Fleet

No one who reads Herodotus’ narrative can underestimate the importance of the naval factor in the two Persian invasions. The Persians were an inland power and possessed no fleet of their own. It says all the more for the organizing ability of the Great Kings – Xerxes in particular – that they were able to muster such vast armadas. It also suggests that their knowledge of Greek seamanship and fighting power was such that they by no means despised the enemy with whom they had to deal.

The largest contingent of the Persian fleet consisted of Phoenician vessels, manned by Phoenician crews. Rather surprisingly, the Persians relied also upon ships and crews from the Greek Ionian cities which they had subjugated. Inevitably, they must have felt some doubts about the loyalty of the Greek contingents of their own fleet. On several occasions during the campaigns, the Ionian effort seems to have been half-hearted, and at the battle of Mycale the Ionian Greeks at last deserted their Persian overlords to aid their compatriots.

Artemisia, the Greek princess who ruled Halicarnassus (subject to Persian goodwill), was present herself on shipboard at the battle of Salamis, fighting on the Persian side. However, she seems to have joined either fleet as circumstances dictated at any particular moment, for when pursued by an Athenian vessel she deliberately rammed and sank another galley of her own contingent. The Athenians, thinking that she had changed sides, abandoned the pursuit and Artemisia made good her escape without further impediment.

The truth is possibly that Xerxes found it less risky to take the Ionian fleet with him than to leave it in his rear. On every ship there was a force of soldiers, either Persians, Medes or others whose loyalty was to be trusted. Persian commanders often took the place of local captains and Xerxes probably kept the leaders of the subject communities under his personal surveillance. Their position closely resembled that of hostages to the Persians.

Apart from the Phoenician and Greek naval contingents, there was in Xerxes’ fleet an Egyptian squadron which was to distinguish itself in the course of the fighting. We hear also of ships from Cyprus and Cilicia. Cyprus contained both Greek and Phoenician cities and the people of Cilicia were largely of Greek extraction. Whether the Cilicians felt any bond of sympathy with the Greeks of the mainland is another question, but only the links of empire united them with the Persians. The proportion of the total naval strength to that of the land army is recorded: the land forces, when counted by Xerxes at Doriscus in Thrace, were, according to Herodotus, 1,700,000 strong: the strength of the fleet is given with some precision as 1,207 vessels, not including transports.

Persian Naval Strategy

It is interesting that Xerxes reverted to his father’s original plan and decided to invade Greece from the north. He must have considered that his channel through the Athos peninsula eliminated the main hazard of this route. Clearly, he could deploy a much larger army in Greece if his land forces could make their own way along the coast. At the same time, the fleet keeping pace on the army’s flank contained transports which considerably eased his supply problem. The land forces carried a good deal of their own baggage and equipment with the help of camels and other beasts of burden. These did not include horses. It was not customary in the ancient world to use horses for such purposes and it is noteworthy that Xerxes transported his horses by sea on special ships. Horseshoes were unknown in the ancient centres of civilization, and it is possible that the Persian cavalry might have reached Greece with lame mounts if their horses had been obliged to make the whole journey by land.

Warships were, of course, necessary to protect both the transports and the land forces. Without naval defence, the Persian army would have been exposed to the danger of Greek amphibious attacks on its flank and its rear. Moreover, it was Xerxes’ hope that he would crush any Greek naval units immediately, wherever he met them.

He met them first at Artemisium, on the northern promontory of Euboea. Several actions were fought there, with varying outcome. The Greek position was well chosen. In the narrow channel between the Euboean coast and the mainland, the Greeks could not be enveloped by superior numbers. At the same time, they guarded the flank of Leonidas’ forces at Thermopylae. If the Persians sailed round Euboea to attack them in the rear, then the Persian land forces would be separated from their seaborne support. What took the Greeks by surprise was the enormous size of Xerxes’ force, which despite all reports far exceeded their most pessimistic estimates. It was possible for Xerxes to send one section of his fleet round the south of Euboea while he engaged the Greeks at Artemisium with the remainder. Such a manoeuvre entailed no loss of numerical superiority on either front. But summer storms gathered over Thessaly and aided the Greeks. The very size of Xerxes’ fleet meant that there were not sufficient safe harbours to accommodate all the ships; a considerable part of it had to lie well out to sea in rough weather. In this way many ships were wrecked. When a squadron was dispatched to round Euboea and sail up the Ruripus strait, which divides the long island from the mainland, this contingent also fell victim to storms and treacherous currents. The task assigned to it was never carried out.

Quite apart from the figures given by Herodotus, events themselves testify to the huge size of the Persian armada. Despite the heavy losses suffered at Artemisium, Xerxes’ fleet still enjoyed the advantage of dauntingly superior numbers when, late in the same season, the battle of Salamis was fought. Even after Salamis, the number of surviving ships and crews was such that the Greek fleet at Mycale hesitated long before attacking them.

Communication Networks – The Royal Road

Reliable and efficient communications throughout the Empire were a necessary component for its success. The construction, maintenance, and guarding of an extensive network of roads and bridges required a great deal of engineering expertise, manpower, and expense. The Persians adopted and adapted their predecessors’ systems, and greatly expanded them, to facilitate communication across vast distances. Individuals or groups on state business carried sealed documents that allowed access to supplies or provisions en route to their destination.

The most famous of these roads, though it was only one of many, was what Herodotus called the Royal Road from Susa in Elam to Sardis in Lydia (5.52–53). Any “royal” road would have, in fact, run through Persepolis and points eastward, so Herodotus’ terminology reflects a Greek view, which usually viewed Susa as the main Achaemenid capital. From the west it ran through Cappadocia and Cilicia in Anatolia to Armenia and then south through Arbela – along the Tigris River – and on toward Susa. Herodotus notes that there were 111 royal staging posts interspersed on it and mentions several of them specifically (5.52). By his calculations this route ran roughly 1,500 miles and took a journey of ninety days. That was for a traveler in no great haste. Royal dispatches could move with surprising speed, a relay system with fresh horses and messengers at each staging post. Herodotus also describes these royal messengers: “There is nothing mortal that travels faster than these messengers … for as many days as the whole route there are horses and men stationed, one horse and one man set for each day. Neither snow, nor rain, nor heat, nor night hinders them from accomplishing the course laid before them as quickly as possible. After the first one finishes his route, he delivers the instructed message to the second, the second does likewise to the third; from there in rapid succession down the line the message moves.” (8.98)

There were similar routes in all directions from the Empire’s core in Fars.11 Ctesias alludes to other roads running from Mesopotamia and Persia proper to Central Asia. The primary route to Bactria across northern Iran is called in modern works either the (Great) Khorasan Road or, for later periods, by its better known appellation the Silk Road. Administrative documents from Persepolis, Syro-Palestine, and Egypt record disbursements to travelers in all directions. From the Persepolis documentation we gain a sense of the itineraries of a number of the network of roads running between Susa and Persepolis. An Aramaic document tracks travelers journeying from northern Mesopotamia to Damascus and on into Egypt, with several stops along the way listed by name.

Large work crews were involved in the construction and maintenance of these roads. Herodotus’ account of Xerxes’ invasion of Greece describes roadmakers at work, not infrequently the army on campaign. The main roads, constructed wide enough to allow chariots or wagons to travel on them, served to move military forces quickly, but they were also used by travelers or merchants to transport cargo. Roads also at times had to cross obstacles such as rivers. Some permanent bridges, such as one spanning the Halys River in Anatolia, were guarded by a fort. Pontoon bridges allowed crossing of other rivers, for example, at many spots on the northern Tigris and the Euphrates Rivers and their tributaries. Temporary pontoon bridges afforded the means for Persian armies to cross into Europe: Darius I over the Bosporus on his campaign against the Scythians and Xerxes’ bridge over the Hellespont against the Greeks. Of course, rivers and larger waterways were sometimes part of the route. Diodorus Siculus (14.81.4) records a journey on a well-known route at sea along the coast of Cilicia, on land from northwestern Syria to the Euphrates, then down the river to Babylon. Similar sea trading routes connected other parts of the Empire to the core, such as through the Persian Gulf and along the southern coast of Iran to the Indus Valley.

Cyrus and the Achaemenids

Around 559 BCE a Persian prince called Cyrus (modern Persian Kurosh), claiming descent from the royal house of Persia and from its progenitor Achaemenes, became king of Anshan on the death of his father. Persia and Anshan at that time were still subject to the Median Empire, but Cyrus led a revolt against the Median king Astyages, and in 549 captured the Median capital, Ecbatana (modern Hamadan). Cyrus reversed the relationship between Media and Persia, making himself king of Persia, making Media the junior partner, and Persia the centre of the Empire. But he did not stop there. He went on to conquer Lydia, in Asia Minor, taking possession of the treasury of King Croesus, legendary for his wealth throughout antiquity. He also conquered the remaining territories of Asia Minor, and also Phoenicia, Judaea and Babylonia, creating an enormous empire that stretched from the Greek cities on the eastern coast of the Aegean to the banks of the river Indus—in extent perhaps the greatest empire the world had seen up to that time.

But without romanticising him unduly, and although it took on much of the culture of previous Elamite, Assyrian and Babylonian empires (notably in its written script and monumental iconography), it seems that Cyrus aspired to rule an empire different from others that had preceded it in the region. Portentous inscriptions recording the military glory of Kings and the supposed favour of their terrible war-gods were a commonplace in the Middle East in the centuries preceding Cyrus’s accession. In the nineteenth century an eight-sided clay object (known since as the Taylor Prism after the man who found it) measuring about 15 inches long by 5.5 inches in diameter, covered in cuneiform script, was found near Mosul. When the characters were eventually deciphered, it was found to record eight campaigns of the Assyrian king Sennacherib (705 BCE–681 BCE). An excerpt reads:

Sennacherib, the great king… king of the world, king of Assyria, king of the four quarters… guardian of right, lover of justice, who lends support, who comes to the aid of the needy, who performs pious acts, perfect hero, mighty man, first among all princes, the flame who consumes those who do not submit, who strikes the wicked with the thunderbolt; the god Assur, the great mountain, has entrusted an unrivaled kinship to me… has made powerful my weapons… he has brought the black-headed people in submission at my feet; and mighty kings feared my warfare…

In the course of my campaign, Beth-Dagon, Joppa, Banaibarka, Asuru, cities of Sidka, who had not speedily bowed in submission at my feet, I besieged, I conquered, I carried off their spoils… I approached Ekron and slew the governors and nobles who had rebelled, and hung their bodies on stakes around the city…

As for Hezekiah the Jew, who did not submit to my yoke: 46 of his strong, walled cities… by means of ramps and by bringing up siege-engines… I besieged and took them. 200,150 people, great and small, male and female, horses, mules, asses, camels, cattle and sheep without number, I brought away from them and counted as spoil…

The way that the pharaohs of Egypt celebrated their rule and their victories was similar, and although Hezekiah, the king of Jerusalem, appears on the Taylor Prism as a victim, some parts of the Bible describing the Israelites and their God smiting their enemies do not read very differently.

By contrast, another not dissimilar clay object, about nine inches by four inches, also discovered in the nineteenth century and covered in cuneiform script, tells a rather different story. The Cyrus cylinder, now in the British Museum, was found where it had been deliberately placed under the foundations of the city wall of Babylon, and has been described as a charter of human rights for the ancient world. This is an exaggeration and a misrepresentation, but the message of the cylinder, particularly when combined with what is known of Cyrus’s religious policy from the books of Ezra and Isaiah, is nonetheless remarkable. The kingly preamble from the cylinder is fairly conventional:

I am Cyrus, king of the world, great king, rightful king, king of Babylon, king of Sumer and Akkad, king of the four quarters (of the earth), son of Cambyses, great king, king of Anshan, grandson of Cyrus, great king, king of Anshan, descendant of Teispes, great king, king of Anshan, of a family that always exercised kingship…

but it continues, describing the favour shown to Cyrus by the Babylonian god Marduk:

When I entered Babylon as a friend and when I established the seat of the government in the palace of the ruler under jubilation and rejoicing, Marduk, the great lord, induced the magnanimous inhabitants of Babylon to love me, and I was daily endeavouring to worship him. My numerous troops walked around in Babylon in peace, I did not allow anybody to terrorize any place of the country of Sumer and Akkad. I strove for peace in Babylon and in all his other sacred cities…

and concludes:

As to the region… as far as Assur and Susa, Agade, Eshnunna, the towns of Zamban, Me-Turnu, Der as well as the region of the Gutians, I returned to these sanctuaries on the other side of the Tigris, the sanctuaries of which had been ruins for a long time, the images which used to live therein and established for them permanent sanctuaries. I also gathered all their former inhabitants and returned to them their habitations. Furthermore, I resettled upon the command of Marduk, the great lord, all the gods of Sumer and Akkad whom Nabonidus had brought into Babylon to the anger of the lord of the gods, unharmed, in their former chapels, the places that make them happy.

Like the proud declarations of Sennacherib, this is propaganda, but it is propaganda of a different kind, presenting Cyrus in a different light, and according to a different scale of values. Cyrus chose to present himself showing respect to the Babylonian deity, Marduk, and declared that he had returned to other towns and territories the holy images that previous Babylonian kings had confiscated. Perhaps it would have been different if Cyrus had conquered Babylon by force, rather than marching into it unopposed (in 539 BCE) after its inhabitants revolted against the last Babylonian king, Nabonidus. Cyrus was a ruthless, ambitious man; no-one ever conquered an empire without those characteristics in full measure. But we know that Cyrus permitted freedom of worship to the Jews too. He and his successors permitted them to return home from exile and to rebuild the temple in Jerusalem (being accorded in the Jewish scriptures a unique status among gentile monarchs in return).

The logic of statecraft alone might have suggested that it would be more sustainable in the long run to let subjects conduct their own affairs and worship as they pleased, but that policy had to be acceptable to the Iranian elite also, including the priests, the Magi, and (leaving aside the question of Cyrus’s own personal beliefs, which remain unclear) it is reasonable to see in the policy some of the spirit of moral earnestness and justice that pervaded the religion of Zoroaster. The presence of those values in the background helps to explain why the Cyrus cylinder is couched in such different terms from the militaristic thunder and arrogance of Sennacherib. The old answer was terror and a big stick, but the Persian Empire would be run in a more devolved, permissive spirit. Once again, an encounter with complexity, acceptance of that complexity, and a response. This was something new.

Unfortunately, according to Herodotus, Cyrus did not end his life as gloriously as he had lived it. Having conquered in the west, he turned to campaign east of the Caspian, and according to one version was defeated and killed in battle by Queen Tomyris of the Massagetae, another Iranian tribe who fought mainly on horseback, like the Scythians.

The Massagetae are interesting because they appear to have maintained some ancient Iranian customs that may shed some light on the status of women in Persian society under the Achaemenids. There are signs in Herodotus (Book 1:216) that the Massagetae showed some features of a matrilineal, polyandrous society, in which women might have a number of husbands or sexual partners (but men only one). Patricia Crone has suggested that this feature may resurface in the apparent holding of women in common practised later by the Mazdakites in the fifth century AD and the Khorramites after the Islamic conquest, which might indicate an underlying folk tradition. Mazdaism certainly permitted a practice whereby an impotent man could give his wife temporarily to another in order to obtain a child (it also sanctioned the marriage of close relatives). But in general the custom of Persian society seems to have limited the status of women, following practices elsewhere in the Middle East. Royal and noble women may have been able to own property in their own right and even, on occasion, to exert some political influence; but this seems to have been an exception associated with high status rather than indicative of practices prevalent in society more widely.

Cyrus’s body was brought back to Persia, to Pasargadae, his capital, to rest in a tomb there that can still be seen (though its contents have long since disappeared). The tomb is massively simple rather than grandiose; a sepulchre the size of a small house on a raised, stepped plinth. This tomb burial has raised some questions about the religion of Cyrus and the other Achaemenid kings (many of his successors were placed in tombs of a different type—rock tombs half-way up a cliff-face). Tomb burial was anathema to later Zoroastrians, who held it to be sacrilege to pollute the earth with dead bodies. Instead they exposed the dead on so-called Towers of Silence to be consumed by birds and animals. Could the Achaemenid kings really have been Zoroastrians if they permitted tomb burial?

Some have explained the inconsistency by suggesting that different classes of Iranian society followed different beliefs; effectively different religions. As we have seen, there probably was some considerable plurality of belief within the broad flow of Mazdaism at this time. But it seems more likely that the plurality was socially vertical rather than horizon-tal—in other words, a question of geography and tribe rather than social class. Perhaps an earlier, pre-Zoroastrian tradition of burial still lingered and the elevated position of all the royal tombs was a kind of compromise. Half-way between heaven and earth: itself a strong metaphor. Around the tomb of Cyrus lay a paradise; a garden watered by irrigation channels (our word paradise comes, via Greek, from the Old Persian paradaida, meaning a walled garden). Magian priests watched over the tomb and sacrificed a horse to Cyrus’ memory each month.

Cyrus had been a conqueror, but a conqueror with an imaginative vision. He was at least as remarkable a man as the other conqueror, Alexander, whose career marks the end of the Achaemenid period, as that of Cyrus marks the beginning. Maybe as a youth Cyrus had a Mazdaean tutor as remarkable as Aristotle, who taught Alexander.

Religious Revolt

Cyrus was succeeded by his son Cambyses (Kambojiya), who extended the empire by conquering Egypt, but in a short time gained a reputation for harshness. He died unexpectedly in 522 BCE, according to one source by suicide, after he had been given news of a revolt in the Persian heartlands of the empire.

An account of what happened next appears on an extraordinary rock relief carving at Bisitun, in western Iran, about twenty miles from Kermanshah, above the main road to Hamadan. According to the text of the carving (executed in Old Persian, Elamite and Babylonian) the revolt was led by a Magian priest, Gaumata, who claimed falsely to be Cambyses’s younger brother, Bardiya. Herodotus gives a similar version, saying that Cambyses had murdered the true Bardiya some years earlier. The revolt led by Gaumata seems to have drawn force from social and fiscal grievances, because one of his measures to gain popularity was to order a three-year remission of taxes—another to end military conscription. Pressure had built up over the decades of costly foreign wars under Cyrus and Cambyses. But Gaumata also showed strong religious enthusiasm or intolerance, because he destroyed the temples of sects he did not approve of.

An Iranian revolution, led by a charismatic cleric, seizing power from an oppressive monarch, asserting religious orthodoxy, attacking false believers, and drawing support from economic grievances—in the sixth century BCE. How modern that sounds. But within a few months, Gaumata was dead, killed by Darius (Daryavaush) and a small group of Persian confederates (a killing that sounds more like an assassination than anything else). The carving at Bisitun was made at Darius’s orders, and it presents his version of events, as put together after he had made himself king and the revolt had finally been put down. The carving itself says that copies of the same text were made and distributed throughout the empire. And what a revolt it had been—Babylon revolted twice, and Darius declared that he fought nineteen battles in a single year. It was really a series of revolts, affecting all but a few of the eastern provinces of the empire. The Bisitun carving illustrates this by showing a row of defeated captives, each representing a different people or territory. Whatever the true nature of the rebellion and its origins, it was no simple palace coup, affecting only a few members of the elite. It was just the first of several religious revolutions, or attempted revolutions, in Iran’s history, and it was no pushover.

Bisitun was chosen for Darius’s grand rock-carving because it was a high place, perhaps already associated with the sacred, close by where he and his companions had killed Bardiya/Gaumata. The site at Bisitun is a museum of Iranian history in itself. Aside from the Darius rock relief, there are caves that were used by Neanderthals 40,000 years ago or more, and by others generations later. Among other relics and monuments, there is a rock carving of a reclining Hercules from the Seleucid period, a Parthian carving depicting fire worship, a Sassanian bridge, some remains of a building from the Mongol period, a seventeenth-century caravanserai, and not far away, some fortifications apparently dating from the time of Nader Shah in the eighteenth century.

Many historians have been suspicious about the story of the false Bardiya. The Bisitun carving is a contemporary source, but it is plainly a self-serving account to justify Darius’s accession. It is confirmed by Herodotus and other Greek writers, but they all wrote later and would naturally have accepted the official version of events, if other dissenting accounts had been stamped out. Darius was not a natural successor to the throne. He was descended from a junior branch of the Achaemenid royal family, and even in that line he was not pre-eminent—his father was still living. Could a Magian priest have successfully impersonated a royal prince some three or four years after the real man’s death? Is it not rather suspect that Darius also discredited other opponents by alleging that they were impostors?

If the story was a fabrication, Darius was certainly brazen in the presentation of his case. In the Bisitun inscriptions, the rebel leaders are called ‘liar kings’, and Darius declares, appealing to religious feeling and Mazdaean beliefs about arta and druj:

[…] you, whosoever shall be king hereafter, be on your guard very much against Falsehood. The man who shall be a follower of Falsehood—punish him severely […]


[…] Ahura Mazda brought me aid and the other gods who are, because I was not disloyal, I was no follower of Falsehood, I was no evil-doer, neither I nor my family, I acted according to righteousness, neither to the powerless nor to the powerful, did I do wrong…

and again:

This is what I have done, by the grace of Ahura Mazda have I always acted. Whosoever shall read this inscription hereafter, let that which I have done be believed. You must not hold it to be lies.

Perhaps Darius protested a little too much. Another inscription in Darius’s words, from another site, reads:

By the favour of Ahura Mazda I am of such a sort that I am a friend to right, I am not a friend to wrong. It is not my desire that the weak man should have wrong done to him by the mighty; nor is it my desire that the mighty man should have wrong done to him by the weak. What is right, that is my desire. I am not a friend to the man who is a lie-follower […] As a horseman I am a good horseman. As a bowman I am a good bowman both afoot and on horseback…

The latter part of this text, though telescoped here from the original, echoes the famous formula from Herodotus and other Greek writers, that Persian youths were brought up to ride a horse, to shoot a bow, and to tell the truth. Darius was pressing every button to stimulate the approval of his subjects. Even if one doubts the story of Darius’s accession, the evidence from Bisitun and his other inscriptions of his self-justification and the use of religion by both sides in the intensive fighting that followed the death of Cambyses nonetheless stands. It is a powerful testimony to the force of the Mazdaean religion at this time. Even the suppressors of the religious revolution had to justify their actions in religious terms. Although Darius by the end reigned supreme, the inscriptions give a strong sense that he himself was nonetheless subject to a powerful structure of ideas about justice, truth and lies, right and wrong, that was distinctively Iranian, and Mazdaean.

The Empire Refounded

Darius’s efforts to justify and dignify his rule did not end there. He built an enormous palace in his Persian homeland, at what the Greeks later called Persepolis (‘city of the Persians’)—thus starting afresh, away from the previous capital of Cyrus at Pasargadae. Persepolis is so big that a modern visitor walking over the site, wandering bemused between the sections of fallen columns and the massive double-headed column capitals that crashed to the ground when the palace burned, may find it difficult to orientate himself and make sense of it. The magnificence of the palace served as a further prop to the majesty of Darius, and the legitimacy of his rule; but helped in turn to create a lasting tradition, a mystique of magnificent kingship that might not have come about but for the initial doubts over his accession. A dedicatory inscription at Persepolis played again on the old theme:

May Ahura Mazda protect this land from hostile armies, from famine, and from the Lie.

The motif of tribute and submission is also repeated from Bisitun: row upon row of figures representing subjects from all over the empire are shown queuing up to present themselves, frozen forever in stone relief. The purpose of the huge palace complex at Persepolis is not entirely clear. It may be that it was intended as a place for celebrations and ceremonial at the spring equinox, the Persian New Year (Noruz—celebrated on and after 21 March each year today as then). The rows of tribute-bearers depicted in the sculpture suggest that it may have been the place for annual demonstrations of homage and loyalty from the provinces. Whatever the grandeur of Persepolis, it was not the main, permanent capital of the empire. That was at Susa, the old capital of Elam. This again shows the syncretism of the Persian regime. Cyrus had been closely connected with the royal family of the Medes, and the Medes had a privileged position, with the Persians, as partners at the head of the Empire. But Elam too was important and central: its capital, its language, in administration and monumental inscriptions. This was an empire that always, for preference, flowed around and absorbed powerful rivals: its first instinct, unlike other imperial powers, was not to confront, batter into defeat, and force submission. The guiding principles of Cyrus persisted under Darius and at least some later Achaemenid rulers.

Darius’s reign saw the Achaemenid empire in effect re-founded. It could have gone under altogether in the rebellions that followed the death of Cambyses. Darius maintained Cyrus’s tradition of tolerance, permitting a plurality of gods to be worshipped as before; and maintained also the related principle of devolved government. The provinces were ruled by satraps, governors who returned a tribute to the centre but ruled as viceroys (two other officials looked after military matters and fiscal administration in each province, to avoid too much power being concentrated in any one pair of hands). The satraps often inherited office from predecessors within the same family, and ruled their provinces according to pre-existing laws, customs and traditions. They were, in effect, provincial kings; Darius was a King of Kings (Shahanshah in modern Persian). The empire did not attempt as a matter of policy to Persianise as the Roman empire, for example, later sought to Romanise.

The certainties of religion, the principle of sublime justice they underpinned, and the magnificent prestige of kingship were the bonds that held together this otherwise diffuse constellation of peoples, languages and cultures. A complex empire, accepted as such, and a controlling principle. The system established by Darius worked, proved resilient, and endured.

Tablets discovered in excavations at Persepolis show the complexity and administrative sophistication of the system Darius established. Although some payments were made in silver and Darius established a standard gold coinage, much of the system operated by payments in kind; assessed, allocated and receipted for from the centre. State officials and servants were paid in fixed quantities of wine, grain or animals; but even members of the royal family received payments in the same way. Officials in Persepolis gave orders for the levying of taxes in kind in other locations, and then gave orders for payments in kind to be made from the proceeds, in the same locations. Messengers and couriers were given tablets to produce at post-stations along the royal highways, so that they could get food and lodging for themselves and their animals. These tablets recording payments in kind cover only a relatively limited period, from 509 to 494 BCE. But there are several thousand of them, and it has been estimated that they cover supplies to more than 15,000 different people in more than 100 different places.

It is significant that the tablets were written mainly in Elamite, not in Persian. It is known from other sources that the main language of administration in the Empire was neither Persian nor Elamite, but Aramaic, the Semitic lingua franca of Mesopotamia, Syria and Palestine. The Bisitun inscription states directly that the form of written Persian used there was new, developed at Darius’s own orders for that specific purpose. It is possible that he and the other Achaemenid kings discouraged any record of events other than their own monumental inscriptions, but these are all strong echoes of that Iranian distaste for writing that we encountered earlier in Mazdaism, and it may go some way to explain an apparent anomaly—the lack of Persian historical writing for the Achaemenid period. It is possible that the histories once existed, and that there were poems written down and all sorts of other literature which have since been simply lost. But later Persian literary culture was strongly associated with a class of scribes, and the fact that the scribes in the Achaemenid system wrote their accounts and official records in other languages suggests that the literature was not there either. There was no Persian history of the Achaemenid Empire because the Persian ruling classes either (like the Magi) regarded writing as wicked or (the kings and nobles) associated it with inferior subject peoples; or both. To ride, to shoot the bow, to tell the truth; but not to write it. That said, no histories as such have survived from the Egyptian, Hittite or Assyrian empires either—it is more correct, in the context of the fifth century BCE, to call the innovation of history writing by the Greeks the anomaly.

To ourselves, at our great remove of time, awash with written materials every moment of our working lives, dominated by the getting and spending of money, a human system that was largely non-literate and operating for the most part on the basis of payments in kind, not cash, even if it be a great empire capable of stunning monuments and great sculptural art, seems primitive. But the history of human development is not simply linear. It is not quite right to see the oral tradition of sophisticated cultures like that of Mazdaism as unreliable, flawed or backward, something we have gone beyond. The Persians were not stupidly trying, with the wrong tools, to do something we can now, with the right tools, do incomparably better. They were doing something different, and had evolved complex and subtle ways of doing it very well indeed, which our culture has forgotten. To try to grasp the reality of that we have to step aside a little from our usual categories of thought, for all the apparent familiarity of Mazdaean concepts like angels, the day of judgement, heaven and hell, and moral choice. The Achaemenid Empire was an Empire of the Mind, but a different kind of Mind.

The Empire and the Greeks

In general Darius’s reign was one of restoration and consolidation of previous territorial expansion rather than wars of conquest like those that had been pursued by Cyrus and Cambyses. But Darius campaigned into Europe in 512 BCE, conquering Thrace and Macedonia, and toward the end of his life, after a revolt by the Ionian Greeks of the Aegean coast of Asia Minor, his subordinates fought a war with the Athenian Greeks that ended with a Persian defeat at the battle of Marathon in 490 BCE. This ushered in what the Greeks called the Persian wars, the shadow of which has affected our view of the Achaemenid Empire, and perhaps Persia and Iran and the Orient generally, ever since. From a Persian perspective, the more serious event was a revolt in Egypt in 486 BCE. Before he could deal with this, Darius died.

The standard Greek view of the Persians and their empire was complex, and not a little contradictory. They regarded the Persians, as they regarded most non-Greeks, as barbarians (the term barbarian itself is thought to come from a disparaging imitation of Persian speech—‘ba-ba’), and therefore ignorant and backward. They were aware that the Persians had a great, powerful, wealthy empire. But for them it was run on tyrannical principles, and was redolent of vulgar ostentation and decadence. The Persians were therefore both backward and decadent—at which point we may be irresistibly reminded (via the judgement of that supreme chauvinist, Clemenceau, that ‘America is the only nation in history that miraculously has gone directly from barbarism to degeneration without the usual interval of civilisation’) of the contemporary French view of the United States. Perhaps the view of the Greeks also was better explained in terms of a simple resentment or jealousy that the Persians rather than the Greeks were running such a large part of the known world.

This in itself is a caricature of the Greek view of the Persians, and cannot have been, for example, Plato’s attitude or the attitude (openly, at any rate) of the many Greeks who worked for or were allies of the Persians at various times. The Greeks were also an imperialistic, or at least a colonising culture, of pioneering Indo-European origin. Perhaps, as at other times and in other places, the hostility between the Persians and the Greeks had as much to do with similarity as with difference. But in contrast to the Persians the Greeks were not a single unified power, being composed of a multiplicity of rival city-states, and their influence was maritime rather than land-based. Greeks had established colonies along almost all parts of the Mediterranean littoral that had not previously been colonised by the Phoenicians (including the places that later became Tarragona in Spain, Marseilles in France, Cyrenaica in Libya and large parts of Sicily and southern Italy), and had done the same on the coast of the Black Sea. Unlike the Persians again, their spread was based on physical settlement by Greeks, rather than the control of indigenous peoples from afar.

Just as Persians appear in the plays of the great Greek playwrights, and on Greek vases, there are examples to show the presence of the Greeks in the minds of the Persians. As well as vases that show a Greek spearing a falling or recumbent Persian, there are engraved cylinder seals showing a Persian stabbing a Greek or filling him with arrows. But it is fair to say that at least initially, the Persians were more present to the Greeks than the Greeks to the Persians. Persian power controlled important Greek cities like Miletus and Phocaea in Asia Minor, only a few hours’ rowing away from Athens and Corinth—as well as Chalcidice and Macedonia on the European side of the Bosphorus. In Persepolis, Susa and Hamadan by contrast, Greece would have seemed half a world away; and events in other parts of the empire, like Egypt, Babylonia and Bactria for example, equally or rather more pressing.

Darius was succeeded by his son, Xerxes (Khashayarsha). The set-piece of Xerxes’s reign in the historical record was the great expedition to punish Athens and her allies for their support of the Ionian revolt, but at least as important for Xerxes himself would have been his successful reassertion of authority in Egypt and Babylon, where he crushed a rebellion and destroyed the temple of Marduk that Cyrus had restored. Xerxes is believed (on the authority of Herodotus) to have taken as many as two million men with him to attack Athens in 480 BCE. His troops wiped out the rearguard of Spartans and others at Thermopylae (when Xerxes asked them to surrender, demanding that they lay down their weapons, the Spartans replied ‘come and get them’), killing the Spartan king Leonidas there in a protracted struggle that left many of the Persian troops dead. Xerxes’s men then took Athens, his hardy soldiers scaling the Acropolis from the rear and burning it, but his fleet was defeated at Salamis, leaving his armies overextended and vulnerable. He withdrew to Sardis, his base in Asia Minor, and his forces suffered further, final defeats the following year at Plataea and Mycale (479 BC). Among other effects of the Persian defeat was the loss of influence on Macedon and Thrace on the European side of the Bosphorus, permitting the subsequent rise of Macedon.

Xerxes’s son Artaxerxes (Artakhshathra) succeeded him in 465 BCE, and reigned until 424 BCE. The building work at Persepolis continued through the reigns of both, and it was under these two kings that many of the Jews of Babylonia returned to Jerusalem, under the leadership of Ezra and Nehemiah. The latter was Artaxerxes’ court cupbearer in Susa, and both returned eventually to the Persian court after their efforts to rebuild Jerusalem. The books of Ezra and Nehemiah give a different picture of the Persian monarchy to contrast with the less flattering image in the Greek accounts.

The wars that continued between the Persians and the Greeks ended at least for a time with the peace of Callias in 449 BCE, but thereafter the Persians supported Sparta against Athens in the terribly destructive Peloponnesian wars, which exhausted the older Greek city-states and prepared the way for the hegemony of Macedon. At the death of Artaxerxes palace intrigues caused the deaths by murder of several kings or pretenders in succession. In the reign of Artaxerxes II (404-359 BCE) there were further wars with the Greeks, and a sustained Egyptian revolt that kept that satrapy independent until Persian rule was restored under Artaxerxes III in 343 BCE. Palace intrigue and murder had already claimed the lives of several of the Achaemenid kings, but a particularly lethal round of events orchestrated by the vizier or chief minister Bagoas caused the deaths of both Artaxerxes III and his son Arses, bringing Darius III to the throne in 336 BCE.

The Iranians must have changed their way of life considerably over the two centuries between the reigns of Cyrus and Darius III. One indicator of social change (as is often the case) was the constitution of their armies. At the time of Xerxes’ invasion of Greece and before, large numbers of Medes and Persians fought on foot, but by the time of Darius III the armies were dominated by large numbers of horsemen and the previous Assyrian-style big units of spear-and-bow armed infantry (and shield-bearers—sparabara) seem to have disappeared (though there were Greek mercenary infantry, and Persian infantry called Cardaces who may have been young men in training for the cavalry). The impression is that the wealth of empire had enabled the Iranian military classes to distribute themselves across the empire and supply themselves with horses, changing the nature of Persian warfare (though there seems also to have been a deliberate policy of military garrisoning and military colonies, notably in Asia Minor). According to Herodotus, Cyrus had warned that if the Persians descended to live in the rich lands of the plain (he probably had Babylonia particularly in mind) they would become soft and incapable of defending their empire. It is too neat to suggest that this is precisely what happened—it may be somewhat the contrary, that by the time of Darius III taxes had risen too high and the Iranians, having had their expectations raised, had become impoverished and demoralised. But whatever their exact nature, fundamental changes had taken place, and Iran had already moved closer to the social and military patterns of the later Parthian and Sassanid empires.




AD 68, to serve new emperor Galba.


Initially, Gallia Narbonensis and Italy.


Misenum, Spain, Mogontiacum, Sirmium, Brigetio, Dacia, Parthia, Brigetio.


Battle of Old Camp, AD 70.

Trajan’s Dacian Wars, AD 101–106.

Trajan’s Partian Campaign, AD 114–116.

Marcus Aurelius’ German Wars, AD 161–180.


Publius Hevius Pertinax, future emperor.


Thrown together during the war of succession, fighting on Otho’s losing side before making a name for itself under Trajan, it would be one of Stilicho’s legions in the last desperate battles before the fifth-century fall of Rome.

A legion with surprising beginnings

In the late spring of AD 68, in a desperate bid to keep his throne, the 30-year-old emperor Nero raised a new legion, taking the unprecedented step of enlisting sailors from the Roman battle fleet based at Misenum, on the east coast of Italy, for legionary service. But the sailors could not save him; or would not. With both the Praetorian Guard and his German Guard bodyguard deserting him, and with the Senate declaring him an enemy of the state and sending troops to arrest him, on June 9 Nero apparently committed suicide. The Senate had already recognized the claim to the throne of 70-year-old Sulpicius Galba, governor of the province of Nearer Spain, and that autumn Galba came marching to Rome from Spain, attended by an entourage which included a new 7th Legion he had raised there, and a large body of cavalry.

In the meantime, Nero’s legion of seamen had sat stubbornly in Rome awaiting developments. With no quarters, they slept wherever they could around the city. At the time, Rome was crowded with legion detachments summoned to Rome by Nero during the last gasps of his reign; those troops, including men from the 11th Claudia and 15th Apollinaris legions, had resorted to sleeping in temples and public buildings. [Tac., H, I, 31] The seamen from Misenum had not been presented with an eagle and standards to signify that their legion was officially constituted, but they were determined to gain recognition of their unit; with that recognition would come a grant of Roman citizenship to each of them.

At this time, seamen and marines serving in Rome’s navy were not citizens. Neither were they slaves. Contrary to the erroneous picture painted by nineteenth-century authors, Rome’s sailors of this era were salaried free men who possessed neither Latin status nor Roman citizenship. [Starr, III, 3, and V, 1] Once the much valued prize of citizenship had been dangled before them by Nero, the seamen from Misenum were determined to win it from the new emperor Galba. Consequently, when news reached Rome in October AD 68 that Galba and his column from Spain were approaching, the 5,000 sailors of the new legion went flooding out of the city gates, joining the thousands of civilians gathered there to greet him.

Three miles (4.8 kilometers) north of Rome, this “disorderly rabble of the seamen,” as Plutarch described them, “those whom Nero had made soldiers, forming them into a legion,” crowded around Galba and loudly demanded “to have their commission confirmed.” [Plut., Galba] Preventing the emperor from being seen or heard by the crowds lining the route into the city, the ex-sailors “tumultuously pressed him, shouting loudly to have colors for their legion and quarters assigned to them.” [Ibid.] Galba tried to put them off, saying he would consider the matter later, and rode on.

But the seamen were not satisfied with this response, “which they interpreted as a denial” of their request. [Ibid.] Growing “more insolent and mutinous” and “some with drawn swords in their hands,” they continued to follow him, yelling their demands. [Ibid.] The sight of the sailors’ drawn swords frightened Galba, and as the column approached the Milvian Bridge over the Tiber river he “ordered the cavalry to ride over them.” [Ibid.] The seamen, the vast majority of whom were unarmed, “were soon routed” by the cavalry. Not a man stood his ground, “and many of them were killed, both there and in the pursuit” as they tried to flee back to the city. [Ibid.]

According to Tacitus, the affair resulted in “the slaughter of thousands of unarmed soldiers” of the unofficial legion by Galba’s cavalry. [Tac., H, I, 6] Cassius Dio, writing of the event more than 150 years later, estimated that “about 7,000 perished on the spot, and the survivors were later decimated,” with one in ten executed. [Dio, LXIII, 3] But 7,000 was certainly an exaggerated figure; an imperial legion only numbered a little over 5,200 men. And there is no other record of the decimation.

Word of Galba’s cold-blooded act of brutality against his own men at the Milvian Bridge soon spread around the empire, and did nothing to endear their new emperor to the Romans. The event was so impressed on the mind of Plutarch, who was at the time a young man in his twenties, and that of fellow historian Tacitus, then in his early teens, that both would observe that this was a bad omen for the new emperor’s reign “that Galba should make his first entry [to Rome] through so much blood and among dead bodies.” [Plut., Galba] Despite this lethal treatment, the surviving seamen hardened their resolve to gain recognition. The legion “which Nero had levied from the fleet” still remained in the congested capital, albeit in custody, and significantly reduced in numbers. [Tac., H, I, 6, 87]

This legion’s tortured beginnings were now about to take another turn. Tacitus wrote that, several months later, the city of Vienna “had recently raised legions for Galba.” [Tac., H, I., 65] This was not today’s Vienna in Austria, but present-day Vienna, in southern France. Roman Vienna was a leading city of the province of Narbon Gaul, through which Galba had passed on his march from Spain to Rome. [Plut., Galba] Situated on the south bank of the Rhône, Vienne, the capital of the powerful Allobroges tribe in Celtic times, had become one of the wealthiest cities in Gaul, even advertising its wealth with an inscription above the city gates, “VIEN FLOR FELIX,” which declared that Vienna was rich and flourishing. Such riches, and such boasts, could only attract the avaricious attention of neighbors who coveted “the gold of the men of Vienna.” [Tac., H, II, 29] And so it was to prove.

From AD 67 to AD 69, Vienna and the neighboring city of Lugdunum, today’s Lyon, were in a state of “perpetual feud.” [Tac., H, I, 65] Rivalry between the two went back as far the first century BC, when Vienna had expelled Roman colonists, who had subsequently been taken in by Lugdunum. When Gallic governor Vindex rose in revolt against Nero in AD 67, Lugdunum immediately threw its support behind Vindex, while Vienna retained its loyalty to Nero. During this period, Vienna had even sent armed men to raid Lugdunum. To keep the peace following the Vindex Revolt, Nero’s Palatium stationed the new 1st Italica Legion in Lugdunum, supporting the 18th Cohort of Rome’s City Guard, which was there to guard Lugdunum’s imperial mint. In an ironic twist, Lugdunum had then switched its support to Nero, and Vienna to Galba.

According to Tacitus, the people of Lugdunum now “began to work on the passions of individual soldiers, and to goad them into destroying Vienna.” [Ibid.] Tacitus says that in trying to coerce the 1st Italica legionaries into attacking Vienna, the people of Lugdunum claimed that while their city had begun as a colony of Roman legion veterans, the people of Vienna were foreigners. This potential threat, of an attack by the 1st Italica Legion, appears to have spurred the people of Vienna to come up with a novel solution, the formation of the first of their “legions for Galba” mentioned by Tacitus, levying young men locally.

Vienna’s first objective was the creation of a force to protect their city from attacks sponsored by Lugdunum, but the elders of Vienna would claim that they were merely creating legions in support of the nearby 1st Legion, the Italica, out of loyalty to their new emperor. Hence, the name taken by this, the first of Vienna’s new legions for Galba, was the 1st Adiutrix, or 1st Supporter Legion; literally, the legion in support of the 1st. Several months later, as Galba passed through their province on his way to Rome, the Viennase would have presented him with their new legion—a unit with perhaps a name but without an eagle, standards, or official standing—and Galba would have added Vienna’s recruits to his train as he marched on.

On December 22, apparently in a Saturnalia Festival act of clemency connected with his birthday, which was just two days away, Galba released some of the imprisoned seamen who had survived the massacre in October outside the city, discharging from military service those considered too old or too unfit to be of further use to the State. [Starr, VIII] The discharge diplomas issued to these men show that up to that point they had not received the Roman citizenship promised by Nero. Meanwhile, the remaining seamen from the Milvian Bridge massacre continued to languish in prison.

At this same time, Galba conveyed eagle and standards to the new legion, officially commissioning it into service as the 1st Adiutrix Legion. [Ibid.] That the legion was officially constituted by Galba, not Nero, is confirmed by Cassius Dio. [Dio, LV, 24] The December 22 timing of this formal presentation ceremony meant that from this time forward the legion would display the astrological birth sign of Capricorn.

Meanwhile, the remaining seamen of Nero’s legion enlistment were still imprisoned. [Tac., H, I, 87] So who was filling the legion’s ranks? It would seem that it was Vienna’s citizen recruits. Twenty-four days later, on January 15, AD 69, Galba was assassinated in Rome by a disaffected soldier of the 15th Apollinaris Legion. The Praetorian Guard at once hailed as their new emperor Otho, the former governor of Lusitania, who had marched to Rome with Galba the previous autumn. The Senate endorsed their choice.

Knowing how unpopular Galba had become with the military, one of Otho’s first acts was to win the loyalty of the fleet at Misenum. Tacitus records how Otho achieved this: Otho “enrolled in the ranks of the legion the survivors of the slaughter at the Milvian Bridge, who had been retained in custody by the stern policy of Galba.” [Tac., H, I, 87] That is, Otho added to the already existing 1st Adiutrix Legion the sailors he now released from custody. On being taken into the legion and joining the Viennase recruits, these seamen would be granted the Roman citizenship for which they had hungered. This diverse mix produced 1st Adiutrix soldiers who were, according to Plutarch, “strong and bold.” [Plut., Otho] As for the rest of the sailors of the fleet at Misenum, to them Otho “held out hopes of a more honorable service in the future”; they, too, might aspire to citizenship eventually. [Ibid.]

With the unit’s official commissioning by Galba, the name of the 1st Adiutrix Legion was formalized, as was its emblem, Pegasus the flying horse. In mythology, Pegasus was the son of Neptune, god of the sea, which would seemingly make the flying horse an appropriate symbol for a legion whose first recruits had come from the navy. Yet, as Starr points out, the seamen of Rome’s battle fleets showed no inclination to worship Neptune. [Starr, IV, 2] Neither did they worship Castor and Pollux, the patron deities of merchant sailors. In fact, the men of the fleet at Misenum venerated Isis, the patron goddess of sailors in Hellenistic times, who was believed to control the weather. [Ibid.]

Another new legion to adopt the Pegasus emblem, the 2nd Adiutrix, was raised the following year. This unit would also have a connection with both Vienna and the Roman navy. Apart from these two units, only one other imperial Roman legion is known to have employed the Pegasus emblem, and that was the 2nd Augusta Legion—a long-established and renowned unit known to use Gallia Narbonensis, a maritime province, as a recruiting ground. It may be that, rather than as a symbol of veneration of Neptune, both Adiutrix legions instead took Pegasus as their emblem in emulation of the 2nd Augusta, the “home” legion of Narbon Gaul, where they began life.

The 1st Adiutrix Legion spent that winter at Misenum, using the fleet’s quarters. Less than two months after the seamen officially joined the unit, the legion was ordered to prepare to march; its first battle was just weeks away. Ironically, the 1st Adiutrix faced the 1st Italica Legion, which it had been founded by Vienna to counter, in its first battle in April AD 69. It fought for Otho against Vitellius’ army at the First Battle of Bedriacum in northern Italy. Otho’s army lost; the 1st Adiutrix surrendered, after which Vitellius sent it to Spain.

In AD 70, the new emperor Vespasian transferred the 1st Adiutrix from Spain to Mogontiacum on the Rhine. Domitian stationed it in Pannonia. By the reign of Nerva it was at Brigetio, on the Danube. From there it took part in both of Trajan’s Dacian Wars, after which Trajan took it to the East for his Parthian campaign. From the reign of Hadrian the legion was back at Brigetio in Lower Pannonia, where it remained for the next 200 years defending the Danube.

In AD 193 the legion joined the march to Rome by the Pannonian legions that installed Septimius Severus on the throne after the Praetorian Guard murdered the popular soldier emperor Pertinax.

The Notitia Dignitatum shows the legion still in existence early in the fifth century, as part of the army of the Eastern Roman emperor, and stationed in the center of modern-day Hungary under the command of the Duke of Valeriae Ripensis.

The Fall of Ctesiphon and the Battle of Jalula

Pushtîghbân Grivpânvâr
[Royal Sassanid Cataphracts]

Farasan Mudar’a
(Islamic Heavy Cavalry

Syria, Palestine and Roman Mesopotamia were not the only regions in which the Muslims were steamrollering an increasingly desperate defence. Much like their victory at Yarmuk, their victory at Qadisiyyah and Sa’d’s ruthless pursuit of Jalinus had exposed the entirety of Persian Mesopotamia and left the road to Ctesiphon open. With its potential as a focal point and possible springboard for a counter-attack whilst still in Persian hands, Umar and Sa’d quickly decided that neutralising or capturing the Sassanid capital should be their next objective. Less than a fortnight after the victory at Qadisiyyah, Sa’d’s army, now reorganised into five separate corps under Zuhra, Abdullah, Shurahbeel, Hashim and Khalid b. Arfatah, set out across the alluvial plains of Mesopotamia towards Ctesiphon. Seemingly aware of the garrison forces along the route to Ctesiphon, Sa’d sent Zuhra on ahead with a strong advanced guard of cavalry, with orders to subdue the garrisons if he could but, should he come up against a sizeable imperial army, he was to await the main column that was proceeding at a more restrained pace.

However, despite the mixture of caution and alacrity, the Muslim advance met with only limited resistance. Zuhra was able to occupy Najaf without any opposition and, while the garrison at Burs resisted, Zuhra defeated its commander, Busbuhra, in single combat and the garrison was quickly overwhelmed. A battle near the ancient site of Babylon is recorded in December 636 against a large concentration of Persian forces, which given that its commanders were Hormuzan, Mihran, Nakheerzan and Beerzan suggests that it was the remnant of the force that Jalinus had extricated from Qadisiyyah. However, given the presence of Beerzan, supposedly killed by Qaqa, and the lack of clear information regarding any battle at Babylon, aside from there being dissension in the Persian ranks and the information that Hormuzan retreated to his homelands in Khuzestan the whole event should probably be downplayed as a major engagement.

Zuhra then continued his pursuit of congregating Persian forces. He is thought to have defeated a Sassanid force at Sura before catching up to Nakheerzan’s force at Deir Kab. Despite the killing of the Sassanid commander in a duel by one of Zuhra’s subordinates, the Persian force seems to have offered stiff resistance. It was only a successful flanking manoeuvre by Jarir that captured a bridge to the rear of the Persian lines that seems to have finally encouraged the Persians to retreat. The last Persian attempt to stall the Muslim advance to the gates of Ctesiphon came in early-January 637 at Kusa, a mere ten miles short of the capital. However, this time all it took was the defeat of the Persian commander, Shahryar, in a duel by one of the mubarizun to force the Persians to retreat.

With the capture of Kusa, nothing now lay between Sa’d’s forces and the walls of the Persian capital. However, despite the rapidity of Zuhra’s advance, properly defended, Ctesiphon would not be easily captured or even surrounded. This was because it was not a single city but a metropolis that incorporated several settlements including Seleucia, Veh-Ardashir, Vologaesocerta and others on the banks of the Tigris, as well as Ctesiphon itself. Indeed, in Arabic, Ctesiphon was and is known as al-Mada’in, meaning `The Cities’. With the direction they were approaching from – the west bank of the Tigris – the Muslims were to come to the sub-cities of Vologaesocerta, Seleucia and Veh-Ardashir first. Of these three, it appears that Yazdgerd and his generals focused their defensive efforts on Veh-Ardashir, probably due to it being the closest to Ctesiphon itself, digging ditches and placing ballistas and catapults. The presence of such siege engines forced the Muslims back from the walls but they quickly evened the odds by employing Persians to build siege engines for them.

By March 637, after almost two months of blockade, the Persian garrison was becoming desperate and sallied forth in an attempt to break the siege. In the subsequent fighting, Zuhra is said to have killed the Persian commander in a duel before being killed himself by an arrow. But one peculiar story stands out most from the siege of Veh-Ardashir: the Persians are said to have used a specially trained lion to disrupt the Muslim cavalry and infantry, with its rampage only being stopped by Hashim, who killed the beast with a single blow with his sword. One cannot help but suggest that this is a prime example of the corruption of the record of a Persian commander either called `lion,’ such as the Greek name Leo, or being described as fighting as fiercely as a lion.

As their sally proved ineffective, the Persians offered to recognise the Muslim conquest of all territory up to the banks of the Tigris in return for an end to the fighting. Sa’d replied by saying that peace would only come when Yazdgerd accepted Islam and paid the jizya. The next morning, the Muslims found Veh-Ardashir abandoned as the garrison had somehow managed to slip across the Tigris to Ctesiphon, destroying many of the bridges and taking any available boats with them.

Despite these measures and the fact that the river seems to have been in flood, the Persians failed to prevent the Muslims from crossing the Tigris. Taking advantage of local knowledge, Sa’d found a location where the river was fordable and sent a contingent of around 600 volunteers under Asim to force a crossing. These were intercepted by Persian cavalry, but Asim’s men were able to fight off this attack, establish themselves on the eastern shore and hold their position long enough for Sa’d to get reinforcements to them. With the Muslim army safely across the Tigris, the Sassanid force in Ctesiphon under Mihran and Rustam’s brother, Khurrazad, decided that any attempt to defend Ctesiphon itself was futile and prompted Yazdgerd to abandon the city with his army and treasury. With that, aside from small pockets of resistance, Sa’d and his Muslim Arab force took one of the ancient world’s greatest cities, along with the large amounts of booty it possessed, without a fight.

This lack of an organised defence of their capital not only demonstrates the poor state to which the Persian military had fallen through its defeats by Romans, Turks, civil war and now Muslim Arabs, but also how unprepared the Persian defences of Ctesiphon were for an attack from the south. Centuries of warfare against the Romans and the nomadic tribes of the Eurasian steppe had concentrated Persian defensive efforts to the north of Ctesiphon. The contrast between the destruction of the bridges over the Nahrawan canal to block Heraclius’ approach in 627 and the ease with which Sa’d approached VehArdashir and then took Ctesiphon in 637 demonstrates the direction in which Persian defences were facing. It could be argued that, by leaving troops in Mesopotamia to slow the advance of the Muslims on Ctesiphon, Yazdgerd assured the capture of his capital by depriving its defence of much needed manpower. However, without garrisons at the likes of Burs, Babylon and Kusa, Zuhra’s advanced guard would have arrived at Ctesiphon before any defensive measures were implemented. Therefore, after the defeat at Qadisiyyah, the Sassanid king and his generals were left with what was a no-win situation with regard to defending Ctesiphon.

However, this Persian evacuation of their capital without a fight meant that there were still sizeable Sassanid armies in the field that needed to be defeated before Muslim control of Mesopotamia could be consolidated. The main Persian force under Mihran and Khurrazad retreated north to Jalula, which, as well as being near the modern site of Baghdad, lay on a strategically important route between the Persian provinces of Mesopotamia, Khurasan and Atropatene. There were also forces congregating to the north at Birtha, usually identified with modern-day Tikrit, as well as the significant garrison of the fortress further up the Tigris recognised as modern Mosul. Its governor, Intaq, appears to have moved south to Birtha with his garrison and along with some survivors from Ctesiphon and new recruits from the local Arab tribes formed a sizeable force.

The relative proximity of Birtha to the main Sassanid force at Jalula meant that Intaq could move to join his forces to those of Mihran and Khurrazad as well as providing a potential route of retreat for the Persian force should it be defeated at Jalula. Therefore, whilst Sa’d sent the majority of his force against Jalula under Hashim in April, he also sent about 5,000 men under Abdullah to preoccupy if not neutralise Intaq. Upon arriving, Abdullah attempted to storm the walls with a lightning attack. However, Intaq’s men held firm and it appears as though Abdullah became concerned about the size of the garrison. To deal with this perceived strength, the Muslim commander attempted to drive a wedge between the elements of Intaq’s force. Muslim spies made contact with the Christian Arab contingent and persuaded them to side with Abdullah rather than Intaq. The Persians seem to have gotten wind of this betrayal or at least suspected it, as they attempted to abandon Birtha along the river. However, they found themselves trapped between the attacking Muslims and their former Arab allies and the Persian garrison was quickly overrun. A few days later, a small Muslim force received the surrender of Mosul without much of a fight.

While Abdullah was cutting off a potential route of retreat and reinforcement for the Sassanids, Hashim had squared up to the Persian forces at Jalula. While the strategic position of Jalula as a crossroads for the Sassanid state meant that it was vital for Mihran and Khurrazad to try to defend it, the position of the town with the Diyala River to the west and foothills of the Zagros Mountains to the east also offered an excellent defensive position. Knowing that the naturally narrow plain in front of Jalula would funnel the Muslim army towards the town and protect their flanks, Mihran prepared diligently for the Muslim attack he knew would come. Jalula itself was turned into a fort, protected by a line of trenches stretching from the broken ground of the Zagros foothills to the Diyala and caltrops to further hinder the Muslim infantry and cavalry. Archers and artillery were also positioned on the fortifications to bleed the Muslims as they approached the walls. Only after inflicting crippling damage on the Muslim ranks would Mihran then leave this defensive position in order to win a decisive victory.

Battle of Jalula, 637: Deployments.

Upon surveying the disposition and defences of the Persian force, Hashim recognised Mihran’s ploy in presenting the Muslims with only one offensive option – a costly frontal assault. This was something that he could ill afford given that the size of the forces arrayed at Jalula were likely very similar, around 12,000 each. Therefore, Hashim decided to draw the Persians away from their defences by employing one of the riskiest manoeuvres in battlefield tactics – the feigned retreat. The danger of this tactic is that a feigned retreat can quickly become an actual one if the morale and discipline of those attempting it is not strong enough and a counter-attack from the opponent is so well pressed and coordinated as to be impossible to resist. Clearly, after the numerous victories they had won up to the battlefield of Jalula, Hashim had every reason to believe in the discipline and prowess of his men to even attempt such a tactic. While there is no evidence to suggest that Mihran’s counter-attack was not well pressed, it could be argued that the presence of their own trenches and caltrops could have prevented the Persians from launching a fully coordinated assault on the `retreating’ Muslims as they had to waste time in placing a bridge over the defences.

The battle therefore began with a Muslim attack on the defences of Jalula, only for them to retreat under the hail of Persian archers and artillery. Mihran took this as a sign that his plan was working and that the Muslim forces were on the verge of breaking and quickly launched his planned counter-attack. Unbeknownst to the Persian commander, his opposite number will have also been pleased that his own plan was going well. His men had fooled the Persians into thinking they were retreating whilst still retaining their own discipline and order. With the Persians now drawn away from their defences, an infantry confrontation took place on the plain before Jalula. Further staged withdrawals by Hashim’s men then opened up a gap between the Persian lines and the bridge route back into the fort and it was then that Hashim launched his counterstroke. Having gathered together a strong cavalry contingent in his rear under Qaqa, Hashim now sent them in an attack around the Persian right flank against the lightly defended bridge. Once word filtered through the battlefront that the Muslims had cut off the only escape route, Hashim ordered his men in a full-scale attack on the Persian lines while Qaqa attacked their rear. Trapped by geography, their own defences and the Muslim forces, the Persian army broke. Despite many men making it back to the fort of Jalula, the defeat of Mihran and the death of Khurrazad had neutralised it as a threat. The exact date for the Battle of Jalula is difficult to pin down from the sources, some of which place the battle at the end of a seven-month siege while others say that the seven-month siege succeeded a battle in April 637.

Whatever the order of events, Jalula had fallen to Hashim by the end of 637. The Muslim general then sent Qaqa after those Persian forces under Mihran who had managed to escape. The cavalry commander caught up to them at the city of Khanaqin, some fifteen miles to the east. Some reinforcements from Hulwan may have reached Mihran but they were not enough to prevent a further defeat and the capture of Khanaqin. It is recorded that Qaqa defeated Mihran in a personal duel, removing one of the more capable Persian commanders as an obstacle. Qaqa was now within 100 miles of Yazdgerd III’s base at Hulwan and was to appear before its walls before the end of January 638. However, upon hearing of the defeat of Mihran at Khanaqin, Yazdgerd had retreated further east into the Iranian heartland of his empire, reaching Qom, around 100 miles south of modern Tehran. This hopping from Ctesiphon to Hulwan to Qom was to become a repeating pattern for the rest of Yazdgerd’s life as he attempted to outrun the Muslim advance whilst at the same time trying to bring together an army strong enough to retake his lost lands.

Battle of Jalula, 637: Qaqa’s Flank Attack.

With the emperor gone and only a modest garrison left to defend it, Hulwan also swiftly fell. Having settled affairs with the citizenry, the ever ambitious Qaqa then sent to his commander, Sa’d, asking if he could drive further into Iran in pursuit of the fleeing Yazdgerd. Sa’d himself appears to have been in favour of such an advance, perhaps thinking that the Persians were sure to return once they had reorganised their forces. However, Umar was unwilling to further stretch his forces given the effects of the `Year of Ashes’ and the Plague of Amwas throughout 638 and 639 and, as he had done in ordering his men to pull back from a potentially decisive confrontation in Roman Anatolia, he denied Qaqa and Sa’d permission to continue east. What is now the border between Iran and Iraq was to be the effective frontier between the lands of the caliph and those of the Persian emperor, albeit temporarily.

The loss of Mesopotamia, let alone their capital at Ctesiphon, was a huge blow, not just to the prestige of the Sassanids but perhaps more importantly to their continued ability to wage war, as those provinces contained a vast proportion of their population and tax revenue. The Persians still held significant territories all the way east to the Oxus and Indus rivers and their Roman neighbours had demonstrated that by identifying the strategic necessity of regrouping such losses were survivable. However, as will be seen, Yazdgerd and his advisers would not exhibit the same restraint and strategic good sense of Heraclius, allowing their loss of dignity to force them into challenging this `Iran-Iraq’ frontier before they had laid any defensive or infrastructural groundwork.

Command Structure of the Macedonian Army

The command structure of the Macedonian army was extremely complex, consisting of many separate layers of authority. At the highest levels it is quite well known; the same cannot be said of lesser ranks, but there are hints that suggest that, even at its lowest levels, it was as complex as the more powerful positions. As with many areas of Alexander’s empire, and particularly within the army, the command structure was continually evolving as new positions were created and others became obsolete. The most significant changes, however, were probably politically motivated. Alexander gradually changed the army from being that of Philip, through the influence of Parmenio and his family, to being his own, particularly after 331/0, when Parmenio’s influence had been removed.

Macedonian Heavy Infantry

At its lowest levels the command structure of the heavy infantry can be deduced from its gradations of pay. The smallest tactical unit of the heavy infantry was the dekas or ‘file’. As the name implies the dekas had once consisted of ten men, but at some point long before the reign of Alexander it had been expanded to sixteen. Of these sixteen men, twelve were rank and file with the other four being of superior status. Of these four, one was the dekadarch or file leader, one was a dimoirites or half-file leader and the other two were dekastateroi or half-file closers. Arrian tells us that the dekastateroi were paid the equivalent of one and a half the pay of a rank-and-file soldier, around forty-five drachmas a month. The dimoirites received double pay of around sixty drachmas a month. Bosworth seems to have made a slight error in interpreting this passage in Arrian, claiming that there were two dimoiritai and only one dekastateros, but Arrian’s text seems quite clear on this point.

Thirty-two decades formed a lochos, consisting of 512 men and being commanded by a lochagos. Three lochoi formed a taxis, which was the fundamental unit of the Macedonian heavy infantry, commanded by a taxiarch. Each taxis therefore consisted of 1,540 men, of whom 1,152 were rank and file and in receipt of the basic one drachma a day. Initially Alexander crossed the Hellespont with six taxeis, later expanded to seven around the time of the invasion of India. Therefore the command structure for a typical taxis of heavy infantry was:

  • Taxiarch
  • Lochagos (x3)
  • Dekadarch (x96)
  • Dimoirites (x96)
  • Dekastateros (x192)
  • Rank and File (x1,152)

The manpower indicated is, of course, paper strength, assuming that each taxis was at full strength. The six taxiarchs appear to have all been of the same rank with none holding superiority. Indeed there was no overall commander of the heavy infantry (excluding the one possible reference in Arrian discussed earlier) as there was for, say, the Companion Cavalry. This is because there really was no such organization as the ‘Macedonian phalanx’, the taxeis themselves frequently being used as separate tactical units, or in groups of two or three. This development came largely after 331 when the army entered northeastern Iran and smaller, more mobile forces were required.

Arrian makes a seemingly strange claim that:

On the right wing of the attacking force Alexander had the guards’ division under his personal command. In touch with them were the infantry battalions, forming the whole centre of the line and commanded by the various officers whose turn of duty happened to fall upon that day.

This is almost certainly evidence of a rotational system within the phalanx. It could be a reference to the order in which the taxeis appeared each day, and we are explicitly told in the sources that that the actual order of each taxis did rotate each day, or it could be that the minor commands within a taxis were rotated to give junior commanders more experience of slightly different roles.

The Macedonian heavy infantry appear to have undergone very few serious changes in the command structure over the course of the campaign. The huge numbers of reinforcements received between the great set-piece battles of Issus and Gaugamela seem to have been incorporated within the existing taxeis, presumably adding to the numbers of rank and file rather than to the officer corps. The first evidence for a seventh taxis does not appear until the time of the invasion of India, where Arrian names seven taxiarchs operating simultaneously.


At the time of the invasion of Persia, another of Parmenio’s sons, Nicanor, was the commander of the hypaspists. The hypaspists were the elite formation of the Macedonian heavy infantry and their tactical and strategic roles were many and varied. The hypaspists were organized into three chiliarchies of 1,000 men, each commanded by a chiliarch. This is true after 331 at least: before 331 it is uncertain. One of these chiliarchies was designated the agema, perhaps commanded by Alexander himself, or more likely by an unknown individual, as Alexander was usually with the Companion Cavalry during set-piece battles. The chiliarchs themselves were of markedly lower status than the taxiarchs of the heavy infantry, being more like a lochagos. This is at first sight rather surprising, considering that the hypaspists were the elite units of the heavy infantry, receiving only the very best of the new recruits into their ranks. We should remember, however, that, unlike the heavy infantry, the hypaspists had an overall commander, Nicanor, who was at a significant level within the command structure, ensuring that their status was considerably higher than that of an infantry taxis.

As with the rest of the army, the command structure of the hypaspists was significantly changed at the end of 331. The chiliarchies were subdivided into two new units, pentakosiarchies, thus adding an entirely new layer into the command structure, albeit a very lowly one. These new officers were again appointed by Alexander on the basis of merit rather than seniority, and again owed their allegiance to the king himself. We saw earlier how important this innovation was to the king, particularly later in his career as he became ever more paranoid.

Companion Cavalry

The numbers of the Companion Cavalry are not certain, although Diodorus gives the figure of 1,800 and most now accept the figure of 1,800 as at least being very close to the actual figure. We do know that by 333 the Companion Cavalry consisted of eight ilai (squadrons) of 200 men, each commanded by an ilarch. We have no information of from the beginning of his reign, but we do know that along with the general reorganization of the army in 331 an ile was further divided into smaller units. Arrian tells us ‘He also – this was an innovation – formed two lochoi (companies) in each cavalry squadron’.

Curtius hints at a reorganization, and confirms that commanders would from this time be promoted on the basis of ability and not of regional affiliations. However he does not mention the division of ile, preferring to concentrate on the hypaspists. From 331 the subdivision of ile was into two hekatostyes of 100 men; there is evidence that each ile was again subdivided into tetrarchia, of which there would have been four per ile. The tetrarchia is only recorded once in Arrian, at the turning of the Persian Gates in January 330, not before or after, but the unit was small enough that this is not surprising as they would mostly have operated in the larger units.

One of the ilai was given the title ile basilike, also frequently called the agema, or royal squadron: this unit was also of higher status than the rest. The royal squadron was of double strength and was charged with defending the king when he fought on horseback; they were his personal bodyguard in all of the set-piece battles. The overall command of the Companion Cavalry was in the hands of Philotas – until his execution in October 330, that is.

The ilarchs seem to have been relatively minor in rank, probably on a par with an infantry lochagos. Ilarchs are seldom mentioned by name in any of the sources and are never given separate commands of their own. The only one that achieved any level of distinction was Cleitus the Black, the commander of the royal squadron.

After the execution of Philotas in 330, the entire Macedonian cavalry was reorganized en masse. The basic tactical formation was now not the ile but the ‘hipparchy’. These new units are first recorded by Arrian during 329. Ilai do still appear in the sources but they become sub-divisions of a hipparchy, each hipparchy comprising a minimum of two ilai and thus 400 men. The ilai were also sub-divided into two lochoi, the commanders of whom were given the title lochagos, as with the commander of an infantry unit. Alexander appointed these commanders personally on a basis of merit rather than superiority, thus breaking with tradition. This was a policy that was entirely consistent with Alexander making the army loyal to him alone. This began to occur some time during 331 BC, probably after Gaugamela, when the last great batch of reinforcements arrived from Macedon. This was the beginning of the policy, that I have noted a number of times, of reducing the army’s ties of loyalty to its individual commanders, ultimately making them loyal to Alexander alone. Thus a new layer of sub-commanders was added in the command structure of the army, one which owed its loyalty directly to Alexander, to some extent breaking the link between the troops and their commanders. There are two possible reasons for this change. Perhaps Alexander came to the conclusion that the ilai were simply too small, at 200 men, to cope with the different style of fighting in the entirely different theatre that was to be their next challenge. The army would no longer be fighting set-piece battles, but would require far more mobility and flexibility as it moved into the northeast of Iran The second possible explanation was a desire on the part of Alexander to increase the relative superiority of the Companion Cavalry over the heavy infantry, each hipparch now being of a higher status than a lochagos.

The term ‘ile basilike’ also disappeared at this time and was replaced by the term agema, the nomenclature becoming the same as for the hypaspists. The actual number of cavalry hipparchies is unknown, but it is assumed that there were eight during the Indian campaign. The position left vacant by the death of Philotas was not directly filled. He was instead replaced by two men: Alexander’s life-long friend Hephaestion and Cleitus the Black. Both men were effectively of equal status within the command structure. Arrian gives the reason for this step as that ‘he [Alexander] did not think it advisable that one man – even a personal friend – should have control of so large a body of cavalry’.

Allies and Mercenaries

The Thessalian cavalry were without doubt the most important contingent of this aspect of the army. They were probably equal in number to the Companion Cavalry and very close to them in terms of quality. Overall command of this vitally-important unit was given to Alexander’s second-in-command, Parmenio. The command structure of the Thessalian cavalry was very similar to that of the Companions, being divided into ilai. They were not, however, allowed their own national commanders, but a senior Macedonian officer was appointed to command each unit. The Thessalian cavalry also had a unit which performed the same role as the royal squadron of the Companions; this was known as the ‘Pharsalian contingent’.

The other allied cavalry contingents, although considerably less important, were again organized along similar lines, being divided into ilai and each having a Macedonian commander. The appointment of a Macedonian commander at the head of non-Macedonian troops, be they cavalry or infantry, was the general policy of Alexander throughout his reign. Even the mercenary contingents were treated in exactly this same fashion, Menander being their overall commander. These Macedonian officers, however, were relatively unimportant in the overall command structure and few achieved any kind of distinction.

The fleets that accompanied the army of invasion were almost exclusively non-Macedonian, being provided by the member states of the League of Corinth. Each ship was captained by a native of the contributing city, and where a city-state provided more than one ship, they also supplied a ‘commodore’ for their particular contingent. As with other non-Macedonian units however, overall command of the fleet was with a Macedonian officer.


The term ‘bodyguard’ is quite a confusing one, as there appear to be two entirely separate groups within the army that carry this title. The first is an apparently-quite-strong detachment of heavy infantry. Arrian, three times, tells us that Alexander took with him the bodyguards and some of the hypaspists; this would seem to strongly suggest that they were not simply a detachment of the hypaspists, who were themselves often called ‘the guards’. Diodorus tells us that at the Battle of Gaugamela, Hephaestion ‘had commanded the bodyguards’. This passage again strongly suggests that we are not here talking about a detachment of the hypaspists, as at this time Nicanor was still their commander and only died later that same year. The bodyguards seem to have been a relatively minor force, perhaps of the order of a couple of hundred strong. The relative position of their commander within the overall command structure of the army is unknown; the only commander named is Hephaestion at Gaugamela, who was of course a very senior figure. Hephaestion’s seniority probably had more to do with his closeness to Alexander than the importance of the bodyguards as a military force: his successor after Gaugamela is never mentioned, for instance. This group could well represent a carry-over from a much older organization that pre-dated Philip’s reforms.

The group that most interests us here are the somatophylakes basilikoi, or ‘royal bodyguard’, originally seven strong, this number being rigidly maintained. The number was probably connected to their historical function of guarding the king’s tent. They were increased to eight in India, however, when Peucestas was promoted to this rank as a sign of gratitude by Alexander for saving his life during the attack on the capital city of the Malli.

The bodyguards occupied a position within the command structure that is difficult to define. The group as a whole formed part of Alexander’s immediate entourage, and seem certain to have been among his closest friends and most-trusted advisors. Membership of the bodyguard was obviously incompatible with any post that involved their being away from court for any length of time: both Balacrus and Menes were replaced as soon as they were assigned to the command of provinces. For reasons that seem less clear, inclusion within the bodyguard was also incompatible with a command within the army. Before Gaugamela, there is no recorded instance of a member of the bodyguard simultaneously holding a senior command. Bodyguards are occasionally reported briefly holding minor commands, such as Ptolemy, who commanded a joint force of hypaspists and light infantry during the siege of Halicarnassus, but this was rare. If any bodyguard were promoted to a senior command, he would immediately lose his title and be replaced. This happened, for instance, when Ptolemy became a taxiarch (this was not the famous historian and son of Lagos, but another very obscure individual: Ptolemy was a very common name in Macedonia). As a group they probably enjoyed the same status as a taxiarch, but did not, as such, occupy any position within the command structure. They were, however, still influential as they were among the king’s closest advisors.

This rather rigid system which applied to the bodyguard, as with almost everything else in the army, evolved considerably over time. After the death of Parmenio we hear of instances of bodyguards receiving senior, albeit temporary, commands. In 328, for instance, Alexander left four taxeis of heavy infantry in Bactria, along with their commanders. The remainder of the army was then divided into five columns, three of which were commanded by known bodyguards. The deaths of Parmenio and Philotas represent something of a watershed in Alexander’s career, as will be discussed below.

Evolution of the Command Structure

One of the major changes that occurred in the command structure over the course of Alexander’s reign was that the cavalry commands became increasingly important, relative to their previously-equivalent infantry commands. By the time of the death of Philotas Alexander was becoming increasingly disinclined to place such large numbers of men under a single commander; with this in mind he divided the command of the Companion cavalry between Hephaestion and Cleitus the Black. Individual hipparchs also became increasingly important in their own right, becoming roughly equal in status to the position of an infantry taxiarch. At the beginning of the invasion of India, the commanders of the heavy infantry who were most highly favoured by Alexander were promoted to command hipparchies of Companion Cavalry; namely Perdiccas, Craterus and Cleitus the White. This again illustrated that moving from an infantry command to that of cavalry was considered a promotion.

During this process, the royal bodyguard evolved into an important position within the command structure. Perdiccas was promoted to a hipparchy, from a taxis of heavy infantry, in 327; by 330 he also had the title of bodyguard. This was a dual function which was also enjoyed by Alexander’s favourite, Hephaestion. The Peithon who was a bodyguard by 325 is very probably the same Peithon who is attested as a taxiarch in 326/5. An infantry command was rare, however; members of the bodyguard were usually given commands within the Companion Cavalry, in alignment with its increasing importance. This desire on the part of Alexander to systematically downgrade the relative importance of the heavy infantry can perhaps be attributed to the fact that Alexander saw them as a potential, and increasing, problem. It was from the ranks of the infantry that the mutinies at the Hyphasis and Opis had come, and it would not be surprising if Alexander had deliberately aimed at increasing the prestige and importance of the cavalry relative to the infantry. This would have been to re-establish the situation that had existed before his reign, where the cavalry were clearly the units of greatest prestige. It is also likely that the heavy infantry had become less prestigious, simply because the army was engaged in northeastern Iran at this time and they were not as heavily involved in the fighting as they had been during earlier campaigns.

From 330, when Alexander entered the northeast of the former Persian Empire, he was faced with an entirely new situation: that of guerrilla warfare. This led to a willingness on the part of Alexander to divide his force, seemingly indiscriminately, between various commanders. Before this time if a second column was required it would consist of allied and mercenary troops, the Macedonians almost always staying with the king. As mentioned above, in 328 Alexander left four taxeis of Macedonian heavy infantry in Bactria and divided the rest of the army into five groups. These new commands were given to a fairly select group of Alexander’s closest friends: Craterus, Hephaestion, Coenus and Perdiccas were usually the first choices, with Ptolemy, Leonnatus and Peithon used where more columns were employed. When Alexander entered India, Hephaestion and Perdiccas were sent ahead to the Indus with a large force comprising around half of the Macedonians and all of the mercenary infantry.

One of the most important features of the changes in the command structure of the Macedonian army towards the end of Alexander’s reign was the increasing mobility of commands. Individual generals still kept their titles, but were expected to command entirely separate units as situations presented themselves. For example, in 327 three taxiarchs, Meleager, Attalus and Gorgias, were detached from their taxeis and were given the commands of a group of mercenary cavalry and infantry. They were then employed on diversionary movements along the river banks. Another example is that of Coenus, a taxiarch since 334, who was employed as a cavalry commander at the Hydaspes.

This move towards an increasing mobility of command was for two main reasons, the first being military. As Alexander entered the next phase of the campaign after 331, he increasingly met with an enemy that operated on significantly different lines from that faced early in the campaign. He was also faced with fighting in a new theatre and in different conditions, all of which required the army to be considerably more flexible than it had previously been.

There is surely a second, and in my opinion significantly more important factor at work here: namely politics. Alexander seems to have been becoming increasingly concerned about assigning large bodies of troops to a single commander indefinitely: there was for instance probably no overall commander of the heavy infantry, and the positions vacated by Parmenio and Philotas were never filled, the Companion Cavalry receiving co-commanders. Alexander increasingly detached individuals from their commands and gave them different assignments. He also employed new layers in the command structure and made promotions according to merit. These changes had a two-fold effect: the commanders became loyal primarily to him, since they owed their positions directly to the king’s favour; secondly, the focus of the army s loyalty was also the king, as their commanders often changed and their territorial origins slowly eroded. Alexander made himself the sole focus of every individual, whatever his rank, within the army.

The Price of Parmenio’s Support

Parmenio was probably the single most important political figure in Macedonia, apart from the king, during the reign of Philip; this is true of the early part of Alexander’s reign too. He, as well as various of his family members, was well entrenched at court and seems to have had political connections with both factions contending for the succession in the last years of Philip’s reign. Thus when Philip was assassinated, Parmenio was in a prime position to act as king-maker. He was in a position to offer the support of most of the lowland barons; this would leave Amyntas or any other potential rival with only the possibility of forming a coalition of the fringes of Macedonia and rebellious Greek cities. Parmenio was evidently a skilled political operator and knew well the strength of his position; Alexander was forced to pay a heavy price for Parmenio’s support, but in 336 he was in no position to argue. When the Macedonian army crossed the Hellespont into Asia almost every key command was held by one of Parmenio’s sons, brothers, or some other kinsman. We have already noted that two of Parmenio’s sons were commanders of the hypaspists (Nicanor) and the Companion Cavalry (Philotas), with Parmenio himself commanding the Thessalian cavalry and essentially being second in command of the whole army. Parmenio’s brother, Asander, probably commanded the light cavalry, and certainly received the satrapy of Sardis as soon as it was conquered. Parmenio’s supporters were also firmly entrenched in positions of power, men like the four sons of Andromenes and the brothers Coenus and Cleander. Many of the commanders of the army of invasion were little younger than Parmenio himself: when Justin tells us that headquarters looked ‘more like the senate of some old-time republic’ he is probably not exaggerating in his description, although it is a far from flattering one.

The Macedonian army down to 330 was at its very core Philip’s; they were his veterans and his commanders. Philip’s influence was always present, and frequently felt by Alexander in the form of Parmenio. This was a situation which Alexander could not tolerate indefinitely. He allowed the command structure to remain relatively unchanged whilst his success was still in the balance, but after Gaugamela Alexander began to make serious and sweeping changes to the army, changes which were made considerably easier by the assassinations of both Philotas and Parmenio. Some authorities have argued that Alexander was plotting for the best part of six years to remove Parmenio’s grip on the army, and see the final execution of Parmenio and Philotas as being the culmination of this plotting; this seems unlikely to be true. Why would he, for example, have left Parmenio in Ecbatana with a considerable part of the army and his treasury if he did not trust his loyalty, or if he was about to act against him? Alexander, on the whole, seems to have been more impulsive and spontaneous than this theory would give him credit for: it is more likely that Alexander seized his opportunity without having engineered it. Alexander was gambling that the army loved him more than they loved the old general, and he was right. Although the Thessalian cavalry perhaps did not take it well, they were not a significant part of the army after this, and indeed were disbanded entirely soon after. Following the death of Parmenio, Alexander would never again allow large bodies of troops to be commanded by any one individual for any length of time. Dissention was met with ruthlessness; the army had at last become his and his alone.