Rome did not employ professional soldiers. Instead, uniquely amongst powerful states by this period, she continued to rely on temporary militias, raised whenever required and then disbanded at the end of a conflict. Every five years a census was carried out of all Roman citizens, listing their property. Soldiers were expected to provide their own weapons and equipment, therefore a man's census rating determined not only whether or not he was eligible to serve, but also in what capacity. The majority of Roman soldiers owned small farms, since land was the main basis of wealth. As citizens they were legally obliged to serve for up to sixteen years or campaigns, but until the Punic Wars such prolonged military service was extremely unusual.1
This bronze Boeotian helmet was found in the River Tigris, but is an example of a type commonly worn by Roman cavalrymen. These helmets were made from sheet bronze which was hammered over a carved stone to give it its distinctive shape.
The word legion (legio) had originally meant levy and referred to all the troops raised by the Republic in one year, but by this period the legion was the basic building block of the Roman army. Our best description of the legion is provided by Polybius and was written in the middle of the second century BC, more than sixty years after Cannae. The historian claims that his description does in fact refer to the war with Hannibal, and his narrative of these campaigns was certainly based upon this assumption. However, it has sometimes been suggested that the army did not assume this form until after Cannae and that at the time of the battle its structure was much less flexible. There is insufficient evidence to solve this question with absolute certainty, but the evidence for a major reform of the Roman army after Cannae is unconvincing, and rests largely on a single passage of Livy describing a local tactical ploy used in 211.2 On the whole it is likely that the Roman military system in 216 BC differed only in minor details from Polybius' description.
The Polybian legion consisted of cavalry, heavy infantry, and loose order skirmishers. Cavalry were provided by the wealthy equestrian order and included the sons of many senators, eager to make a name for courage and so help their future political careers. Their equipment had been copied from the Greeks and consisted of bronze helmet, mail armour or a metal or linen cuirass, circular shield, sword, spear and javelins. Later Roman horsemen employed the four-horned saddle, which provided an excellent seat, and it is distinctly possible that this was already in use. The basic organization was the turma of thirty, subdivided into three groups of ten each led by a decurion. Normally there were ten turmae per legion, providing a cavalry force of 300, but we also read of legions with only 200 cavalry, so this probably varied.3
The main strength of the legion was its heavy infantry, who were divided into three lines on the basis of age and experience, since all possessed the same property qualification. The first line (hastati) consisted of young men in their late teens or early twenties, the second line (principes) were men in their prime (which for the Romans was considered to be the late twenties), whilst the third line (triarii) was composed of the experienced, older men. Each line was divided into ten basic tactical units, the maniples, but for administrative purposes these were split into two centuries each commanded by a centurion. The centurion of the right-hand century was senior to his colleague, and commanded the whole maniple when both officers were present. Centurions were appointed or elected from amongst the ordinary soldiers. Each was assisted by his second in command (optio), a standard bearer (signifer), trumpeter (cornicen), and a guard commander (tesserarius).4 The soldiers in all three lines carried the same defensive equipment of a bronze helmet, a pectoral or chest plate, probably a greave for the left leg, and a bodyshield (scutum). This was oval in shape, about 1.2m (4 feet) in length and 60cm (2 feet) in width and constructed from three layers of plywood, each laid at right angles to the next. It was thicker in the centre and flexible at the edges, making it very resilient to blows, and the top and bottom edges were reinforced with a bronze edging to prevent splitting. Good protection came at a price, for the Roman shield was very heavy, around 10kg (22 pounds), and in battle its entire weight was borne by the left arm as the soldier held the horizontal handgrip behind the boss. Wealthier soldiers replaced the bronze or iron pectoral with a cuirass of mail or scale armour which, although heavier, offered far better protection. All soldiers carried a short thrusting sword, which probably was already of the type known as the Spanish sword (gladius hispaniensis) - the classic sidearm of the Roman soldier for over five centuries. Most also carried a dagger. The triarii were armed with thrusting spears, up to 2nr. (8-9 feet) in length, but the hastati and principes both carried the famous Roman pilum. This was a heavy javelin consisting of a wooden shaft some 1.2m (4 feet) in length attached to a narrow iron shank 60cm (2 feet) long, topped by a small pyramid-shaped point. All of the weapon's weight was concentrated behind the small tip, giving it great penetrative power. The length of the metal shank gave it the reach to punch through an enemy's shield and still go on to wound his body, but even if it failed to do so and merely stuck in the shield it was very difficult to pull free and might force the man to discard his weighed-down shield and fight unprotected. The piluin's maximum range was about 29m. (c. 100 feet), but its effective range something like half that. According to Polybius each soldier carried two pila, one lighter than the other, but the archaeological evidence suggests rather more variety than such a simple, clear division.5
Antenna type sword, the most common sidearm of the Spanish infantry at Cannae.
Supporting the heavy infantry and cavalry were the light infantry skirmishers or velites, recruited from the poorer citizens and those as yet too young to serve in the hastati. They were armed with a small round shield, sometimes a helmet, a sword, and a bundle of javelins, but it is unclear whether they were organized into units and how they were commanded. Many wore pieces of animal skin, especially wolf skin, attached to their helmets and Polybius believed that this was intended to allow senior officers to recognize individuals and reward or punish their behaviour, but is vague as to who these officers were. The number of triarii was fixed at 600 in ten maniples of sixty, but the remaining infantry strength of the legion was divided equally between the hastati, priticipes and velites. A legion normally had 4,200 foot, and therefore there were 1,200 men in each of these contingents, but in times of particular crisis the total might be increased to 5,000 or even more. As a result, the size of a maniple of Hastati or priticipes could vary from 120 to 160 men when the legion was first formed and before any campaign losses had occurred.6
In battle the three lines of heavy infantry were formed one behind the other. In each line there was a gap equivalent to its frontage between each maniple. The maniples in the next line were stationed to cover the gaps in the line ahead, forming a quincunx pattern, like the 5 on a die. It has often been doubted that the legion actually fought in such an open formation, and various theories have been developed to explain how the intervals between maniples were closed just before contact, but such views are unconvincing and there is not a shred of evidence from our sources to support them. All armies formed battle lines with some intervals between their units, otherwise it was impossible to move without the units merging into one mass too large for its officers to control, and the gaps in the Roman formation were wider than usual. The open formation gave the manipular legion great flexibility and allowed it to move across fairly broken country without losing order. With more than half of the legion in the second or third line, and thus uncommitted at the beginning of a battle, the Romans had plenty of fresh troops with which to plug a gap in their own line or exploit a break in the enemy's. Above the sixty centurions there were six military tribunes in command of each legion. A pair of these officers held supreme authority at any one time, but all were available to direct the legion in battle.7
Part of the relief on the first century BC altar of Domitius Ahenobarbus from Rome, this depicts soldiers wearing equipment very similar to that worn by the legions at Cannae. This scene shows a sacrifice of popanum cakes and a bull, part of the complex rituals required to prepare a Roman army for war. To the left of the altar stands an officer, most probably a military tribune.
The military tribunes, like all of the senior officers of the Roman army, were not professional soldiers but elected magistrates. The Romans did not maintain the strict division between army and politics common in modern democracies, and senators followed a career which brought them both military and civilian responsibilities, sometimes simultaneously. The two consuls elected each year were the senior magistrates and also provided the commanders for the most important of the State's wars. By modern standards they were amateurs, who received no formal training for command and instead learned by experience of service with the army in various junior capacities. The amount of military experience possessed by a consul inevitably varied considerably, but most displayed talent as leaders of men, even if they lacked the more technical skills of an army commander. Roman magistrates rarely stood for election on the basis of particular policies, instead relying on their reputation for ability. It was a system which heavily favoured a small group of wealthy aristocratic families who were skilled at promoting the virtues and successes of former generations and implying that as much or more could be expected from younger members of the family. With only two posts per year, competition for this high honour was intense, especially since a mixture of law and tradition prevented anyone attaining the rank before their early forties, and was supposed to prevent it being held twice within ten years. The vast majority of the 300 or so senators never became consul, and it was very rare even for the members of the established families to win the office more than once.8
The standard Roman army was commanded by a consul and consisted of two legions supported by soldiers from the Italian allies. The latter were organized into wings (alae) with roughly the same number of infantry as a legion, but as many as three times the cavalry. Each ala was commanded by three praefecti who were Romans, but very little is known about their internal organization and tactics. The alae were divided into cohorts, each provided by a single community, which appear to have varied in size from about 400 to about 600 men. It is unclear whether these were in turn subdivided into maniples, perhaps one for each of the three lines, or how often the cohort itself was used as a tactical unit. Our sources pay little attention to the allies, and give the impression that an ala operated in much the same way as a legion. The normal formation for a consular army was with the infantry of the two legions in the centre and an ala on either flank, so that the latter were usually known as the 'Left' or 'Right' ala. The cavalry of the two legions are usually depicted as stationed on the right wing, the place of honour, whilst the Latin and Italian horse formed on the left, but, given that there were often three times as many of the latter as the former, this may be an oversimplification.
The pick of the alae were drawn off to form the extraordinarii, elite cavalry and infantry at the immediate disposal of the consul, and sometimes these were used as a distinct tactical unit in battle. The entire consular army usually consisted of at least 20,000 men, but sometimes the military situation required a smaller force and a single legion and an ala might be employed. In this case the army was usually commanded by a praetor, the next senior magistrate, four of whom were elected in each year.9
The Roman army in this period functioned best at the level of the consular army and it was very rare for any enemy to pose so great a threat that the two consuls were required to join forces and give battle together. On the rare occasions that this was considered necessary, as when Hannibal invaded Italy, it was normal for the consuls to hold supreme command on alternate days. Deeply embedded in the Roman political system, and the military hierarchy was essentially an extension of this, was the desire to prevent any one man gaining overwhelming power. Therefore, just as in politics any grade of magistrate had several members, all with equal power, so also in the military organization there were three decurions to a turma, two centurions to a maniple, three prefects to an ala and six tribunes to a legion. Only with the appointment of a dictator was this principle abandoned for a set, six month period. Differences of opinion between consular colleagues Scipio and Sempronius Longus figure heavily in our sources before the defeat at Trebia in 218 and recur when Minucius Rufus was granted power equal to the dictator in the following year. It is clear that these narratives have been partially distorted by the desire of some senatorial families to absolve their members from blame for these defeats. However, this should not obscure the fact that the divided command was a weakness in the Roman military system when it was called upon to wage war at this level.10
Probably the greatest strength of the Roman military system was the vast reserves of manpower which underlay it. The precise figures may be questioned, but Polybius’ survey of the male citizens and allies eligible for call-up in 225 BC produced a total of over 700,000. This gave Rome the capacity to absorb casualties which would have forced any other state to seek peace. It was especially difficult for the Hellenistic kingdoms to cope with heavy losses, both because of the time taken to train soldiers and also because of the relatively small population from which their recruits were drawn. In civilian life Roman citizens had considerable protection under the law, but nearly all of their rights were sacrificed as soon as they enlisted, legionaries willingly subjecting themselves to an extremely harsh system of discipline. The death penalty was inflicted even for such crimes as theft within the camp, and the punishments for flight, failure to perform duties, or for desertion were as harsh. The Roman army was highly organized and disciplined, and in these respects compared well with more professional forces. However, its essential impermanence was often a weakness. It took time to absorb recruits, train them to fight as units, accustom them to trusting each other and their officers. The longer a Roman army remained in existence, assuming that it did not suffer constant defeats, the more effective a fighting force it became. By the end of the Second Punic War some legions had been in constant service for over a decade and were as well drilled and confident as any professionals. Yet as soon as an army was discharged the whole process had to begin again. Each new levy usually included men with prior service, but they had not served together in the same legions and maniples under the same officers before, so still needed extensive training. Most Roman armies had the potential to be very efficient, but it took time and considerable effort on the part of its officers at all levels to realize this potential.11