Cannae

by Adrian Goldsworthy

  Carthage maintained a much clearer divide between war and politics than Rome, and it was very rare for a serving magistrate ever to be given a military command. Punic generals were appointed and frequently served for many years with little interference or supervision of their actions from the civil authorities. As a result many were more experienced than their Roman counterparts, who were appointed annually. However, we should not exaggerate the difference, since Punic commanders were drawn from the ranks of the same aristocratic families who dominated politics. Military appointments appear to have owed more to wealth and influence than impartial assessment of military ability. Some Carthaginian generals were very able men, but, in spite of their longer commands, the majority did not prove themselves markedly superior to their Roman counterparts.

  We do not have a detailed breakdown of Hannibal’s army at the beginning of the expedition to Italy. The troops left behind in Spain or sent to Africa to guard against Roman attacks consisted of infantry and cavalry from at least five Spanish tribes, Balearic slingers, small numbers of Libyan cavalry and a considerable force of Libyan infantry, Liby-Phoenician (a people of mixed Punic and African stock who enjoyed limited rights) cavalry, Numidian horsemen from at least four tribes, and a small band of Ligurians from Northern Italy. The authority for the breakdown of these forces was a pillar erected in Italy on Hannibal’s orders, which makes it all the more frustrating that less information has been preserved about his own army. When he arrived in Italy, Hannibal had sizeable contingents of Libyan foot, Numidian horse and Spanish cavalry and infantry, as well as a number of Balearic slingers, supported by war elephants (although all of the latter had perished before Cannae). He was soon joined by strong contingents from his new-found Gallic allies, who came to supply almost half of his field army. It is possible that Hannibal also had small contingents from some of the other ethnic groups, but if so their numbers were not great.13

  The Libyan foot were the most reliable element in the army. Most fought in close order, although it seems likely that Libyans were also included amongst the lonchophoroi, Hannibal’s specialist javelin skirmishers. (Many translations of Polybius render this inappropriately as ‘pikemen’.) The heavy infantry began the war dressed in a version of the standard equipment of Hellenistic infantry. They wore bronze helmets and body armour, probably made from stiffened linen, carried large round shields and probably fought with spears. In 217 BC Hannibal re-equipped them with the spoils of the Roman dead at Trebia and Trasimene. It is not clear whether this means that he gave them only Roman defensive armour of helmet, mail cuirass, and oval scutum, or whether they also adopted the pilum and gladius. The Libyans were well disciplined and drilled, capable of complex manoeuvres, and in every respect the equals, and sometimes the superiors, of any Roman legionaries. If there were any Libyan or Liby-Phoenician cavalry with Hannibal’s army, then they would have fought in close formation and carried Hellenistic-style equipment, not too dissimilar from Roman cavalry.14

  The other African contingent was provided by the Numidians, most of whom fought as light cavalry. These men rode small, agile horses without saddle or bridle, wore a simple tunic and had only a small round shield for protection. Their tactics emphasized swift movement and avoidance of actual contact, sweeping in to throw javelins and then retreating before the enemy could close. At the beginning of the war the Romans were unprepared for these tactics and had great difficulty coping with them. The Numidians were linked by language and culture but divided into fiercely independent tribes, themselves often wracked by civil wars as rival members of their royal families fought for power. However, the bond between Numidian royalty and Carthaginian aristocracy was often close; Hamilcar Barca was one of several Punic noblemen to form marriage alliances with princes from the various tribes. Such a means of cementing alliances was employed elsewhere by Punic noblemen, both Hasdrubal and Hannibal himself marrying Spanish princesses.15

  The Spanish contingents provided close order cavalry, and both heavy and light infantry. The close order foot were known by the Romans as scutati, and carried a flat oval body shield. Some were also equipped with heavy javelin similar in size and effectiveness to the pilum. The skirmishers, or caetrati, carried a much smaller, round shield and a bundle of javelins. Foot and horse alike were also armed with high quality swords, either the short thrusting pattern copied by the Romans or the curved falcata, shaped rather like a Gurkha’s kukri. The falcata is more prominent than the straight-bladed sword in the archaeological record, but this is most probably due to its use in ritual and sacrifice and it may in fact have been less common. The normal Spanish costume was a white or off-white tunic with a purple border, but there was probably considerable individual and regional variation. The slingers from the Balearic Islands were renowned for their skill and savagery in the ancient world. The combination of these men armed with longer-ranged weapons and well-trained and highly motivated javelinmen gave Hannibal a marked superiority over the Romans in light infantry actions.

  The Gallic tribes provided Hannibal with both horse and foot. The cavalry fought in massed formation like the Spanish and were certainly using the four-horned saddle, which may in fact have been a Celtic invention. Helmets and especially armour were very rare amongst both Gallic and Spanish warriors, usually restricted to chieftains. For most warriors their sole defensive armour was provided by an oval or rectangular shield. The Gauls employed a variety of spears and javelins for throwing and/or hand-to-hand combat. Those Gauls rich enough to afford a sword tended to use a much longer, heavier weapon than the Spanish. Their blades might be 90cm (3 feet) or even more in length and frequently lacked a point. Instead the warriors relied upon the edge, slashing at their opponent and relying on strength more than finesse. We do not know whether some Gallic warriors fought as light infantry. Little prestige was associated with this type of fighting in tribal warfare and it seems to have been left to the poor or the young, but it is possible that some skirmishers were present with Hannibal’s army.

  We do not know the size of the basic tactical units in Carthaginian armies. It may be that the more regular elements, notably the Libyan foot, were formed into units of uniform size. However, it is probable that the other peoples provided war bands varying considerably in numbers. There are a few references to groups of 500 which may have been single units, and another to 2,000 Gauls divided into three bands, although in this case it is not clear whether these were in turn subdivided into smaller groups. Probably the basic units in the army consisted of a few hundred men, but some may have been as small as Roman maniples and others significantly larger.16

  There really was no such thing as a typical Carthaginian army, since these varied tremendously in their ethnic composition, mixture of troop types and general effectiveness. Each general had to develop a system of controlling and co-ordinating the movements of the diverse elements within his army, a process which took time. Hannibal’s army in the early years of the Italian expedition was the finest fighting force ever put into the field by the Carthaginian state. Its solid nucleus was provided by the men who had fought under Hamilcar, Hasdrubal and Hannibal’s own command in Spain. These men were tough, experienced and highly disciplined. They knew and trusted their officers at all levels, and were personally loyal to the Barcid family who had given them victories and rich rewards in former campaigns. Most of the senior officers, who included Hannibal’s younger brother Mago, were used to working under their young general and knew what he expected of them. The quality of Hannibal’s senior subordinates was markedly higher than in the Roman army at Cannae.

  Around this core of veterans, Hannibal was able to incorporate the Gallic warriors who joined him in Italy without any loss of the army’s tactical flexibility. It took time for the Gallic contingent to become as reliable as the other troops, but this process of absorption had largely been completed by the date of Cannae. At Trebia the Gallic infantry were stationed in one sector of the line and appear to have fought under their own leaders just as
if this were a tribal army. In the next year our sources noted the poor march discipline and low stamina of the Gauls, who were unused to such rigorous campaigning. By the time of Cannae small units of Gauls were interspersed with Spaniards, suggesting that the tribal structure had been substantially replaced by the same system of command which directed the rest of the army to perform the will of its Carthaginian officers. Still later, at the capture of Tarentum in 212, Hannibal made use of bands of Gauls specifically chosen for their speed and discipline, showing the extent to which these tribesmen had grown in efficiency during their long service. A similar process had probably occurred with most of the Spanish troops before 218; training, experience and discipline adding to the fierce bravery and individual skill at arms of these men drawn from warrior societies to produce highly effective soldiers.17

  At Cannae Hannibal’s soldiers formed a more united and cohesive force than their Roman opponents in spite of their much greater mix of languages and cultures. The Roman army in 216 was a very mixed bag, unused to working together and unfamiliar with many of its officers, a theme which we shall explore in more detail in the next chapter. Hannibal’s army also possessed a much better balance between the different troop types. The Punic army had one cavalryman to every four infantrymen, compared to the Romans’ ratio of one to thirteen. The Roman cavalry were significantly outnumbered and generally lacking in confidence. They also lacked the flexibility of Hannibal’s men who included both light and heavy types. On a man to man basis the Carthaginian soldiers at Cannae were at least the equal of their Roman equivalents, and in the case of the cavalry and light infantry markedly superior. The real superiority of the Punic army came at higher levels, where its cohesion and efficient command structure gave it far greater flexibility in grand tactics and manoeuvre. The problem for the Romans in 216 was how to overcome this.

  THREE

  The Campaign of 216 BC

  FABIUS’ DICTATORSHIP HAD given the Republic some respite after the disasters at Trebia and Lake Trasimene. A Roman field army had been rebuilt and had gained a little confidence through winning a few skirmishes with the enemy. The Romans had also enjoyed some success in Spain. Yet Hannibal, the man whose actions had provoked the conflict, was now spending his second winter in Italy at the head of an army undefeated in any serious engagement. Since 218 he had beaten two armies comprehensively, and destroyed or severely mauled several smaller forces, inflicting heavier losses than the Romans had suffered since the First Punic War. In many ways these disasters were worse, for they had occurred in the fields of Italy rather than on some distant sea, and the losses had fallen heavily on the wealthier classes who served in the legions instead of the poor and politically less significant men who crewed Rome’s fleets. The army, the source of Rome’s military pride, had proved incapable of preventing an invader from marching where he willed, burning and despoiling the Italian countryside with impunity. This was a massive humiliation for any state, and a terrible admission of weakness. However well Roman forces did in Spain or Sicily, this was scant consolation for the continued presence of an undefeated Hannibal in Italy itself. Over the winter of 217–216 BC the Senate prepared to mobilize a large part of the Republic’s resources to mount a massive war effort in the coming year. The other theatres, Spain, Sicily, and Cisalpine Gaul, were not ignored, but the main concern was always to be Hannibal.

  THE LEADERS

  In 218 and 217 the Senate had intended both consuls to join forces and defeat Hannibal, but Scipio had been wounded before Trebia and Flaminius killed and his army destroyed before Geminus’ legions had reached them. This time the two new consuls would set out from Rome together, and the Republic’s most senior magistrates hold joint command from the very beginning of the campaign. These men, Lucius Aemilius Paullus and Caius Terentius Varro, were elected to lead the largest army Rome had ever fielded to the anticipated great victory. In spring 216 it is doubtful that anyone could have guessed at the scale of the subsequent disaster, but all our sources were written with the benefit of hindsight. This makes it extremely difficult to separate myth and propaganda from truth, and so gain some genuine insight into the characters of these men.

  Aemilius Paullus was the grandfather of Scipio Aemilianus, Polybius’ patron, and as a result receives very favourable treatment from the Greek historian. In Polybius’ account it was he who made the greatest efforts in organizing and encouraging the newly raised legions. Varro is mentioned as Paullus’ colleague, but then does not appear again until the consuls have led the army to within sight of Hannibal. At this point a dispute broke out between the two Roman commanders over what to do next, Paullus advocating a far more sensible course than the less experienced Varro. It is only really at this stage and in the ensuing battle that Polybius depicts Varro as an incompetent general. Paullus had already held the consulship in 219 BC, successfully campaigning in Illyria along with his colleague, Marcus Livius Salinator. This victory had been marred by scandal involving the distribution of booty, and although the chief blame had fallen on Salinator, Paullus had not emerged entirely unscathed. This seems to have made him especially eager to avoid criticism in his second consulship. The Illyrian War had involved combined operations between the fleet and army as the Romans operated along the Adriatic coast, but there had been no pitched battles. Command in such a conflict certainly made great demands on a general, but it should be noted that the skills required were not precisely the same as those needed to control a massive field army.1

  In Livy’s account, and the narratives of all later authors, Varro appears from the beginning as a dangerous fool. He had been quaestor in 222, was aedile in 221 and the praetor in 218, but is first mentioned by Livy as allegedly the only senator to support the bill granting Minucius equal power to the dictator in 217. Varro was a ‘new man’ (novus homo), one of that small number in any generation of Roman politics who were the first in their family to reach high office. Livy dismissively describes Varro’s ancestry as ‘not merely humble, but sordid’. He repeats, without saying whether or not he believed it, a claim that Varro’s father was a butcher on a small scale, and that as a youth he worked in the business. This suggestion is rarely given any credence by modern historians, since it was typical of the vulgar abuse which was the common coin of political debate at Rome. The minimum property qualification required for membership of Rome’s highest census rating, and thus the Senate, would also mean that, even if Varro’s father had begun in business on a small scale, he had built up a considerable fortune by the time that the son inherited and embarked on a political career. The claim that Varro established a reputation in the courts by winning cases on behalf of dubious clients is again a fairly conventional accusation to level at a rival politician. Livy’s Varro, like his Flaminius and Minucius, conforms to a theme running throughout his History which held rabble-rousing politicians responsible for most of the ills to befall the Republic, but even his own account of the election in 216 BC suggests that things were much more complicated than this.2

  Livy says that when the first ballot was held only Varro was elected. He implies that the five unsuccessful candidates were far more distinguished and responsible, but it should be noted that none of these men ever seem to have gained high office, although it is possible that this was because some or all were killed during the war, perhaps even at Cannae. As a result of his success, Varro presided over a second election which chose Aemilius Paullus as his colleague. Normally the presiding magistrate could do much to influence the outcome of a ballot, but it is unclear to what extent Paullus could be considered Varro’s choice as partner, since we are told that all the other candidates withdrew. It has been speculated that there was some connection between the two, even that Varro may have served under Paullus in Illyria, but this must remain conjectural. A ‘new man’ needed considerable political ability to win elections against rivals with famous names and many clients who would vote for them. It was rare for candidates to advocate specific policies, since the electorate was
more concerned with the individual’s virtues and character, but a ‘new man’ could not parade the achievements and quality of his ancestors and needed some other way of fixing his name in the public eye. Discontentment with Fabius Maximus’ cautious leadership in 217 presented Varro with a popular cause which gave a boost to his election campaign. He may well have exaggerated his own aggressiveness and the perceived faults of the well established senatorial families as military leaders, and so by implication what could be expected from some of the rival candidates, for ‘new men’ tended to overemphasize their own attributes in an effort to compete. Latching onto a popular cause in this way was risky, but a novus homo often needed to take risks if he was to succeed. Finally, it is vital to remember that the voting system at Rome heavily favoured the wealthier classes. Varro’s supporters came not primarily from the landless poor, a group always seen as foolish and fickle in our ancient sources, but from all levels of Roman society. To have achieved such overwhelming success in the first ballot, he must have had the support of many, perhaps even the majority, of the wealthy equestrian order and probably also the Senate. The desire to confront Hannibal, and thus to choose an aggressive commander who would do this, was widespread at all levels in Rome.3