Although Hannibal had advanced the centre of his main line to form the crescent-shaped formation before the battle, these troops do not subsequently seem to have moved any further forward. Instead, once the skirmishers had fought their indecisive combat and withdrawn through the narrow gaps in the main lines, it was the Roman infantry who attacked. Elsewhere Polybius twice tells us that it was the Roman custom at this time to advance noisily, the men cheering and clashing their weapons against their wooden shields, whilst the trumpeters carrying the curved military horn, the cornu, added their blare to the cacophony of noise. Visually the massed ranks of Roman infantry can only have been an intimidating sight. At a distance, and the armies may have begun up to about a kilometre apart, the small intervals between maniples were probably scarcely visible and the Roman centre must have appeared an almost solid mass of rank upon rank of armoured men behind oval shields. The legionaries and allied troops were probably not uniformed in the modem sense, since each man supplied his own equipment and we do not know for instance whether legions painted their shields in a certain colour or with a specific device, but the minor differences in appearance would only have been apparent at close range. It is probable that, like soldiers in later Roman armies, the men in 216 had taken care to dress well for a battle, polishing armour and helmets, and donning the tall crests which added to each man's apparent height. By the time that the Roman centre began to lumber forward, the men were undoubtedly covered in a thin layer of the dust, clouds of which continued to be thrown up by their marching feet and whipped around in the gusting wind.32
The idealized barbarian: a Gaul who has already killed his wife stabs himself. At Cannae this Roman stereotype, noble but defeated, was far from reality.
The greatly outnumbered Gauls and Spaniards watched as the grand Roman attack came straight towards them. They too raised their battle cries, clashed weapons together and blew their trumpets, including no doubt the tall camyx which was said to produce an especially harsh note. Perhaps individual warriors ran a little way forward to show off their prowess and display their contempt for the enemy, for such acts of bravado were common in tribal warfare. Our sources emphasize the wild and frightening appearance of these tribesmen, the Spanish supposedly in their usual white tunics with red or purple borders, the Gauls clad in trousers but with bare torsos. The noise and displays served the same purpose for both sides. It was hard for troops to advance or wait in silence to fight a visible enemy, and shouting relieved the tension and helped the men to cope with their growing fear.
They shouted louder to show themselves that they were not truly afraid, and the more their comrades joined in the more they encouraged each other. Thus soldiers urged themselves on, whilst the noise they made and the appearance of confidence they presented would hopefully intimidate the enemy. The early phases of a battle were fought as much in the mind as with physical weapons, for if one side shouted louder and appeared more formidable then the other side’s spirit declined and might even collapse. In extreme cases appearance alone was enough to convince troops that they could not win and put them to flight before a blow was struck. It was said that the German tribes could tell which way a battle would go simply by listening to the shouts raised by the rival armies.33
In this case both sides were highly confident and do not seem to have been unduly intimidated by the opposition. The Gauls and Spaniards had the confidence of past victories over similar Roman armies, and perhaps there was pride too, for they had been chosen from all the Punic army to be the first to meet the enemy's main attack. This was an opportunity to prove their courage in plain view and may have exploited a similar urge to the one which had led the naked Gaesatae to run out ahead of the main line and challenge the enemy at Telamon in 225 BC. The advancing Romans trusted to the superiority of their numbers, but some may have remembered that even in their recent defeats legionaries had often prevailed over such unarmoured warriors. Both sides were encouraged by the many officers in or near the front of the formation, moving around and urging on the soldiers.34
Eventually the Romans came within range of missiles. Tests with reconstructed piln suggest that this heavy throwing spear had a maximum range of between 25 and 30m, and effective range of about half that distance. Other javelins, including those probably used by the Gauls and some of the Spaniards, may have had a slightly, but not substantially longer range. Whether all soldiers waited until they were within the most effective range before throwing their pilum or javelin is highly questionable. Modern studies of combat suggest that only a minority of soldiers actually fired their personal weapon during a firefight and that even fewer did so with care and took trouble to aim. Both sides had been yelling for some time, nervously watching as the gap separating their own and the enemy's line grew narrower. Shouting helped them to fight against their fear, but the urge to do something to strike at and frighten off the approaching enemy must have been overwhelming. Throwing a missile at that enemy was the best way of striking at and perhaps driving off the foe.
In the later, professional Roman army, the tactical doctrine was to advance slowly, in good order, and complete silence towards the enemy. Then, probably within 15m - the pilinn's effective range - they delivered a devastating volley of pila and immediately charged, at last breaking their silence to yell out a war cry and letting the trumpets blare. The first century AD Jewish historian Josephus, who gives us our only account from a non-Roman of what it was like to face such an attack, spoke of the terrifying moment when legionaries finally broke their silence and charged. This method of fighting required an immensely high level of discipline which was only the product of good training. Even so, there appear to have been cases when this discipline was not enough and the attack was not pressed, degenerating instead to a more sporadic exchange of missiles. Most armies in the third century BC, and especially the legions at Cannae, simply did not have this level of training. Polybius tells us that each legionary carried two pila, one lighter than the other. It would not be physically possible for a man to throw two missiles with a range of less than 30m whilst he ran charging towards the enemy. This was especially true if that enemy was in turn charging towards him. Nor was it possible for a soldier to hold the second pilum with his left hand and still employ his shield properly. Not only was the Roman scutum very heavy, but it was held with a horizontal handgrip making it impractical to clutch this firmly and hold onto the shaft of a pilum at the same time. If both pila were carried in battle, which seems likely but is not certain, then the Romans must have halted for a while, close to the enemy line, to allow time to throw these before actually charging home. Probably a degree of hesitation a short distance apart was normal if both sides had failed to frighten the enemy sufficiently with their appearance, noise and confidence as they advanced. A recent study of the tactics of Roman infantry in the Republic suggests that exchanges of missiles could occupy considerable time.35
As the Roman centre started to come within missile range men on either side began to throw spears, javelins or pila. At first most probably dropped short or lacked the momentum to drive through the wooden shields which protected most of a man's body. Later, as the distance separating the two lines narrowed, some of the missiles began to strike home with greater force, punching through shields and perhaps even helmet or armour. Men crouched behind their long shields to gain as much coverage as possible, the Romans doubtless walking forward with heads bowed as if walking into a wind in the timeless posture of infantry advancing under fire. The majority of wounds were probably to the unprotected lower legs and occasionally to the face. Such casualties were probably led away to the rear, although there is some evidence of men fighting on with a number of non-incapacitating wounds in certain circumstances. The Roman pilum frequently had a barbed head and was designed to be very difficult to remove once it had punctured a shield. Some of the Gauls and Spaniards were most likely faced with the choice of dropping their shields after a pilum hit or fighting on with the shield awkwardly weighed down by the heavy weapon.
The whole Roman line appears to have halted once it came close to the enemy centre, even though much of the Roman front was still a fair distance from Hannibal's refused flanks. Close contact with the enemy often appears to have resulted in such inertia in ancient battles and the movements of armies were far more tentative than we might expect. In the central sector huge numbers of missiles were thrown by either side, but the vast majority fell short or struck harmlessly against shields. On either side only the front, and to a lesser extent the second ranks could actually see the enemy and make any effort to aim. The men behind were simply lobbing their weapons blindly forward in the hope that they would land somewhere amongst the enemy mass. There were roughly the same number of Gauls and Spaniards as there were hastati in the first Roman line and it is probable that both were formed in considerable depth. As a result many men, perhaps over half of each line, were too far away to have any hope of reaching the enemy with a thrown javelin. Fatal casualties were few on either side, but the sheer number of long- shafted javelins whizzing through the air made all the participants vividly aware that they were now in physical danger. This added to the pressure already created by the closeness of the enemy, their appearance of strength and confidence and the noise of war cries and trumpets, all of which continued throughout the exchange of missiles.