Late Republic

By 100 bce the Roman army had become effectively a professional one. The legion was now made up entirely of heavy infantry, organised into circa 600-man cohorts. Legionaries were trained to both fight and serve as combat engineers, a dual function that led to great Roman success at siege warfare.

In what is known as the Social War (91-88 bce) the Italian allies rose up against Rome, during which, according to Velleius Paterculus, 300,000 were killed on all sides, a figure which matches that of the Second Punic War but is much less likely.^[486] While the fighting certainly was severe, it is worth noting that the war ended in large part through Roman accommodation and not through brutal suppression. The one exception to this was the treatment of the Samnites, whose lands were severely devastated. The fact, however, that this damage was carried out systematically six years after the war ended, suggests it might have been due to a political decision and not to emotion.

There was increasing civil unrest in Rome, often breaking out in violence, in which the army increasingly participated. In such civil strife, cruelty could be common: for example, the soldiers of Cinna displayed the heads of their political opponents in Rome in 87 bce.

Julius Caesar's writings were rhetorical and tendentious, but he describes battle scenes that he personally witnessed in great detail and in general accurately. His numbers, however, are another matter. While he claims to be using the Helvetii's own records, his figure of 263,000 for the tribe's total population, and 153,000 for those killed fighting the Romans, is not credible.^[488] Figures from other sources need to be treated even more sceptically. Velleius Paterculus states that 400,000 Gauls died in the ten years of the Gallic Wars.³¹ Plutarch gives even higher figures: he writes that 3 million Gauls fought in the wars, a million were killed and a million captured.^[489] Such high numbers are questionable, to say the least, and ought not to be accepted uncritically.

Archaeologists from the Vrije Univesiteit Amsterdam announced in 2015 that the remains of a battle near the city of Kessel had been carbon dated to the first century bce, suggesting a connection to Caesar's victory over two German tribes, the Tencteri and the Usipetes in 55 bce. Descriptions of this discovery routinely cite a number of 150,000 Germans killed, though this number is not in Caesar's description.

While isolated pieces of evidence can be illustrative, one has to be very careful in building a case based on them. In dealing with ancient literary passages or archaeological remains we must be cognisant of the fact that very little survives, and that what we do find is difficult to interpret. Of course, simply disregarding literary evidence because it is rhetorical (it is all rheto­rical) or archaeological evidence because it is ambiguous (it is all ambiguous) is as misleading as simplistic credulity. Nevertheless, there is simply no basis for a definitive conclusion on the nature and extent of Roman military violence during the Republic.