The main developments in military theory and practice before the Punic Wars had all occurred in the Greek world. Greek city states had first developed the hoplite phalanx, a dense mass of heavily armoured infantrymen which advanced and fought their way through anything in their path. On level ground such Greek spearmen had proved superior to any other type of soldier down to the earlier fourth century. It was a system of fighting ideally suited to the Greek soldier-farmers, who wished to resolve a campaign quickly so that they could return to working their farms. It required little technical skill or training, for which Greek citizens apart from the Spartans had little time, but needed considerable courage and group solidarity, things which the hoplites of the city states possessed in abundance. Tactics were simple, especially when two similar phalanxes confronted each other in a war between rival cities, and such subtleties as reserves were virtually unknown. Many wars lasted a matter of weeks and were resolved by a single day's clash on one of the few level plains in the Greek Peninsula. Greek warfare developed as society changed and in the late fifth and fourth centuries professional soldiers appeared in increasing numbers, campaigns tended to last much longer and not be so closely tied to the agricultural year, whilst generalship and tactics grew in importance. Philip II and Alexander the Great of Macedon led well-trained and drilled professional soldiers, forming armies that included heavy and light cavalry, and light infantry as well as the heavy infantry of the phalanx, who were now armed with two-handed pikes instead of the spears. It was with such an army that Alexander swept across the Near East and on into India in little more than a decade. The Hellenistic military system had proved itself superior to anything else in the world at that time, but after Alexander's death and the break-up of his empire into a number of bickering Successor Kingdoms Macedonian style armies most often found themselves facing another similar force. Where both sides employed the same tactical system and equipment, a decisive victory was much harder to achieve. As a result armies began to experiment with all sorts of unusual weapons, such as scythed chariots, war elephants and heavily armoured cataphract cavalry, in an effort to gain some advantage over the enemy. More emphasis was also placed on the role of the commander, who tried to force a battle on terms most favourable to his own army and avoid contact if the odds were against him.
Neither the Romans nor the Carthaginians possessed a modern army based on the Hellenistic model, but the campaigns between them were to be fought largely in the manner of contemporary Hellenistic warfare. The most important and decisive element in warfare remained the pitched battle, although raiding and sieges now played a far more significant role than had ever been the case with hoplite warfare. A clear victory in a massed battle was the best way of putting pressure on the enemy, but there was always the possibility of defeat and a battle was not something to be risked lightly. Heavy casualties were difficult to replace quickly, since both Carthaginian mercenaries and Roman levies needed time to be turned into a battle-worthy force. Even if the majority of a defeated army survived an engagement, their morale was severely reduced and made it unlikely that they could face the same enemy again with any prospect of success until some time had passed. Battles were rarely if ever fought for any strategic purpose greater than destroying the enemy field army. Therefore a good commander sought battle when he felt that there was a good chance of victory and avoided a confrontation if there was not.
The area covered by the campaigns of the Punic Wars was very large, whilst the armies involved were relatively small. With long-range strategic intelligence usually poor and sometimes non-existent, it was rare for either side to have a clear idea of the enemy's location until the two armies neared each other. Armies in this period tended to march carelessly, moving as quickly as possible towards the anticipated campaigning area, only becoming cautious when the opposing army was located within a few days' march. The march of a large army was difficult to conceal from scouting parties, the cloud of dust thrown up by tens of thousands of feet and hoofs being visible for many miles. It was normally expected that the presence of a hostile army would be recognized before it came close enough to pose a direct threat. Roman armies in particular tended to have a casual attitude towards reconnaissance, in part because their aristocratic cavalry had little taste for the rigours of patrolling. In the third and second century BC Roman columns managed to get themselves ambushed with monotonous regularity. Even with more experienced and professional armies it was not unknown for either or both sides to lose track of an enemy, and accidental encounters, some of which resulted in battle, were not unknown. A good commander took care to seek as much information about the enemy's location, strength and intentions as possible before planning his own actions.35
Once close to the enemy, the movements of the rival armies became extremely tentative and hesitant. The rate of march slowed, until the armies camped within a few miles of each other. It was not uncommon for them to remain in these positions, perhaps as little as half a mile apart for days or even weeks before a battle occurred. Skirmishing and single combats between the cavalry and light infantry of the two sides occupied much of the time, and either one or both commanders might march out and deploy their army for battle. There was a strong element of ritual about this posturing. How far forward a side chose to advance its battle line towards the enemy army or camp displayed how confident it was of victory. Camps were usually pitched on high ground so that a force which deployed close to its camp was in a strong uphill position which no enemy was likely to attack. If an enemy army stayed too close to their camp when the other side advanced towards them, or remained behind the ramparts and refused to deploy at all, then this allowed a general to tell his men that the enemy were afraid of them. This was one way of raising their morale, and during the days of delay before battle this was what a commander attempted to do, encouraging his army and trying to give it as many advantages as possible. Cumulatively, even very slight advantages, such as manoeuvring so that the enemy fought with the wind blowing towards them or the sun in their eyes, could contribute to victory. Sometimes though, after spending weeks facing each other, the two armies parted without fighting a battle, one or both sides unwilling to risk forcing a fight. Disengaging when so close to the enemy was a difficult and dangerous operation, but could be preferable to fighting in unfavourable conditions. There was a strong degree of mutual consent about the battles of this period. It proved very difficult even for highly skilled commanders to force an unwilling opponent to fight.36
If and when a battle finally occurred, both sides marched out from their camps to deploy into battle order just as they had when challenging the enemy to battle in earlier days. There was of course always the danger that when they did this, perhaps merely with the intention of displaying their confidence, the enemy would rise to the challenge and join battle. The normal procedure was for an army to form into a column with each unit in the order it was to take up in the fighting line. This column would march out of the camp to a point roughly on the left flank of the planned battle line, then wheel to the right and march along parallel to the enemy until the head of the column reached the extreme right of the line. Then the unit at the head of the column formed up as the right flank unit of the army and each unit took its place beside it. In the case of the Romans, it was normal to form three such columns, one each for the triplex acies, and if there was a possibility of encountering the enemy during a march a Roman army might march for some distance in this formation. It took a considerable time to form up an army in this way and by the time that the process was complete many units may have marched several miles, apart from having the frustration of stopping and waiting for each unit ahead of them to contract from looser marching order into battle formation. It required considerable supervision on the part of senior officers to ensure that this process ran smoothly and that the army ended up in the right order and in the right place. There is some evidence to suggest that in the Roman army this task was a particular concern of the tribunes. In most cases while this process was going on the enemy was engaged in a similarly laborious process, but it was normal to send out cavalry and light troops to protect the vulnerable columns as they deployed.37
The majority of soldiers in both Roman and Punic armies carried some sort of missile weapon, either a throwing spear or javelin, or a longer-ranged sling or bow. Even these weapons could not harm an enemy more than a few hundred yards away. The missile exchanges in ancient battles probably occupied more time than hand-to-hand combat, but it was the latter which was usually decisive. Ultimately victory went to the side most willing to close and attack the enemy with spear and sword. Massed hand-to-hand combat is very difficult for us to picture, in part because it has been exceptionally rare in the warfare of the last two centuries, even in battles involving armies armed primarily with close combat weapons. We have no detailed description of the fighting in a single battle from the Punic Wars, but it is possible to create a composite picture of what infantry combat was like, drawing from all accounts and those of other battles in the same period. These give the impression that close combat was often far more tentative than our imagination, or cinematic representations, may suggest.
The opposing battle lines might begin an action as much as a mile or as little as a few hundred yards apart. Unless holding a strong position, it was normal for both sides to advance, since moving forward gave the men confidence. As they came towards each other, both sides attempted to intimidate the opposition, by trying to look as confident and frightening as possible. They yelled their war cries, blew trumpets and clashed weapons against shields, hoping to make more noise than the enemy. Individual appearance, such as tall plumes, brightly painted shields and highly polished armour, helped to make a man feel more confident and dismay the opposition. Ideally the advance of such an impressive and noisy body of troops was enough to break the morale of the enemy and cause them to retreat or break and run, but such easy victories only occurred when one army had a massive advantage in morale over its opponents. Normally both sides closed to within missile range, perhaps 30 yards or so, and began to throw whatever missiles they possessed. It is probable that the advance checked whilst they did so. Each Roman legionary carried two pila, neither with a range of more than 100 feet (30.5 m) and with an effective range of half that. There was insufficient time to throw both of these missiles whilst running towards an advancing enemy, and it was impossible for a man to hold one of the heavy javelins in his left hand and still use his heavy shield effectively.
It is unclear how long such an exchange of missiles normally lasted, but at some point one or both sides built up enough confidence to surge forward and close the last short distance to contact the enemy. Again, the confidence of the enemy onslaught, the sound of his battle cries and trumpets, or perhaps the casualties caused by thrown missiles might just have been enough to crack the opposing unit's spirit and cause them to rout. Otherwise the two lines met and fought in a cacophony of yells and clashing blades. Only the men in the front rank of each formation could actually strike at the enemy, although in a spear-armed unit some of the second rank might have been able to thrust over the shoulders of the men in front. Relatively few wounds were instantly fatal, and cuts to the right arm or lower legs (especially the left leg nearest to the enemy) most common. The head, unprotected by the shield, was vulnerable to a serious or incapacitating wound and for this reason a helmet was the single most desirable piece of defensive armour after a shield. The objective was to knock down, kill or force back an opponent in the enemy front rank, and then step into his place to begin making inroads into the opposing formation, although this was a dangerous thing to do. A unit's confidence relied very heavily on coherence of its formation, since men were more inclined to remain where they were if they believed themselves to be surrounded by trusted comrades. If the enemy penetrated the front rank of a unit then its members became nervous and were highly inclined to panic and run. It was then, as one side turned to flee, that the greatest casualties were always suffered, the winning side frenziedly striking at the backs of the running men. Those who were too slow in joining the flight, and the injured, especially those with leg wounds, were usually caught and cut down.
Hand-to-hand fighting with sword and spear was hard physical effort combined with massive emotional stress. Such combats could not last for long before the participants in the opposing front ranks were physically exhausted and unable to continue. It is extremely unlikely that such combats ever lasted for more than fifteen minutes, and most were probably far shorter. Many such conflicts, perhaps the vast majority, did not end with one side fighting its way into the enemy formation and routing them, but were indecisive. In these cases the two sides seem to have separated, a gap of a few yards opening up between the lines, one or both units taking a few paces back. There was then a lull as the rival fighting lines faced each other, recovering their strength and confidence, perhaps yelling or throwing any remaining missiles at the foe. Eventually one or the other side was able to surge forward again in a fresh charge and renew the combat.
With each time that the opposing lines came into actual contact and parted without a decisive result, it must have become harder after the next lull to urge the weary men forward once more. Both sides steadily became exhausted, the front rank men from the physical effort of fighting, the men in the rear ranks from the strain of waiting, unable to see much of what was going on and knowing that at any moment their formation might collapse and a vengeful enemy appear to slaughter anyone who did not run quickly enough. Casualties were comparatively few - judging from Greek battles probably less than 5 per cent until one side fled - but fell most heavily on the boldest men, the ones who tried to cut their way into the enemy ranks. Less confident men were inclined to drift to the rear. Greek military theory recommended putting the bravest men in the front and rear ranks of a unit. The ones in front did the actual fighting, whilst the ones in the rear prevented everyone else from running away. The Romans stationed theiroptiones behind the line to push the men back into place if they tried to flee. The number of leaders, both formally recognized officers or chieftains and those particularly bold individuals willing to go forward and lead a new onslaught, was vitally important. This highlights the importance of the large number of junior leaders in the Roman legion and its encouragement of boldness, especially individual boldness, in its soldiers. The confrontation between two lines of infantry may well have lasted for an hour or more, since we hear of battles lasting at least two to four hours altogether. Sometimes we hear of one line being forced back without breaking for several hundred yards. In the end these contests were decided by two major factors, stamina and aggression. A fighting line needed stamina to endure such a long slogging match. Discipline and experience added to a unit's endurance, as did the depth of its formation, since the men in front could not flee until the ranks behind collapsed. Aggression was needed to persuade the men to advance one more time and close to contact with the enemy, since this was the most likely way of causing his eventual collapse.38
This is a rather different picture of ancient combat, but makes the tactics of the period, and especially those of the Roman legion, far more intelligible. In the context of such tentative fighting, the presence of gaps between the units forming a line becomes less important. In fact it is clear from our sources that all armies maintained intervals separating their units. We hear of light infantry moving forward to skirmish and then retiring through the gaps in the main lines of Roman and Carthaginian heavy infantry at Trebia in 218. It would in fact have been impossible to move an army across even the flattest country without significant breaks in the line, since otherwise the units would inevitably collide, merging into one and becoming very difficult to control. The main difference between the Roman and other systems was that the intervals between maniples were especially large, not that the gaps existed at all. Charging enemies, even 'wild barbarians', did not sweep through these gaps and swamp the maniples in the first line because the charges delivered in reality appear to have been far less concerted and rapid than those of popular imagination. Such charges were themselves not delivered by an enemy line that was solid anyway, but one made up of distinct units or groups with small gaps between them. Far more importantly the intervals in the Roman line were covered by the maniples of the next line.39
All armies apart from the Romans tended to concentrate the vast majority of their infantry strength in a single line. Hellenistic armies, for instance, preferred to deepen their phalanx rather than form troops into a second line and made little or no use of reserves. This was in part because their commanders, usually monarchs, were obliged by tradition to fight in person at the head of their Guards and were in no position to send orders to reserve formations. The deepening of the phalanx also gave it great stamina in combat. In every battle over half of the Roman infantry were initially kept uncommitted in the second and third lines. Deeper lines had more stamina, but even the men in the rear ranks were affected by the exhaustion of prolonged combat. The Roman system allowed fresh men to be fed into the fighting line, renewing its impetus and leading a surge forward which might well have been enough to break the wearying enemy. The wide intervals between maniples made it easier to reinforce a combat in this way. Committing the reserve lines required careful judgement on the part of a Roman commander. Too early and the fresh troops risked being absorbed by the front line and sharing their exhaustion. Too late and the fighting line might collapse, perhaps even sweeping the second and third lines away in its rout. A good commander kept a tight rein of his reserve lines and restrained them from joining the combat on their own initiative, as excitable and nervous men and centurions were eager to join the fight. The triariitraditionally squatted or kneeled down in the third line. The posture made it easier for them to brace their spear-butts on the ground and present a hedge of point to the front, but it may also have been intended to discourage them from moving forward prematurely. The triarii numbered fewer than half of either of the first lines and traditionally offered a refuge for these to retire behind, hence the expression 'the affair came down to the triari? which applied to any desperate situation.40
The Roman military system was directed to the single end of applying massive, steadily renewed pressure to an enemy in front. The second and third lines were not true reserves in the modern sense, and only in the most experienced legions were they capable of any form of manoeuvre. The legion's drill and tactics were ideally suited to the formal, almost ritualized battles of the period. The marching camp with its formal layout and the wide lanes between the tent lines and behind the rampart allowed the troops to form up in the columns used to deploy into battle order and then each march out through one of the gateways. The large number of officers with the army helped to regulate and control this process. Indeed the system placed considerable responsibility in the hands of the general and senior officers, which in itself belies the old view that these men were inexperienced and scarcely necessary for the army to function. With much of the army kept in reserve at the beginning of an action, it was important that someone, usually the commander himself, took the decision to commit these fresh troops. Roman commanders did not charge spear in hand at the head of their Guard cavalry like Alexander the Great or the Successor Kings. On some occasions, especially in a desperate situation, a Roman general might choose to lead a charge, but he did not expect to spend the entire battle this way. Roman generals tended to stay near to the fighting without actually joining it, riding around just behind the fighting line. From this position they were able to encourage their men and also, through the noise made by and appearance of the men in combat, judge how well the action was going and issue orders to commit their reserves accordingly. The general needed to guess where the most important fighting would occur and move to that point in the line, although all along the line tribunes, allied prefects and the general's immediate subordinates or legati were usually stationed to cover each section of the front. It was a style of command which made great demands on senior officers and put them at considerable risk, for their close proximity to the fighting line put them at risk from missiles and the attacks of lone enemies. Roman commanders needed to be mobile, moving from one crisis point to the next or riding back to fetch reserves in person when these were required quickly and there was not time to send a message. For this reason it was normal for Roman commanders to lead their armies on horseback, and even the dictator, who was banned from riding by archaic taboo, by this period automatically sought permission to ride. Roman soldiers fought better when they believed that their general was with them, able to observe and either reward or punish their behaviour.41
The Roman army was well suited to formal pitched battles, where it could form up against an enemy to its front and attack straight forward, throwing in men from the reserve lines to reinforce the main attack, to plug a breakthrough in their own line or exploit a penetration of the enemy's. Until well into the war with Hannibal Roman commanders were indeed inclined to seek such a confrontation as swiftly as possible. Hannibal in particular was to prove far more skilful in the careful manoeuvring before a battle, exploiting the instinctive desire of his Roman opponents to meet him as soon as possible to, ensure that the battle was in fact fought in a situation and place of his own choosing. Yet it was a striking feature of the Romans, especially in their military enterprises, that they were willing and able to learn from their opponents and adapt.