Ptolemy II’s chief concern early in his reign was to secure his rule against the possibility of any interdynastic disputation of the throne. When his sister Arsinoe arrived in 280 or thereabouts from the northern Aegean, in flight from her disastrous marriage to Ptolemy Ceraunus, he made her marry him and adopt his children, so that there were no loose ends. Marrying his full sister was an extraordinary step for Ptolemy to take. There were only faint traces of such a practice in Egyptian and Persian pasts (though Mausolus and other fourth-century dynasts of Caria had married siblings), but Ptolemy gloried in it: around 272 he inaugurated a joint cult of Alexander and the Sibling Deities—himself and Arsinoe, though both were still alive (she died in 270). Nor were these minor cults: in both the Seleucid and Ptolemaic kingdoms, the priests of dynastic cults were always important men, and royal appointees. There came to be several such cults in each kingdom, often conjoined with those of Alexander, and the Greek cities followed the kings’ lead.
Brother–sister marriage was supposed to guarantee the purity of the bloodline, to advertise the solidity of the royal family, and to secure stability by eliminating the possibility of rival claimants to the throne; the king was effectively cloning himself, and so every generation of Egyptian kings took the same name. Perhaps surprisingly, there is no real evidence of genetic deterioration over the more than two hundred years that the Ptolemies, or some of them, practiced sibling marriage. Strife within the royal family became increasingly savage, but savagery has characterized many courts throughout the ages. Sibling marriage was a symbol of power, a way for the Ptolemies to claim that conventional morality did not apply to them. If the satirical poet Sotades’ reaction is typical, the Greeks were appalled: “Unholy the hole into which you push your prick.” Sotades paid for the quip with his life.
The mainspring of Ptolemaic foreign policy was the need to keep intact the extensive buffer zones that the first two Ptolemies had put in place around Egypt by the middle of the third century. Apart from their defensive function, these overseas possessions made up for Egypt’s deficiencies in minerals and ship-quality timber, and enabled them to control the trade routes of the eastern Mediterranean, the Aegean, and the Black Sea approaches. They were the major suppliers to the Greek world of grain and other commodities, and they needed to make sure that their cargoes were safe.
In Greece and Anatolia, as we have seen, the Ptolemies supported whichever state or states seemed best able to check Antigonid and Seleucid ambitions. In the Aegean, they did their best to retain their possessions against a strengthening Macedonian navy. In Palestine, Coele Syria, and Phoenicia, they fought a series of wars against the Syrian kings. In Cyrenaica, the earliest Ptolemaic external possession and one of the most important, they used diplomacy to keep the peace and allowed the rulers to think of themselves for a while as royalty. In Africa, they extended south into Nubia, especially to safeguard the provision of war elephants and gold.
From the moment he ascended to the Syrian throne in 223, Antiochus III intended to recover the entirety of the kingdom when it had been at its greatest extent, under Seleucus I, as though he still had rights to it. He was delayed by Molon’s rebellion, but once it had been put down, Antiochus drove the Ptolemaic forces out of Coele Syria and coastal Phoenicia. This task occupied the first two years of the Fourth Syrian War. But Ptolemy IV, who had come to the Egyptian throne in 221, belied his reputation for being more interested in poetry than politics, or took the advice of his powerful chief ministers. Having restructured his army and greatly increased the number of native Egyptians serving in it, he inflicted a massive defeat on Antiochus at the battle of Raphia in 217. With over 140,000 men (and 175 elephants) between the two sides, which were fairly evenly matched, this was the greatest battle since Ipsus. After acknowledging defeat, Antiochus withdrew to northern Syria, and over the next few weeks almost every single place that he had gained or regained returned to Ptolemaic control.
This was a great victory for the Egyptians, but it proved to be a peak from which they could only fall. Trouble had been brewing for a long time, with occasional outbursts, since a good number of Egyptians, especially in the south, resented being a subject race and the exploitation of their land by foreigners. Before the battle of Raphia, the Ptolemaic governor of Coele Syria had gone over to Antiochus, and Ptolemy’s queen is said to have offered every soldier in the Egyptian army two gold minas. Even with the exaggeration, it seems that the Ptolemies were finding it hard to retain the loyalty of their men.
To judge by the concessions that were made when the troubles were over (reductions in tax, for instance, and concessions to the priesthood), social discontent was the major factor. Very probably, Egyptian priests were behind the disturbances; during the decades of Persian rule, the temples had grown hugely powerful, forming a kind of nationalist underground, much as the Greek Orthodox Church did during the Turkish rule of Greece, and the Catholic Church did in Ireland under British occupation. When the Ptolemies arrived, they did their best to appease the powerful priesthoods, by performing all the rituals appropriate to their position as pharaohs, by allowing the temples to prosper, and by personally funding the building and rebuilding of temples. Many of the monumental Egyptian remains that survive today date from the Ptolemaic era.
Nevertheless, it was clear that this velvet glove concealed an iron fist. There were garrisons everywhere; soldiers were a common sight on any town or city street, especially since the country was so often on a war footing; the kings presented themselves as warriors. Polybius described the inhabitants of Alexandria as Egyptians, Greeks, and mercenaries, “heavily armed, numerous, and coarse.” Ptolemy II’s far-famed parade, held in Alexandria perhaps in 278, included eighty thousand soldiers; even Adolf Hitler’s fiftieth birthday in 1939 was celebrated by only fifty thousand.
Disturbances began not long after Raphia, both in the Delta and in Upper Egypt, and the Egyptian soldiers who had fought in the battle were right at the center of them. Although the kings were never seriously threatened, and probably retained control of the Nile valley, there were occasions between 205 and 186 when men in Thebes were calling themselves pharaohs, or perhaps were being allowed to call themselves pharaohs. The country was seriously weakened by two decades of internal strife, and, as we shall see, only Roman intervention stopped it falling to the Seleucids.
Egypt was a relatively self-contained unit, geographically speaking; it consisted of the Nile Delta on the Mediterranean and a thin strip of fertile flood plains a thousand kilometers (620 miles) south up the river valley to the First Cataract (the first stretch of shallows), never wider than thirty kilometers (twenty miles) at any point and bounded by desert to east and west. The kingdom comprised about 23,000 square kilometers (about 8,880 square miles) and had a population of four or five million.
Settlements along the river were perched on high ground, to avoid the annual mud-depositing floods, the source of the country’s great fertility and wealth; at the time of the floods, they were turned into islands. There were three main areas of settlement. Lower Egypt, the Delta region in the north, was densely settled; it was on the far west of the Delta that Alexander chose to site Alexandria. To the southwest of the Delta lay a large, fertile depression called the Fayyum, where the arable land was hugely increased by a massive drainage and canalization project initiated by the Ptolemies. Then Middle and Upper Egypt sprawled up the Nile, and included two great cities: Memphis in the north, the religious center of Egypt, and Thebes in the south, famous for the temples of Karnak and Luxor. The many-streamed and marshy Delta was hard to cross, so Memphis (near modern Cairo) was the usual gateway to Egypt from the east—though there was the Sinai Desert to cross first.
Egypt had been a major center of culture for hundreds of years before the Macedonians arrived to form its thirty-first and final dynasty, and Ptolemy I had less city-building to do than the Seleucids. Many Egyptian Greeks therefore lived in non-Greek environments, in close relationships with the native populations. Ptolemy’s only large foundation (or refoundation: it replaced a smaller Greek settlement) was Ptolemais in the southern Thebaid, which, with its different dialect and ethnic makeup, had a perennial tendency to regard itself as a separate state, and so needed a regional administrative center.
The Ptolemies, like all Hellenistic kings, also founded many smaller settlements (for instance, by settling mercenaries on the land, like the Seleucids), but there were only ever the three Greek cities in Egypt itself (not counting Egypt’s overseas possessions)—Alexandria, Ptolemais, and Naucratis. But Alexandria by itself was an enormous project. Founded in 331, it was still largely a building site when Ptolemy designated it his capital, perhaps in 313, and marked the occasion by moving Alexander’s body there from Memphis, where he had first laid it to rest after the hijacking. The city was divided into three sections: one for Greeks and Macedonians (who were the only full citizens and were privileged with tax exemptions), one for Egyptians, and one for everyone else, who were mainly Jews—the second largest Jewish population after Jerusalem. Until the growth of Rome, Alexandria was the greatest city in the Mediterranean. Even in the first century, one visitor could say: “It leaves all other cities a long way behind in terms of its beauty, size, financial liquidity, and everything that contributes to graceful living.” But it was also beset with all the usual urban problems, from corruption to ethnic tension.
The winds of change echoed along the Tiber as the year 216 began, and though a skeptic might have heard the winds howling disaster, most Romans seemed confident they were blowing toward a quick and decisive victory.
The strategy seemed sound; pressure would be applied in all the appropriate directions. Marcellus, the reliable and hyperbelligerent spolia opima winner, was sent to keep an eye on Sicily, the fleet there having been augmented for a potential invasion of Africa. More on target, in late 217, Publius Scipio, now recovered from his wound, had joined his brother Cnaeus and his two legions in Spain with eight thousand fresh troops and a small fleet. Both Scipios had been given the proconsular imperium to tear up Barca land and rob Hannibal of his base. Nor was Hannibal’s Gallic connection overlooked. Twice-consul L. Postumius Albinus was given two legions and dispatched north to break the rebellion in Cisalpine Gaul and seal off any further support from that quarter. But the central objective, the overwhelming priority, was to directly confront Hannibal and crush him beneath the weight of Rome’s key advantage, military manpower. Everything points to a corporate decision to stage a great battle and obliterate the invader once and for all. Fabius was out, the bludgeon was in.
Viewed from the comfortable perspective of subsequent events, the reasoning that led to Cannae is easy to dismiss. But it was far from implausible. Arguably, there was a fundamental Hannibal problem: if you didn’t beat him, you couldn’t get rid of him. On the other hand, if the Punic force were to lose even one significant battle, it was too far from any secure base to survive. Just one Roman triumph, a single day’s victorious fighting, would put an end to the invasion. The string of previous defeats could be convincingly attributed to impulsive commanders, impiety, bad weather, bad luck, bad timing … The excuses were endless. Meanwhile, Romans still had good reason to believe in their military system—after all, its fundamentals would provide security for nearly another half millennium. They had merely to supersize it and leave nothing to chance.
Of course they were wrong, and Fabius Maximus had been right. Lacking a secure base, Hannibal probably could have been attrited out of existence. But the victory at Cannae would allow him to sink his claws deep into Italian soil, and then he would prove far harder to uproot. So the battle proved to be much more than a human tragedy and a tactical debacle; it was the strategic basis of fifteen more years of Hannibal, what must have seemed at times like perpetual Hannibal. About the only things the Romans could salvage from Cannae were Scipio Africanus and perhaps ten thousand disgraced survivors, and one day they would avenge themselves and Rome by drawing the Carthaginian away and then defeating him nearly as badly as he had defeated them. But that day was still far in the future.
Deciphering any political environment is difficult, more so an environment twenty-two hundred years old and littered with deceptive contradictions, patronage relationships, and family alliances. Although modern historical scholarship has clarified the climate of opinion and motivation to some degree, we will never know exactly what Romans were thinking in 216. Therefore, while it is possible to say that as the year began, attitudes had hardened and grown more overtly aggressive, certain issues remain veiled in obscurity.
For example, Livy (22.33) tells us that a Carthaginian spy, who had gone unnoticed for two years, was caught right around this time. His hands were cut off, and then he was let go. In the same breath Livy adds that twenty-five slaves were crucified for forming a conspiracy in the campus Martius, the field where Roman troops customarily drilled. The two events seem related. Why else would they be mentioned together? Also, from this point Hannibal’s intelligence advantage begins to diminish, or at least it appears to, on the basis of available narratives. Was this spy the Punic mole, and were these slaves his spy ring? It can be inferred as such, but not with certainty. It may be that the Romans avoided saying too much about what could have been considered an embarrassment and a vulnerability.
Other deceptions are more apparent. Both Livy and Plutarch would have us believe that the consular elections of that year, which determined the commanders at Cannae, were basically contests between the impulsive “people,” whose choice was the lowborn knave and demagogue C. Terentius Varro, and the prudent patriciate, who managed to secure the elevation of the wise and experienced Lucius Aemilius Paullus as a brake on his hotheaded and foolish colleague. The historians even stage a tête-à-tête during which Fabius and Paullus agree that the former’s strategy of avoiding battle is the best approach and that the impulsiveness of Varro is virtually as dangerous to Rome as Hannibal. Livy even insinuates that on the day of the battle, Varro issued his orders to fight without bothering to inform Paullus. Polybius, while less hyperbolic in his denunciation of Varro, is nevertheless plainly sympathetic to Paullus and largely absolves him from blame. But all of this becomes more difficult to swallow in light of the fact that after Cannae the apparently incompetent Varro was given a number of other important commissions and even military commands—although this also may have been a means of shifting the blame. Meanwhile, Polybius’s exculpatory portrayal of Paullus fades somewhat when it is realized that Paullus was the grandfather of the historian’s patron, Scipio Aemilianus.
Modern historians have come to understand that a more likely explanation is that Varro, the first member of his family to rise to the consulship, and largely without illustrious descendants, was tagged by later generations as Cannae’s designated scapegoat, while Paullus’s reputation was rescued by the later propagandizing of his powerful family.6 Actually, Varro may have served under Paullus during his first consulship three years before, when they’d been campaigning in Illyria, and both were probably now on the same side of the debate over how to fight Hannibal.
This amalgam of confrontationists was likely built around the powerful families of Aemilii and Cornelii, particularly the Scipionic branch, and included Minucius and Metilius, the tribune who’d worked to elevate Minucius to equality with Fabius Maximus. Probably they were opposed by the Fabii and the older, more conservative members of the senate, who could be assumed to have stood on the side of patience and the gradual attrition of the invader. Yet the policy of patience was plainly in eclipse, perhaps even among some of its adherents. After all, they were every one of them Romans, and the Roman default position was to fight. A measure of this enthusiasm was that as many as a third of the senate joined the army at Cannae, and most of the other senate members had close relatives among the ranks. This showdown with Hannibal was intended to be the magna mater of all battles, and an analysis of those selected for magistracies in 216, especially as military tribunes, shows them to be considerably more experienced in military matters than was usually the case. Plainly much of the leadership was ready to stake their future and the future of their respective gene pools on this gigantic roll of the dice.
So were the rest of Rome and Rome’s allies. The contemplated instrument of destruction was to be an army roughly twice the size of any previously assembled by the Romans to operate as a unit. Varro and Paullus would each command double armies of four legions plus equivalent allied units, but the whole mass was expected to fight together—eight legions and eight alae, in effect a quadruple consular army. Given that a Roman army operated best as a maneuver unit when it was composed of two legions and two alae, there was reason to suspect that this monster might prove inherently unwieldy—a lumbering Frankenstein of a force at best, and at worst a paralytic, a quadriplegic consular army. And this raises the question of who actually would be in charge overall. Meanwhile, to compound the effect, each of the legions, and presumably the alae, was increased from forty-two hundred to fully five thousand, which added up to a grand total of eighty thousand infantry. As we shall see, quantity had a quality all its own … but not the one the Romans expected.
The sole area where the force seems somewhat shorthanded was cavalry—six thousand, two thirds of it allied, when the normal legionary and alae component might have been expected to yield almost ten thousand. Apparently recent losses, especially those of Centenius, had taken their toll, and this too would prove telling at Cannae.
More specifically, the army that would confront Hannibal had two basic components. The first was the force left to keep an eye on him at Gerunium, an experienced element with a history of heart-stopping ups and downs—mostly the latter. Its core was built around the two legions that Publius Scipio had managed to salvage from the defeat at Trebia, soldiers who earlier had been repeatedly ambushed by Gauls. The legions had later been taken over by Geminus, then transferred to Fabius Maximus to chase and lose Hannibal, and then they’d nearly been destroyed under Minucius. To make up for casualties and other attrition, they were bound to have been reinforced on multiple occasions, but at least the veterans had served together and under the same officers for a period of years.
The second element was essentially virginal, the Roman portion consisting of four new legions all recruited around the beginning of the year. While these troops as individuals appear to have received the rudiments of military training as part of their upbringing, the process of integrating them into maniples and teaching them to fight as units not only took time—presumably the spring and early summer—but would have resulted in only a thin behavioral veneer of mutual trust and confidence, which, without the experience of actually fighting together, could be ripped away fairly easily in an emergency to reveal a substrate of panic and helplessness. Next to nothing is known about the allied components, but if this was a newly recruited force, it’s hard to imagine they were any more tested than the Romans, nor would they have been used to their officers, who were also Romans.
It is not clear when the two forces joined together. Polybius (3.106.3) talks about sending new recruits forward for experience skirmishing, but these recruits appear to be reinforcements for the legions already at Gerunium. Although Livy (22.40.5) maintains that the new legions arrived before Hannibal left winter camp and headed toward Cannae, modern opinion favors a delayed linkup, as late as less than a week before the battle. Given this, it’s hard to imagine that the Romans’ juggernaut was in any meaningful way integrated; rather, it remained two separate armies that on the day of battle would be cut up and welded together for its moment of truth, tactically a dubious proposition.
Yet it could be argued, and probably was at the time, that the Roman military system made their troop formations inherently interchangeable, and therefore more easily mixed and matched. No doubt the injection of experienced leadership was counted upon as a lubricant. And there was the intangible of morale. The allied forces in particular were furious over Hannibal’s rampage across the Italian countryside and were vengefully eager for combat. Meanwhile, the Roman troops seem to have been embarrassed, not daunted, by previous defeats and were now grimly determined to prevail.
To seal the deal, the establishment did something unprecedented. Livy (2.38.2–5) tells us that once the allied levies arrived, the consuls had the military tribunes formally administer an oath to all the infantry and cavalry that they would depart from their ranks only to secure a weapon, kill an adversary, or save a comrade. Previously, the historian notes, this had been a voluntary pledge among the soldiers themselves; now flight in the face of the enemy was against the law. To some societies and some armies this might have seemed a mere formality, but the Romans were literalists and legalists. And as we shall see, it was this oath that would determine the fate and futures of those who might have thought themselves otherwise lucky to escape the death trap at Cannae.
As a fighting force, the fated quadruple consular army was large and full of Romans, both good things. But it was also full of vulnerabilities. It had a substantial number of light troops, perhaps as many as twenty thousand, but they were of suspect quality. If not the “armed servants” one source calls them, they were clearly not as effective as their Carthaginian equivalents. These soldiers had been scattered like chaff at Ticinus and Trebia … though they did seem to stiffen somewhat under Fabius and Minucius. Still, it is probably telling that Rome’s staunch ally, old King Hiero of Syracuse, looking for ways to help, thought it wise to donate one thousand light troops of his own, some of them archers (apparently the only archers at Cannae).
The cavalry was probably even weaker; it had already suffered savage losses, and the skills involved are not easily replicated on short notice. The ranks were probably swelled a bit by members of the senate, who were by definition equestrians, yet many would have been old and past their military prime. Besides, the majority of cavalry were allies, and Hannibal’s well-advertised leniency toward them could be expected to have an impact on their fighting spirit in a pinch, especially among forces with the greatest ability to get off the battlefield in an emergency.
The obvious strength of the Cannae army was its heavy infantry. Even if it could not be screened effectively by light troops and its flanks protected by cavalry, it was big enough to be relatively immune from harassment, provided it could maneuver and win decisively with some degree of dispatch. Yet this was also a force subject to emotional volatility—its better half, the experienced element, had been defeated more than once by Hannibal, and the other part was a mass of neophytes, with all that that implied. Temperamentally, it was an army likely to overreact—prone both to excessive enthusiasm and passive despair. Judging by his plan and its results, this was exactly what its Barcid nemesis anticipated.
It is not certain when Hannibal left Gerunium and headed south, but he likely would have waited until the early summer, when the crops were ripening and his troops could forage. Livy concocts a tale of an attempted ambush and then a night escape in the face of the already united Roman army, but Polybius’s version that Hannibal marched out past the guarding force of Geminus and Regulus, who were under orders not to engage until Varro and Paullus arrived with the rest of the army, is simpler and more plausible.18 The historians agree on one point: the Punic force was hungry. Whether on the basis of good luck or good intelligence, Hannibal gravitated to Cannae, around sixty miles south of Gerunium, nearly to the Adriatic coast. Here he captured a grain storage and supply dump in the ruined citadel on the heights of the otherwise abandoned town. His food problem solved, at least temporarily, he made no further effort to move, and this was telling.
Cannae sat at the bottom edge of an immense treeless plain, the largest south of the Po.19 It was ideal for cavalry operations and large-scale maneuvers, exactly the terrain for magnifying Punic military advantages. By this time Hannibal probably understood that the Romans were intent on a showdown and were putting together a monster of an army for the occasion. So far, his efforts to chip away at the Italian alliance had come to nothing. He needed a really spectacular victory to generate the kind of political impact to begin breaking off Rome’s affiliates. This was the perfect time and place to inflict it. Staying here was tantamount to accepting the challenge. He had only to await his opponent’s arrival.
But, being Hannibal, he probably kept himself busy preparing the reception. If we accept Polybius’s version of the events leading up to the battle, the Punic army may have been alone at Cannae for several weeks. This is a dangerously long time to give any commander to prepare a battlefield, much less a commander with Hannibal’s fertile military imagination. By this time he was likely on a first name basis with every rise and depression, every twist in the River Aufidus, every potential campsite, every approach and escape route, every possible advantage he could squeeze out of the surroundings and then blend with a battle plan that seems to have been derived from his cumulative observations of Roman fighting tendencies and the capabilities of his own troops. As always, much remained to be determined by chance and the circumstances of the actual engagement, but it’s a safe bet that during this respite Hannibal’s mind seldom wandered from the upcoming test.
His plan when it was finally hatched implied a great deal of faith in his army. This faith was not misplaced. The gang of desperadoes that had stumbled off the Alps not much more than a year and a half earlier had been but a scrawny prototype of the force that now awaited the Romans at Cannae. Freeze-dried no more, the men and horses had regained their health, had eaten their fill, and were rested. We know that key elements had been systematically rearmed with the best of the captured equipment, and it is likely that many others had picked up bits and pieces of what had once been Roman weaponry.
Another change had to do with the Gauls. By this time they were much more reliably integrated into the fighting force. They still fought together, to take advantage of their peculiar tactical characteristics, but at Cannae small units of Gauls were interspersed with Spaniards, indicating that their tribal allegiances had been effectively superseded by the command system that controlled the rest of the army. Very likely the process that had begun for the Gauls on the slog through the Arno marshes was now complete. They were now not only fierce and brave individually; they were also disciplined, well trained, and above all reliable at the unit level. And as such they would play a critical role at Cannae.
Psychologically, this was an army that had known nothing but the most decisive sort of success since it had entered Italy. In a life-and-death struggle, confidence is crucial, and the recent past had given these men every reason to believe in their own fighting skills, as well as their commander’s ability to drive opponents into positions of utter vulnerability and near helplessness. Many must have already killed Roman soldiers personally and must have also observed them reduced to an abject state, begging for mercy. That was Hannibal’s point when he reassured an officer worried about the size of the opposing force at Cannae. (“In all this multitude there is no one who is called Gisgo.”) For Carthaginians, more Romans simply meant more Romans to kill. This was the dark side of a truly professional fighting force, especially one that fought with edged weapons; they were used to killing, inured to it. They would kill without hesitation. It was a terrible advantage that the Carthaginians had and that most of the Romans at Cannae lacked.
Nowhere was this more evident than in the cavalry, probably the most lethal Punic fighting component. As had been true since Trebia, the Spaniards and Gauls rode together as a shock element, although now they were almost certainly better trained and integrated. The Spaniards carried two light throwing spears, a sword, and a round shield, or caetra. The Gauls, primarily composed of nobles, were more heavily armed and armored, with chain mail, metal helmets, and a stout thrusting spear. The two groups would have been an impressive one-two combination, with an initial hail of javelins followed by closer, more decisive engagement. This was a force more than capable of taking on anything the Romans had on horseback, and likely predisposed to fight in the same very confrontational way—one entirely different in its ethos from the other face of Punic cavalry.
They were the Numidians, Hannibal’s version of killer bees, proverbially swarming their opponent if given even the slightest opening. The Numidians were the closest thing a western Mediterranean battlefield saw to an inner Asian steppe horseman. They lacked only the steppe horseman’s deadly composite bow, relying instead on a brace of light javelins and a slashing dirk. Characteristically, Numidians pinned and herded their foes through absolute mastery of their hyper-agile ponies, and then ran the enemy down with ruthless efficiency, able to cut their hamstrings even at a full gallop. Like the steppe horsemen, they were fatally easy to underestimate. Riding bareback and carrying only a light shield for protection, they avoided hand-to-hand combat and were largely incapable of direct confrontation. Polybius (3.72) describes them as “easily scattered and retreated, but afterwards wheeled round and attacked with great daring—these being their peculiar tactics.” Yet in the hands of a commander as opportunistic as Maharbal, they could destroy an entire force once it became even slightly demoralized and ready to bolt.
All together, Hannibal’s cavalry now numbered around ten thousand, two thirds more than when he’d entered Italy, and more to the point, they were enjoying a five-to-three quantitative edge over the Cannae-bound Romans, whose horsemen were by far inferior in quality also. Looked at another way, the Carthaginian force had one horseman for every four foot soldiers, while the Roman ratio was one to thirteen, a strong indication that the Punic army was far better adapted to the flat terrain on which the battle almost certainly would be fought. All in all, it constituted a yawning gap, and one that would soon send the Romans stumbling down the initial steps toward tactical ruin.
Numerically, the Carthaginian advantage in cavalry was nearly reversed with regard to infantry. Polybius and Livy both agree that forty thousand foot soldiers would have been available to Hannibal at Cannae, a figure modern sources support. But if the infantry were outnumbered two to one, the quality of the Punic soldiers was better, and not just in terms of confidence and prior experience killing Romans. The Carthaginian force was notably less homogenized than its Roman equivalent, and in those various parts were vested a variety of fighting skills tailor-made for a commander with Hannibal’s protean military imagination. The ancient sources provide no specific figures for the various contingents, but modern historians have made a number of informed estimates that seem basically in agreement.
The approximately eight thousand Punic light troops were probably proportionately even more outnumbered by the Roman velites than was the case with their comrades in the other infantry arms, but the relative difference in personal capabilities was equivalently lopsided in the other direction. Basically, Roman skirmishers were men either too young or too poor to take their place in the maniples. The Carthaginians were specialists—screening and harassment was their business. Numidian javelin men, perhaps six thousand of the total light troops, proved particularly adept at cooperating with their horse-borne countrymen and seem to have intensified the effects of the cavalry’s swarming tactics. Though less numerous, the other major component of the Carthaginian light infantry—the Balearic slingers—were, if anything, individually even more lethal. They were both feared and coveted as mercenaries throughout the western Mediterranean. Much overlooked by modern historians, the sling was capable of launching a projectile toward its target at up to 120 miles per hour—fast enough to kill a man at fifty paces. While light troops in general played a secondary role at Cannae, at least until the later stages, one shot delivered by a slinger early in the battle may have played a critical role in compromising the Roman leadership.
Hannibal’s close order heavy infantry probably numbered around thirty-two thousand at this point and came in three varieties: Gauls, Iberians, and Libyans. Despite the heavy infantry’s having suffered relatively severe casualties at Trebia and Trasimene, there probably remained around sixteen thousand Gauls, whose shock value and increased reliability we have already considered. There were likely about six thousand Spaniards left at this point, a fraction of the original contingent that had made up the bulk of the army when it had departed New Carthage. In Darwinian terms, we can assume the fittest survived. Nevertheless, Polybius (3.114.4) leads us to believe they wore no armor, but only a purple-bordered linen tunic, and possibly not even a helmet; but since a lot of captured Roman equipment was available, this may not be entirely accurate. Tactically, these Iberians were most interesting because they fought like Romans, first throwing a heavy javelin not fundamentally different from a pilum, and then weighing in with either a straight or a curved short sword and a large oval shield. Interspersed with Gallic units, they could be construed as having had a stabilizing effect on the critical center at Cannae, allowing the more impulsive Gauls to rush forward, hack away for a while, and then fall back, leaving the Spaniards to fight the pursuing Romans on their own terms.
Finally, there were the Libyans, presumed to be Hannibal’s best-drilled and most elite maneuver element, since they were the first to receive Roman equipment and because they formed the jaws of the trap that snapped shut on the Romans at Cannae. Thus far he had used them scrupulously and in ways they would take few casualties, so of the twelve thousand who had come down off the Alps, probably around ten thousand remained. But if these African spear-fighters formed the teeth of the Punic force structure, the other elements were the claws and muscle and sinew of this beast of battle. Hannibal’s genius as a commander was his ability to devise and execute a plan that used all the parts in concert to swallow and digest a much larger prey.
We pick up the Roman juggernaut where we left it at the beginning of Chapter II, slouching toward Cannae, proceeding with the utmost caution. The Romans may have found the flat coastal terrain reassuring, since it gave Hannibal nothing to hide behind should he try to stage an ambush. No doubt they sent the cavalry out to reconnoiter just in case.
According to Polybius’s version, the two halves of the great army probably joined up on the road in late July, with Geminus (Regulus, the other proconsul, seems to have been sent back to Rome because of advanced age, to be replaced by Minucius) having followed Hannibal south at a respectful distance, and Varro and Paullus intercepting Geminus near Arpi, roughly two days’ march north of Cannae. With the army combined, there were eighty-six thousand mouths to feed. So it made sense to keep the contingents separate for as long as possible. The hunger of the army would also place time constraints on the commanders to seek decisive combat once they got within striking distance of Hannibal. Ironically, it seems perhaps the dinner tables had turned; though Livy (22.40.7–8) would have us believe that as battle approached, Hannibal was also running out of food. If true, and not merely the historian’s way of saying Fabius had been right all along, both sides needed a fight quickly.
The Romans had no trouble locating Hannibal, since he was hardly hiding, and they set up camp initially approximately six miles to the east on the broad plain that runs down to the Adriatic. There ensued an elephantine pas de deux, as the two armies warily closed the distance while at the same time trying to gain some tactical and psychological advantage.
The Romans, however, were literally of two minds, since Varro and Paullus followed the tradition of alternating command daily when consuls operated together—just what Fabius had refused to do the year before, which had thus enabled him to save Minucius. Because of the curtain of blame later cast over Varro, it is hard to separate actual disagreement between the two consuls from aspersions dumped on Varro retrospectively. If it is conceded that both consuls wanted to fight, and most modern historians do concede this, the evidence such as it is points to Paullus as the more cautious of the two, particularly worried about being caught on the flat ground ideal for Carthaginian cavalry.
The vicinity of the anticipated field of battle was dominated by high ground to the southeast, where the abandoned town of Cannae and Hannibal’s first camp were located, and bisected by the River Aufidus, a shallow, narrow watercourse running in a northeasterly direction toward the sea. The terrain to the northwest, over and beyond the left bank of the river, was broad and flat. The area between the right bank and the highlands toward Cannae, while still level, was more broken and constricted. Hannibal almost certainly preferred the left bank but could and would fight on the right side; both Roman consuls wanted to avoid the left side and stage the battle in the most confined area possible. The days preceding August 2, 216, were an extended test of wills that saw Hannibal unsuccessfully harass the Romans with light troops as Varro initially approached. Then Hannibal moved his camp to the left bank and formally offered battle on this side, first to Varro and then to Paullus, only to be refused. Finally, Hannibal sent Numidians after the Romans’ water bearers, and this gesture provoked the Romans into action, albeit on the right side of the Aufidus. Meanwhile, the Romans had moved too close to withdraw safely, and so they split their army into two camps, leaving two thirds on the left side of the river and the remainder in a smaller enclave on the right bank. The stage was set.
The Roman battle plan at Cannae can be summarized in three words: “pack the middle.” Because this approach would play into Hannibal’s own scheme and lead to a great disaster, it is easy to dismiss the plan as nonsensical. It wasn’t. Rather, it had a clear purpose, to maximize the Roman numerical advantage in infantry while minimizing the obvious Carthaginian superiority in cavalry. It was also based on past experience. At Trebia, ten thousand legionaries had finally managed to hack their way through the center of the Punic line, and had they been able to do it sooner, they could have split and pivoted to crush the Carthaginians attacking on each flank. Even amidst the surprise and demoralization of Trasimene the impetus of around six thousand Romans had carried them through the Carthaginian stopper force, only to be captured later. We can assume that Varro and Paullus and those advising them were confident in the ability of their troops to puncture the heart of the Punic line, and were intent on doing this as rapidly as possible. Geometrically this called for a narrow, thick formation, exactly the configuration on the day of the battle described by Polybius (3.113.3), “placing the maniples closer together than was formerly the usage and making the depth of each many times exceed its front.”
Breaking though in this manner was decidedly not a matter of simple momentum, like some gigantic rugby scrum pushing inexorably forward. Romans fought primarily with short swords, so the cutting edge was by definition the first line of combatants. True, pila could be launched from several rows back, but any soldier behind around line eight would probably have hit a fellow Roman up front.
The real arguments for this type of human geometry had to do with order, endurance, and psychology. Long narrow columns are easier to keep together, and, they therefore move faster and more cohesively on the battlefield. The many lines to the rear also insured an almost inexhaustible supply of fresh fighters to take the place of the fallen and exhausted, a kind of conveyor belt of shark’s teeth. Finally, a great many of the Roman participants at Cannae lacked combat experience; the middle of such a formation was a safe, psychologically reassuring place for them. One source equates this to the human instinct to herd together for mutual comfort, but without considering that this was actually prey behavior. The thickened manipular order could be expected to have massive combat endurance, which would make the formation almost impossible to defeat by frontal attack, and would thereby allow the unit to move steadily forward. But what would happen if it faced the unexpected, was hit from an unanticipated direction? At this point herding behavior might become just that, dissolving the maniples into a crowd compressed to the point of mass helplessness. The legions would lose the ability to replace frontline fighters through now nonexistent gaps between units. It was a prospect not pleasant to contemplate, and one we can be pretty sure the Romans failed to consider.
Hannibal may have. Just how much he knew of the Romans’ plans prior to battle is impossible to say. Though Livy (22.41.5–6) maintains that “all the circumstances of his enemies were as familiar to him as his own,” whether Hannibal understood beforehand the degree to which the Romans would pack the middle remains open to question. Yet his experience fighting them would have warned him of their will and of their ability to break through in the center, and the likelihood of their trying it again. Also, given his knowledge of Greek military practice, he was doubtless familiar with the Athenian tactics at Marathon in 490, when the Greeks withheld their center and crushed the Persians with their wings. As the day of battle approached, all of this must have been taking root in Hannibal’s fertile brain as he roughed out the framework of an even more lethal trap. As we shall see, the final details would await the contingencies of the battlefield, but the basic plan of using the Romans’ own greatest strength against them was inherent in Hannibal’s deployments and therefore had to have been plotted in advance.
Though there is a tradition of viewing Hannibal as simply being up to his usual tricks—hidden attackers and fake surrenders—the key deception at Cannae was far more subtle. In essence, the trap was hidden in plain sight, something that even today does not seem to be fully understood. Basically he planned to string a line of combined Gauls and Spaniards between two very deep columns of Libyans positioned on either flank, so that viewed from above the formation would look like a backward block letter C. The idea was that as the legionaries rushed forward, the Gauls and Spaniards would give way in a measured fashion (this was critical), leading the Romans farther and farther in between the two columns of Libyans, who would then be in a position to stage a devastating simultaneous attack inward from either flank, stopping the Romans dead in their tracks and leaving them all but surrounded.
Maps of the battle, which are invariably drawn from a bird’s-eye perspective, make the net results clear enough, but also reveal the central deception in a way that leaves open the question: “Why would anyone be dumb enough to walk into such an obvious trap?” But from ground level it would have been far from obvious.
The analogy of American football is helpful. This very intricate game can be enjoyed and understood by the public precisely because it is viewed from on high; the deception is designed to be seen at ground level, and from this perspective deceptions are profoundly confusing, requiring all manner of coaching, cues, and experience so that players are not fooled on every play. As the Romans approached the Carthaginian line, all they would have seen was a continuous line of men, with no way of knowing the varying depth at either side. As the Romans pushed forward, their attention would have been focused straight ahead and toward the center, where they were making the greatest progress. When the Punic flanks attacked, most of the Romans would not have realized it was even happening. They would have known only that their body of men had strangely come to a halt. By this time it would have been too late. They were as good as dead.
Assuming that Hannibal did not have direct knowledge of the Romans’ plans and simply had to anticipate what they might do, the Carthaginian’s plan faced several worrisome contingencies. Expecting his adversaries to pack the middle implied they would deploy on a fairly narrow front, not much wider than his own. Should the Romans march onto the battlefield in a more normal formation, their advantage in numbers would leave the Carthaginian line seriously outflanked on both sides, affording a perspective that would not only betray the depth of the Libyan columns, but also would force the Carthaginians to abandon the ambuscade by pivoting outward to make up the difference. Also, the Roman and allied cavalry could not be ceded the initiative; if they were allowed to sweep around to either side of the Carthaginians, the jaws of the trap would be revealed and their commanders could be warned before it was too late. None of this happened; instead the Romans played into Hannibal’s hands as if choreographed; but such schemes are always vulnerable to the unexpected and this could account for the shadow of what is possible to interpret as Plan B in the sources—Appian’s (Han. 20) story of Hannibal concealing some cavalry and light troops in ravines on a hill (presumably the bluffs leading up to Cannae) with orders to attack the enemy rear as at Trebia, and Livy’s (22.48) tale of 500 Numidians staging a fake capitulation, being conducted behind enemy lines, and later producing hidden swords and assaulting from this quarter. Neither is taken very seriously by modern historians; but they were hardly out of character for a fox full of tricks.
As the sun rose on Tuesday, the second day in August, the scarlet tunic signifying battle could be seen displayed above the tent of Terentius Varro, whose turn it was to command the Cannae army. Polybius says Varro’s men were eager for the fight, were at a near fever pitch of anticipation from the waiting. Orders would have been distributed to the tribunes in the night. The tribunes then would have assembled the men and cavalry in time to march out of camp just after dawn, cross the river, and join the troops in the smaller encampment on the right bank. All were now present, with the exception of ten thousand (probably a legion and an ala) left to guard the main camp and stage an assault on the Punic encampment during the battle. It is likely that the men guarding the main camp were the bulk of those fated to survive the day and become the living ghosts of Cannae. Those less fortunate, around seventy-six thousand men, would move into the customary formation—velites out front; triplex acies, compacted in the middle; and cavalry on the flanks—all awaiting the Carthaginians and destiny.
But exactly where were they? The short answer is that we will never know the precise site of the battle for sure; but that said, the issue has stirred up enough controversy over the years to make it worth considering. Geographically, there are basically two reference points—the location of Cannae itself and the River Aufidus, now called Ofanto. There are also two reliable historical artifacts: we know from Polybius that the battle was fought on the same side of the river as the smaller Roman camp, and the Roman line faced roughly south, with its right flank anchored by the river. It also makes sense that the Romans would have wanted their left flank resting against the highlands on which Cannae was perched, the idea being to make it impossible for Hannibal’s cavalry to sweep around either side to envelop them. The problem is that the distance between the bluffs and the modern Ofanto is far too narrow to accommodate anything like the size of the Roman army, no matter how compacted.
This led a number of respected scholars to propose that the battle was actually fought on the left side of the river, or on a broad plain to the east of Cannae. But the problem with the first view is that it clearly contradicts Polybius, who seldom made this sort of mistake; the drawback to the second is that the flat area to the east is easily wide enough to give Hannibal’s cavalry complete freedom, which raises the question of why Varro would have bothered crossing the river to fight there. Yet all of these interpretations assume that the course of today’s Ofanto is identical to that of the Aufidus, likely a bad bet, given the passage of twenty-two hundred years. This assumption is questioned by modern historians Peter Connolly and Adrian Goldsworthy. Their ingenious alternative is that the ancient river ran considerably to the north as it passed Cannae, leaving flats of about 1.3 miles, wide enough to fit the Roman order of battle as it was assembled that day. This hypothesis remains open to conjecture, but this alternative location seems to be the most plausible for what would shortly become the most prolific killing ground in the history of Western warfare.
If this was indeed the point of deployment for the Romans, it must have inspired great confidence. The inexperienced citizen and allied cavalry, stationed at the extreme ends of the line, right and left respectively, had been relieved of any offensive responsibilities; the cavalry had simply to guard the flanks while the infantry did its work.
Similarly, the numerous but qualitatively inferior velites that were spread out in front of the army had no particular mission once the force was deployed, and they could conveniently retreat between the maniples if pressed.
Meanwhile, the reinforced triplex acies seemed unstoppable, and if anything slowed it down, it was at least impenetrable. It must have appeared to Varro and Paullus that they had finally positioned their forces in a way that even Hannibal could not bend to his own advantage.
Now it was his turn to do just that. Hannibal apparently sent the Balearic slingers and Numidian foot soldiers across the river at about the same time the Romans crossed, but the mission does not seem to have been to interfere with enemy deployments so much as to set up a perimeter behind which the Punic cavalry and heavy infantry could line up. When this was done and he was certain the Romans were ready to fight, Hannibal followed. We can conjecture that the cavalry crossed the river first to reinforce the screeners. Next the Gallic and Spanish infantry joined them, lining up in the center, followed by the two bodies of Libyans, who took their place on either side but remained in columns to form the backward block letter C. By this time the horsemen would have split apart and moved to the flanks, the Spanish and Gallic heavy cavalry on the left facing their Roman equivalents, and the Numidians on the right matched against the allied mounted elements.
As orderly and purposeful as these pre-battle rituals seem in print, the real thing must have provided, even before the fighting started, plenty of distractions and cause for disorientation. At this point the field must have been a jumble of cacophony—horns blaring, drums pounding, swords beating on shields, shouts and war cries reverberating back and forth, to and fro—all the sounds that men muster as they steel themselves to face death and intimidate those they hope will be their victims. Also, more than 125,000 men and in excess of 15,000 horses tramping about in a confined area must have kicked up huge quantities of dust, and it appears that Hannibal’s familiarity with the environment now dovetailed with his desire that the Romans not accurately perceive the true nature of his infantry formation. He had apparently observed earlier that a southeasterly wind, the Vulturnus, gusted with increasing force during the morning, and could be counted on to whip up the dust and blow it into the Romans’ faces, a vexation apparently confirmed by a fragment from the near-contemporary poet Ennius.
Finally, and probably most important, this was August in southern Italy; we can count on it having been hot, and it was bound to get hotter as the day progressed. Most of the Roman heavy infantry and at least the rearmed Libyans would be burdened with between fifty and eighty pounds of arms and armor as they fought for their lives throughout the day. If Trebia had been orchestrated by hypothermia, Cannae was destined to be an inferno where untold thousands were likely to be felled by heat exhaustion, and access to drinking water may well have allowed the Carthaginians to persist in their butchery during the last and most murderous phases of battle.
So it was as they began to fight. The ancient sources agree that the light troops were first to engage, and from Polybius we hear they were evenly matched, neither side gaining an advantage before withdrawing, as was customary, behind their respective lines of heavy infantry. If it had been otherwise, presumably we would have heard more from other sources. So it seems the velites’ numerical preponderance had won the Romans at least a standoff—an auspicious start, considering the multiple drubbings they had taken on earlier occasions. Still, the Punic auxiliaries may have inflicted a very significant casualty at the outset. Livy (22.49.1) reports that the consul Aemilius Paullus, who was with the Roman citizen cavalry, was dealt a severe head wound from a slinger just as the battle commenced—an injury bad enough to leave him unable to ride a horse and bad enough to force his bodyguard to dismount in order to protect him. Polybius does not mention this incident, but it is still suggestive, considering what was about to happen.
The first decisive Punic move came when the Spanish and Gallic cavalry under Hasdrubal—leader of the service corps and destined to perform brilliantly on this day—charged headlong into the opposing horse on the Romans’ right flank. With their adversaries anchored on the river and outnumbering the Romans by more than two to one (around sixty-five hundred to twenty-four hundred), there were none of the normal wheeling maneuvers. Instead the Carthaginians seemed intent on going right through the Romans.
The combat that ensued, Polybius (3.115.3) tells us, “was truly barbaric.” In large part this was because it was mostly on foot. Roman cavalrymen had a decided proclivity toward fighting on the ground, and many of these troopers must have been inexperienced and new to their mounts. But they also may have chosen to dismount because of Paullus’s wound. Plutarch maintains that when the consul was forced from his horse and his attendants got off theirs to protect him, the cavalry assumed a general order had been given and so also dismounted—a development that supposedly caused Hannibal to comment: “This is more to my wish than if they had been handed over to me in fetters.” While it is unlikely the Punic commander actually observed the cavalry getting off their horses, the act nonetheless proved fatal to most of the outnumbered Romans, who were basically annihilated. This is also where many of the Roman senators and others of the equestrian class would have gathered to fight, and ended up making their last stand. It is not clear if Paullus died here—he and his staff may have escaped to join the rest of the army—but Livy’s version makes it seem that this was his end. So at a point when the battle had barely begun, it seems logical that the republic had already been dealt a grievous blow. And it would only get worse. Rather than chase down the last of the survivors who managed to get back on their horses and flee, Hasdrubal reeled in his men from the pursuit, then rested and re-formed them to inflict further mayhem on another part of the battlefield.
Meanwhile, the heavy infantry engagement had begun and had already taken shape, literally, in an unexpected way. As the line of Gallic and Spanish infantry advanced (one source estimates the formation was roughly 840 men wide and 26 deep46), it bowed outward to form a crescent. While some maintain this was natural for a line of men moving forward, others believe it was a last-minute decision on Hannibal’s part. Whichever it was, this convex formation had an immediate and beneficial effect for the Carthaginian side. For as the Roman hastati charged and reached pila-throwing distance, the shape of the line left only a narrow group in the Punic center vulnerable to this potentially devastating missile barrage, and may have led many legionaries to waste their shots while still out of range.
The same thing would have happened as the sides closed for sword-play. Initially at least, the Roman manipular order and their own training would have more or less automatically kept their line straight, and so only the center group of Spaniards and Gauls would have been engaged. The key to Punic success turned on the interior line retreating slowly and in a controlled manner. This was why Hannibal and his brother Mago (presumably joined by other officers and Celtic nobles) stationed themselves here, immediately behind the front, to better manage the action and encourage these most critical of fighters. And the initial geometry of battle served exactly its purpose by committing only a relatively few combatants, and by keeping the huge Roman force at bay until the legionaries in the middle managed to push those at the Carthaginian center back.
As we imagine clusters of bare-chested Gauls flailing their broadswords, interspersed with Spaniards fighting from a crouch behind their shields, all seeking to fend off the surging Romans—themselves bashing forward with their scuta, seeking an opening for their gladii—we should not forget that this sort of combat, essentially a series of individual duels, was both physically and emotionally exhausting. It could not be sustained for more than a few minutes. Once the Punic line failed to collapse immediately, these spasms of violence had to be followed by rest periods when both sides drew back to catch their collective breath for a few minutes. War cries and insults might have been hurled back and forth, followed by pila and other projectiles picked up or passed forward, and then close combat would have been reinitiated. Over time the lulls would have grown longer and the mêlées shorter.
This interrupted rhythm of violence also was to the Carthaginians’ advantage, allowing them to regroup, regenerate, and fall steadily backward in relatively good order. Seeing this, the Romans naturally pushed ahead with increased confidence and growing excitement, focused on their objective of breaking through at the center as quickly as possible. As this happened, the retreating Punic line began to assume an increasingly concave shape, and a critical juncture was reached. Polybius (3.115.6) reports that the Gallic and Spanish infantry in the middle were forced into such a rapid retreat that the Punic line started to break up. As the Roman tide surged forward, it cast caution and training aside and followed the line of least resistance, crowding inward toward the center. The intervals separating the three lines of the triplex acies, and the spacing between maniples, disappeared, and its general organization started to disintegrate. Collectively, the legionaries thought they could see victory just ahead, but it was a mirage; instead, as-yet-unnoticed defeat stared them down from either side in the form of two serried blocks of Libyan heavy infantry, the jaws to the Carthaginian trap.
The moment of Hannibal’s killer epiphany had arrived. The order went out, and man by man the Africans on both the left and right sides pivoted inward, dressed their ranks, and in unison fell upon the Roman flanks, most likely the location of the least-experienced citizen and allied troops. There was little the Romans could do in response besides turning as individuals to face the threat; as units, their formations were too compressed and disorganized to maneuver effectively. They were reduced to a crowd of loners trying to fight off a coordinated engine of destruction. Meanwhile, the emotional shock waves rippled inward, spreading paralysis throughout the Roman ranks and halting the forward momentum of the entire army. Their fate was all but sealed.
That took place in another quarter. Terentius Varro, the overall Roman commander, was with the allied cavalry numbering around thirty-six hundred on the left wing, doing not a lot in the face of a roughly equal number of Numidian horse under either Hannibal’s nephew Hanno (Polybius 3.114) or the resourceful Maharbal (Livy, 22.46.7). The Numidians skirted and swarmed their adversaries as best they could, but were probably thwarted by the Cannae bluffs anchoring the Romans’ position creating the kind of standoff the Romans had wanted on their flanks, a standoff likely satisfactory to Varro. It was around this time, Livy maintains, that the Numidians supposedly staged their fake surrender. But even if this did occur, the Roman consul, who would have had little idea of what had taken place on his opposite flank, was shortly in for an even more unpleasant surprise.
Hasdrubal, fresh from obliterating the Roman horse on the other side of the combat zone, led his reconstituted force of Gallic and Spanish heavy cavalry across the battlefield behind the deployed lines of legionaries, and was soon bearing down on the allied horse with a force that was nearly twice the size. But even before the Carthaginians could bring the charge home, their intended victims evaporated in a panicked stampede, apparently sweeping Varro and his attendants along in their wake. The Numidians, devastating in pursuit, took out after them, killing or capturing all but three hundred of the allies, though Varro escaped to nearby Venusia with seventy of his bodyguards. With Paullus very possibly killed on the right, and Varro removed from the left, Hasdrubal may have already shorn the quadruple consular army of both its consuls.
Yet the focus of his appetite for destruction remained unerring. Rather than rising to the bait of the chase, yet again the Punic commander re-formed his cavalry and instead headed toward the rear of the Roman infantry, quite apparently intent on closing their last avenue of escape. Here, Polybius tells us (3.116.8), Hasdrubal delivered multiple charges at different points, seemingly with devastating effect. On first glance this seems puzzling, since the triarii in the rear should have been well equipped to turn and resist, lining up on one knee with their shields resting against their shoulders and their long spears protruding to form a barrier that horses would not charge. But instead of a solid wall of triarii, it is far more likely the Roman rear was cluttered with a soft mass of up to twenty thousand velites, who had withdrawn behind the maniples shortly after the battle had begun. Most of them were probably adolescents, were very lightly protected, and were lacking room to throw their javelins, and with no avenue of escape, they were virtually the perfect prey for heavy cavalry. Terrified by the horses and the slaughter of their comrades, they would have recoiled inward, exposing their backs and hamstrings to spear thrusts and sword slashes while they pressed desperately against an ever more compacted and undifferentiated human mass.
The same crowding into helplessness must have been taking place on the flanks, as the Libyans on either side continued to press home their attack, an ever-tightening human vise. Meanwhile, the nearly routed Gauls and Spaniards, no longer pursued by the Romans at the forward edge of battle, would have been given time to regroup and turn the tide in their favor. A terrible dynamic was taking place. Assailed from all sides, beyond the control of its officers, with those on the outer edges having no place to go but inward, the Roman army, by pressing itself into paralysis, was becoming if not the instrument of its own destruction then at least complicit in the process.
Somewhere between sixty-five thousand and seventy thousand Romans and allies—depending upon how many had already fallen—were now surrounded. Tactically the battle was over, but the killing had just begun. There was no alternative. The army was still too large and full of fight to be taken prisoner; besides, with its leadership immobilized in the press, it had no real means of surrender. The only choice was its effective extermination, a task the Carthaginians accomplished through systematic butchery almost until the sun set on this terrible day.
Not only does the process beggar description, but exploring the details of the massacre might seem to serve little purpose beyond pandering to some bloodlust with a kind of pornography of violence. Yet war is truly terrible, and to turn our eyes away from its results is in itself an act of cowardice. Hannibal’s great victory, his tactical masterpiece celebrated through the ages, produced, in the end, little more than corpses. But this is probably better illustrated by recounting the subsequent course of events than by moralizing over it now. Nevertheless, there is a more prosaic but still historically valid reason for prying into the details of this exercise in mass homicide. As one source put it, “What remains unclear is how encircled troops, with nowhere to run, could be slaughtered in such a one-sided fashion.” We live in an age when killing is cheap, virtually automated; that was far from the case at Cannae. Other than those who succumbed to the heat, each of the men who died had to be individually punctured, slashed, or battered into oblivion. One modern source estimates that in order for the necessary killing to be accomplished in the eight hours that Appian (Han. 25) estimated the fighting lasted, over one hundred men had to be dispatched every minute. Yet even this astonishing figure underestimates the swiftness and profusion of the slaughter, since the estimate assumes that the killing took place at a regular rate throughout the day and not in a great spasm toward the end, as actually happened. In essence, so many victims, so little time, and that doesn’t even attempt to reflect on the ruthlessness and horror of it all.
Nevertheless, logic tells us that the liquidation of the Roman army at Cannae, if it is possible to reconstruct at all, must have been a matter of mechanics and motivation. We can start here. Among the ancient sources, only Polybius (3.116.10–11) has left us something approaching a plausible description of what must have been the most horrific several hours in all of Western military history: “The Romans, as long as they could turn and present a face on every side to the enemy, held out, but as the outer ranks continued to fall, and the rest were gradually huddled in and surrounded, they finally were all killed where they stood.” In other words, they were finished off from the outside in, peeled like an onion. This makes sense, at least basically, but there were likely to have been other lethal dynamics at work.
Hannibal’s skirmishers—the Numidian foot soldiers and Balearic slingers—having earlier withdrawn to safety behind his main line, must have been intact and available. It’s hard to imagine that he wouldn’t have had them let loose a hail of javelins, stones, and even expended pila onto the stationary mass, a deadly barrage that could hardly have missed in such a target-rich environment, nor avoid inflicting serious injury on Romans who were either too crowded or too exhausted to raise their heavy shields for protection.
Meanwhile, the infantry of Libyans, Gauls, and Spaniards would have continued with their grim work at the circumference. One modern source in an otherwise believable reconstruction of the carnage describes victims “dispatched with frenzied blows, usually to the head.” This seems to miss the mindset implied in the quantity and rapidity of the butchery. Hannibal’s soldiers were practiced killers; very likely most had adopted the cooler, utilitarian approach of the predator, having drawn on our emotional heritage as hunters of the most prodigious and ruthless sort. Moreover, they would have known how to kill quickly and efficiently. If the victim’s back were turned, then a spear or sword thrust to the kidneys would have been so painful as to instantly paralyze, and would have killed within seconds through massive internal bleeding. Or if the victim were facing forward, an equivalent stab to the lower abdomen would have produced the same results almost as fast. Yet such a death stroke—or even more so, multiple death strokes rapidly delivered—implies a certain acceptance, or at least passivity, on the part of the recipient.
This is something the ancient sources—all of them Romans or Roman sympathizers—deny. They would have posterity believe Cannae was, in the words of Polybius (3.117.1–2), “a battle in which both the victors and vanquished displayed conspicuous bravery,” a battle in which the Romans fought stubbornly to the bitter end. Given what has been learned through modern combat studies, this does not seem likely. Even among elite units, when sufficient casualties have been suffered, the whole group slides into a state of apathy and depression more extreme than is encountered in almost any other kind of human experience. “Unable to flee and unable to overcome the dangers through a brief burst of fighting, posturing, or submission,” writes military psychologist Lieutenant Colonel Dave Grossman, “the bodies of modern soldiers quickly exhaust their capacity to energize and they slide into a state of profound physical and emotional exhaustion of such a magnitude and dimension that it appears to be almost impossible to communicate it to those who have not experienced it.” Were Romans tougher and more stoic than modern combatants? Perhaps, but as far as we know, the mental makeup of the ancients was similar to our own. How else would so many of their recorded deeds make sense to us? It seems, then, that the reason why it was possible to kill so many so fast is that most of the victims faced death without resistance. Would this paralyzing combat fatigue have afflicted everyone, and to the same degree? Probably not. Anecdotal evidence indicates that some would have gone down fighting no matter what the circumstances. But the circumstances were really bad, well beyond the limits of most.
If it is possible to conceive of hell on earth, this human abattoir at Cannae must have been the equal of any hell that history in all its perversity has managed to concoct. Thousands upon thousands packed together, unable to move, beset by the cries of those in extremis, many of them dressed in now useless chain suits and cooking-pot helmets beneath the broiling sun, without prospect of water, only death offering any relief whatsoever. As time passed, more and more men would have fainted from the heat, slid to the ground, and been trampled beneath the feet of their comrades, their bodies and discarded shields tripping still others who would then have fallen similarly to their deaths. At the outer edges especially, but also in the interior, where javelins rained, the ground would have grown slick with Roman blood, which would have brought down still others. As at Lake Trasimene the hopeless would have begged their fellows to finish them—presuming there was room for even a short sword thrust—or simply would have done the deed themselves. The stink of death and all the bodily functions that accompany it must have come to pervade the atmosphere and compound the wretchedness of those condemned to take their last breaths there. There was no place worse.
Here and there it is possible to catch a glimpse, even if it’s only a statistic, of a shadow of an actual person caught in the grip of this misery. We know that of the forty-eight military tribunes at Cannae, twenty-nine did not survive. Most would have died in this central killing field, since it was their job to lead the legions. Both quaestors, Lucius Atilius and L. Furius Bibaculus, were likely here also, as were Geminus and Minucius—all of them dead. If Paullus had not been killed earlier on the wing, as Livy suggests, then he too met his end here—according to Polybius (3.116.15), “in the thick of the fight.” This brings to mind the fate of Paullus’s son-in-law.
Nineteen-year-old Publius Cornelius Scipio was young for a military tribune, but he had already seen a lot. He had saved his father’s life at Ticinus and had likely been part of the disaster at Trebia, and perhaps even at Trasimene. At Cannae we know he was attached to the Second Legion, and, given his social standing and his relationship to Paullus, it doesn’t seem likely that the young Scipio was with the ten thousand left to guard the camp. So probably he found himself caught in the dwindling remains of the Roman infantry, once again ensnared by Hannibal’s trickery. It must have been a learning experience, but perhaps a futile one in the face of almost certain death. Yet, contrary to all expectations, he would survive and elude capture, as did thousands of others also apparently hopelessly trapped. Here again this seems to have been a matter of mechanics.
Body buildup had to have become the central problem of the Carthaginian executioners, piles of dead obstructing them from getting at more Romans, not to mention all that slippery blood. A point of diminishing returns must have been reached and a new approach required. Logically this would have suggested a change of venue, a shift in the killing field to less cluttered terrain. The controlled release of clumps of legionaries away from the main mass would have effectively served this end. These Romans could then be run down and killed (or taken prisoner). But this also would have opened a window of opportunity for the Romans, who might have organized into wedges capable of defending themselves until they could reach either of the two camps and some sanctuary, however temporary. Many never could have made it, especially if they got separated. But chance, this opening, and the inevitably growing exhaustion of their Carthaginian tormentors would have collaborated to sweep away from this disaster the core of a class of survivors who would live to fight another day. Young Scipio, it seems, was one of them.
Finally the killing must have trailed off. Polybius maintains that Hannibal rounded up approximately two thousand Romans who had climbed up and hid in the ruins of Cannae, and also captured both Roman camps immediately after the battle.65 While Hannibal may have captured the refugees in the ruins, it doesn’t seem likely that his troops were in any shape to overwhelm a fortified area, no matter how dispirited the inmates. Sleep was probably the only item on their agenda.
If this makes sense, then Livy’s story (22.50.4–12) of what went on in the two Roman camps that night appears believable. Most of the men seem to have been in shock. But those in the larger enclave, having avoided the main disaster and having participated in only a brief failed attempt on Hannibal’s camp, were probably in better shape. These men were still organized and were being led by their officers, who undoubtedly were aware that their present position was untenable. They sent a runner to the smaller camp and ordered them to break out and join forces, so that both elements could slip away under the cover of darkness and make for Canusium, a walled town about twenty-five miles to the southwest. The message fell on deaf ears, until one of the surviving military tribunes, P. Sempronius Tuditanus, made an impassioned speech and got six hundred (Frontinus says it was only sixty-two) men to follow him out to join the others. Not everybody from the big camp was willing to leave. But Livy indicates that some four thousand legionaries and two hundred cavalry in this group arrived safely at Canusium, where they were eventually joined by several thousand other survivors; meanwhile, another forty-five hundred found their way to Venusia, where Varro had taken refuge. All of these men were destined to be reorganized and branded with the stigmatic title legiones Cannenses, the living ghosts of this terrible battle.
How terrible? Dawn of the next day revealed approximately 45,500 legionaries and twenty-seven hundred cavalrymen strewn about a space not much larger than a single square mile. As the Carthaginians set about despoiling the bodies and searching for their own among the dead and half dead, even they were shocked by their handiwork. Livy, the ancient cinematographer, leaves us a scene as surreal as any other in military history:
Here and there amidst the slain there started up a gory figure whose wounds had begun to throb with the chill of dawn, and was cut down by his enemies; some were discovered lying there alive, with thighs and tendons slashed, baring their necks and throats and bidding their conquerors drain the remnant of their blood. Others were found with their heads buried in holes dug in the ground. They had apparently made these pits for themselves, and heaping the dirt over their faces shut off their breath. But what most drew the attention of all beholders was a Numidian who was dragged out alive from under a dead Roman, but with a mutilated nose and ears; for the Roman, unable to hold a weapon in his hands, had expired in a frenzy of rage, while rending the other with his teeth.
If this does not give pause, it is possible to resort to statistics. By way of approximation we can consider each Roman weighed 130 pounds—they were lighter than modern men. Then there would have been well in excess of six million pounds of human meat left to rot in the August sun—the true fruits of Hannibal’s tactical masterpiece, at least for an air force of vultures.
The fate of the others remaining at Cannae was not much better, particularly if they were Roman citizens. According to Livy’s timetable, Hannibal, after allowing his troops much of the day for looting, next made short work of the two camps, gathering up nearly thirteen thousand prisoners. When these men were added to those taken from the ruins on the hill and to those taken from the battlefield, the total was slightly more than nineteen thousand captives. Many of the Romans would end up as slaves in Greece and Crete, still there more than two decades later—another of Cannae’s many legacies.
Hannibal too was left to wrestle with the outcome of Cannae. The fight had cost him between fifty-five hundred and eight thousand men, but at least half of these had probably been Celts, and the army was basically intact. Meanwhile, his men had recovered gold rings numbering in the hundreds, some taken from captives but most pried from the lifeless fingers of senators and equestrians. In a single day Hannibal had decimated a substantial proportion of Rome’s leadership, a blow that some might well have considered mortal. Maharbal, Hannibal’s brilliantly opportunistic cavalry commander, was apparently one who thought so. Livy tells us (22.51.1–4) that sometime after the battle, amidst the congratulations of the Barcid’s henchmen, Maharbal warned that no time was to be lost, and held out instead the prospect of dining in the enemy capital within five days. “Follow me: I will go first with the cavalry, that the Romans may know that you are there, before they know you are coming!” It was the most audacious of proposals. March on Rome! Finish it now! When Hannibal hedged and refused to make an immediate decision, Maharbal’s reply was equally impulsive: “So the gods haven’t given everything to one man; you know how to win a victory, Hannibal, but you don’t know how to use one.”
Assuming the incident actually took place—it was very much characteristic of both men: Maharbal seizing the main chance, and Hannibal the gambler growing cautious in the face of overwhelming good luck—it strikes at the heart of Punic prospects and is therefore hotly debated. On balance, scholarly opinion seems to support Hannibal for not trying it. Some argue that he was short on pack animals and the logistical support needed to move his army 250 miles to Rome with the necessary alacrity. (This argument seems odd, given the journey from Spain and the Carthaginians’ tromp through the swamp.) Other scholars maintain that even if he had gotten to Rome, he couldn’t have done much productive, and he lacked siege equipment. (He could have built some.) Still others are of the opinion that Hannibal was better off trying to break the Roman alliance, win on a solid base of support, and then negotiate. (We shall see how that worked out.) In fact there were many good reasons for not marching on Rome, and only one good reason for going.
Unlike the scholars, Field Marshal Bernard Montgomery, a soldier, thought Maharbal was right. Maharbal seems to have understood that when a more powerful adversary is down, it has to be dispatched. Rome still had huge manpower reserves; there was no such thing as a peace party; this was a state that bargained only with the defeated. Hannibal’s single chance of winning the larger war was to begin marching his army toward Rome. Even if it had taken him a month to get there, the tension in the city would have only built with reports of his coming. And ultimately his appearance outside the walls might have broken the spirit to resist, or might have led to Rome’s sending an ill-prepared force out to another catastrophic defeat and ultimate capitulation. Or not. In the end it still would have been a long shot. But it was his only shot. Instead, Hannibal chose another route, and the war became only a matter of time.
Still, if Rome was not about to collapse, there were certainly cracks … and not just in the façade. At Canusium it seems things almost reached the point of falling apart. The survivors were treated kindly by the locals, especially a wealthy woman named Busa, who gave them food and fresh clothing. Among the survivors were four military tribunes (for some reason Tuditanus is not mentioned): Lucius Bibulus; Quintus, the son of the former dictator Fabius Maximus; Appius Claudius; and Publius Scipio, who, despite being the youngest of the group, emerged as its dominant personality in what was shortly to become a crisis situation.
The episode began when a reliable source informed them that within the group of survivors, a cabal of young nobles led by M. Caecilius Metellus and P. Furius Philus, whose father had shared the consulship with Flaminius in 224, were ready to give up on Rome, abandon Italy, and become mercenaries abroad. When the other tribunes agreed to form a council to discuss this stupefying piece of intelligence, Scipio would have none of it, demanding immediate action. Leading a few followers, he burst in on the conspirators with drawn sword and took them into custody, but not before demanding on pain of death that they swear an oath of allegiance to the state.
The nascent mutiny put down, Scipio and Appius learned of Varro’s presence in Venusia and sent him a message asking if the consul wanted them to deliver their forces to him or wanted them to remain in Canusium. Varro promptly marched his own troops over to them. This may be significant.
Besides the obvious motive of concentrating forces, the reasons for Varro’s decision and the later treatment of the legiones Cannenses could be connected. Had the conspiracy gone beyond the young nobles and extended to the troops, making it necessary for Varro to reach Canusium quickly, in order to stabilize the situation? Or alternatively, did the errant consul rescue his own reputation at the men’s expense, by giving the appearance of restoring order where there had not necessarily been disorder? It’s impossible to know. What is known is that Varro, having exited the battlefield at a full gallop, was later remarkably well treated by his countrymen, while the Cannenses, who left under rather more duress, were effectively banished. Rank certainly has its privileges, but the contradictions here are hard to ignore or explain.
Back in Rome the city’s population was on the ragged edge of panic, “expecting Hannibal every moment to appear,” Polybius (3.118) tells us shortly before he effectively signs off, the remainder of his description of the war surviving only in fragments. For better or worse Livy becomes our primary oracle, framing subsequent events with a dramatist’s eye.
Accordingly, Rome’s streets are described as echoing the wailing of lamenting women, because the initial reports indicated that the Cannae army had been crushed and there were no survivors. The senate met to take measure of the situation, with Fabius Maximus arguing for gathering more intelligence, sending the women indoors, and preventing anyone from leaving the city. It was only after a letter arrived from Varro verifying the disaster—but adding that he was with ten thousand survivors at Canusium and that Hannibal was still at Cannae not doing much of anything—that the cloud of terror began to dissipate and enough traction was gained for the senate to begin serious planning. What emerged was a characteristic combination of superstition, practicality, and adamantine stubbornness.
Existentially, beating back the dread and propitiating the gods called for extraordinary—what we would call barbaric—measures. Perhaps conveniently, two of the vestal virgins were found not to be so. One of the two women committed suicide before she could be buried alive with the other, while the seducer was beaten to death by the pontifex maximus, the chief priest. Meanwhile, the priest’s colleagues were consulting the Sibylline Books for other goddess-calming measures, and found the answer in more live interments—this time two couples, Greek and Celtic, male and female. And if human sacrifice did not prove sufficient, the city fathers thought to send fellow senator and historian Fabius Pictor to Greece to consult the Delphic oracle for more ideas on atonement. In a further attempt to restore emotional equilibrium, the senate officially limited mourning to thirty days, but the senate still had to call off the annual festival to the goddess of the harvest, Ceres, since the rituals required married women who weren’t in mourning, and few were available.79 Rome did regain its composure, but these steps, plainly meant to be viewed as extreme, reflect the degree to which the news of Cannae had shaken the inhabitants.
Yet beneath the veil of ritualistic excess, the senate remained clearheaded, making the leadership and personnel decisions necessary to deal with the immediate crisis and to restore Rome’s capacity to defend itself. Almost immediately the stalwart Marcellus, apparently no longer in Sicily, was sent to Canusium, where he would reorganize the Cannenses and put them back in fighting shape, while Varro was sent specific instructions to return home, possibly to nominate a dictator. (Upon arrival he would be greeted rapturously “for not having despaired of the Republic.” Acidly, Livy reminds us that a Carthaginian arriving back in Carthage under similar circumstances would have been punished with the utmost rigor—that is, he would have been crucified.) The man chosen as dictator was very experienced—M. Junius Pera, a former consul and censor, with the highly capable Tiberius Sempronius Gracchus acting as master of horse. Together they set about rebuilding Rome’s force structure.
In time there would be a huge manpower pool available, but in just two years Hannibal had killed at least a hundred thousand of Rome’s soldiers, and the recruiters on the Tiber behaved as if they were more than a little shorthanded. Just one thousand new cavalry could be raised, a number reflecting Hannibal’s prodigious attrition of the equestrians. To levy more foot soldiers for two new legiones urbanae, the draft age was lowered and boys of seventeen or even younger were called up, along with reinforcements from the Latin allies. Yet more telling was the enlistment of six thousand criminals and debtors, who had to be equipped with the Gallic arms taken by Flaminius for his triumph in 223. Finally, and most significant, the city’s slaves of fighting age were promised freedom upon discharge if they were willing to join the war effort, a call that was answered by eight thousand of them, who were subsequently known as volones, or volunteers. Their owners were compensated with state funds, the cost of which, Livy (22.57.11–12) notes ominously, exceeded the amount that would have been required to ransom the prisoners held by Hannibal.
Back in Apulia, the Carthaginian was in an avuncular mood. As he had after Trebia and Trasimene, he let the allies he held go free, yet again professing his goodwill. He then turned to the Roman captives and sought to explain himself, which was something new. He was not pursuing a war to the death with Rome, he explained; he was fighting “for honor and empire.” Just as his Carthaginian predecessors had yielded to the success of Roman arms, now it was time for Rome to accept defeat in the face of his own skill and good fortune. It was a speech that might have been given by Pyrrhus or any other Hellenistic monarch, a perfectly reasonable speech. The rules of war as he saw them dictated that, after such a string of beatings, the vanquished, presuming they were in their right minds, admit defeat. That was the way the “great game” of the Mediterranean basin was played; it was time Rome got used to it. He was prepared to be generous. The captives were to be ransomed for a reasonable price; ten of their own number would be sent to Rome to work out the details. Carthalo, a Carthaginian cavalry officer, would accompany them to present Punic peace terms. It’s impossible to know if Hannibal really expected his initiative to work, but it seems unlikely that he anticipated the reception that the delegation actually got.
As the group approached the city, the senate had the dictator, Pera, send a lictor to meet them and inform Carthalo that he would not be received and that he had to leave Roman territory by nightfall. So much for a negotiated peace. There was some sympathy for the captives, but not enough. In the speech the leader of the prisoners gave to the senate, Livy (22.59) has him argue that ransoming their number would be cheaper than purchasing the previously mentioned volunteer slaves, and comparing themselves favorably with those who took refuge at Venusia and Canusium, “men who left their swords on the field and fled.”
These pleas fell on deaf ears, especially those of T. Manlius Torquatus. He delivered a savage rebuttal. Although he did concede that the troops at Canusium were better judges of courage and cowardice than the captives, he revealed little regard for either group. The negligence of the captives was twofold: first, “they fled to the camp when it was their duty to stand firm and fight,” and second, they surrendered the camp. It was left unsaid but still implied that all those who had left the battlefield, captives and escapees alike, had violated the oath administered before the battle never to break rank except in the pursuit of duty.
The point, for the moment at least, was that the captives were not to be ransomed. The senate even went so far as to forbid their families to raise money privately to free them. This plainly went against precedent; just the year before, Fabius had paid prisoners’ ransoms with the proceeds from the farm Hannibal had left untouched. The Roman leadership wanted to send a message not just to its own soldiers, but to Hannibal, to shock him with the degree of their determination.85 Whatever he might think, in their eyes this was a fight to the finish.
When the delegation reluctantly returned with the bad news, Hannibal’s mood—though not necessarily his strategy—hardened. Appian (Han.28) maintains that Hannibal had those of senatorial rank fight as gladiators for the amusement of the Africans; some were slaughtered; the rest were sold into slavery. The last, at least, we know was true. Polybius in a fragment (6.58.13) reports that Hannibal lost his joy over the victory at Cannae; he now knew he was in for a long fight.
But as hardheaded and hard-hearted as was the image presented along the Tiber, the Roman leadership still had to work within its means. After Cannae much of southern Italy was leaning toward Hannibal, and Rome needed a presence to fend off the momentum toward the Punic side. The new legiones urbanae and the scratch force of slaves and criminals were not yet trained. The only trained men were the Cannenses.
We next hear of Marcellus in the autumn of 216, first at Casilinum, then at Nola, parrying Hannibal’s thrusts at the latter town with his army of survivors. Livy conflates what was probably a series of desultory skirmishes into a tactical victory featuring a surprise sortie out of the city gates, but even he questions the number of losses inflicted on the Punic force. It was not much in the way of revenge. Still, the Cannenses, now divided into two legions, showed themselves to be once again an effective fighting force and one ready to take the field against its nemesis. If nothing else, the men had amply demonstrated their loyalty to the state. Yet they were not forgiven, even in the face of further disaster.
As the terrible year 216 came to an end, Rome settled down for a change of leadership. But no sooner had L. Postumius Albinus—who had been sent in the spring to Cisalpine Gaul with two legions plus allies to break the rebellious Celts—been elected in absentia to his third consulship than news filtered into the city that he and his entire army had been ambushed and annihilated. To add insult to injury, the victorious Boii beheaded the fallen consul-elect, hollowed out his skull, and subsequently used it as a drinking cup. But more to the point, Rome was down another twenty-five thousand troops. It did not matter; by December of 216, the new legiones urbanae were ready and were given to Marcellus.
No longer needed, the Cannenses would now get what they deserved, at least in the eyes of the senate. Taxes had been doubled so that all soldiers could be paid in cash immediately, except for those who’d fought at Cannae. They got nothing. But this was secondary compared to being shipped to Sicily. Here they would stay until 204, removed from their families and their livelihood, effectively banished. It was a terrible punishment, inflicted upon them because they were seen as having broken an oath never before required, which had made them, technically at least, deserters. Rome had lost a great battle and needed a scapegoat. Rather than blame the strategists and commanders who had planned it, the powers that be turned on the survivors. The logic, the same as for decimation (“pour encourager les autres”), might have made sense at the time. But these ghosts of Cannae would live to haunt the republic. For one day, legionaries would look to their generals and not Rome for a future, and that perspective would spell civil war and absolute rule. This more than anything else was the battle’s legacy.
As far back as the reign of Shamashi-Adad (1813-1791 BCE), the Assyrian military had shown itself an effective fighting force. In the period known as the Middle Empire, kings like Ashur-Uballit I (1353-1318 BCE) were employing the army with great efficacy in the conquest of the region of the Mitanni and the king Adad Nirari I (1307-1275 BCE) expanded the empire through military conquest and crushed internal rebellions swiftly.
Adad Nirari I completely conquered the Mitanni and began what would become standard policy under the Assyrian Empire: the deportation of large segments of the population. With Mitanni under Assyrian control, Adad Nirari I decided the best way to prevent any future uprising was to remove the former occupants of the land and replace them with Assyrians. This should not be understood, however, as a cruel treatment of captives. Writing on this, the historian Karen Radner states:
The deportees, their labour and their abilities were extremely valuable to the Assyrian state, and their relocation was carefully planned and organised. We must not imagine treks of destitute fugitives who were easy prey for famine and disease: the deportees were meant to travel as comfortably and safely as possible in order to reach their destination in good physical shape. Whenever deportations are depicted in Assyrian imperial art, men, women and children are shown travelling in groups, often riding on vehicles or animals and never in bonds. There is no reason to doubt these depictions as Assyrian narrative art does not otherwise shy away from the graphic display of extreme violence.
Deportees were carefully chosen for their abilities and sent to regions which could make the most of their talents. Not everyone in the conquered populace was chosen for deportation and families were never separated. Those segments of the population that had actively resisted the Assyrians were killed or sold into slavery, but the general populace became absorbed into the growing empire and they were thought of as Assyrians. This policy would be followed by the kings who succeeded Adad Nirari I until the collapse of the Assyrian Empire in 612 BCE.
During their zenith from the 10th century BC to 7th century BC, the Assyrians controlled an enormous territory that extended from the borders of Egypt to the eastern highlands of Iran. Many historians perceive Assyria to be among the first ‘superpowers’ of the ancient world. In a brilliant insight by historian Mark Healy, quite paradoxically, the rise of Assyrian militarism and imperialism (from 15th century BC) mirrored their land’s initial vulnerability, as it laid inside the rough triangle defined between the cities of Nineveh, Ashur and Arbil (all in northern Mesopotamia).
Simply put, this terrain rich in its plump grain-lands was open to plunder from most sides, with potential risks being posed by the nomadic tribes, hill folks, and even proximate competing powers. This, in turn, affected a reactionary measure in the Assyrian society – that led to the development of an effective and well organized military system that could cope with the constant state of aggression, conflicts, and raids (much like the Romans).
Such an intrinsic scope of the military being tied with the economic well-being of a state resulted in what can be called a domino effect. So in a sense, while the Assyrians formulated their ‘attack is the best defense’ strategies, the proximate states became more war-like, thus adding to the list of enemies for the Assyrian army to conquer. Consequently, when the Assyrians went on a war footing, their military was able to absorb more ideas from foreign powers, which led to an ambit of evolution and flexibility (again much like the later Romans). These tendencies of flexibility, discipline and incredible fighting skills became the hallmark of the Assyrian army that triumphed over most of the powerful Mesopotamian kingdoms in Asia by 8th century BC.
The scope of Assyrian military development and expansion was intrinsically tied to the economic prosperity of the rising realm. In the period circa 1450 BC, such a military system was required for actually protecting the vulnerability of the Assyrian lands locked between the powerful Mesopotamian states of Mitanni in the north and Babylonia at the south, which supported the land’s economic stability. But as the centuries went by, Assyria morphed into the aggressor with the aid of its continually developing military prowess.
Suffice it to say, more conquered lands brought more plunder in the form of various valuable resources, ranging from metals, horses to skill-based populations. This was complemented by the control of crucial trade routes that crisscrossed through various parts of Mesopotamia. In essence, waging wars (along with conquering and raiding) became organized ventures conducted for the betterment of the Assyrian realm’s economy. So simply put, by 11th century BC, the security needs of the state became indistinguishable from the prosperity of the ascending empire – with the Assyrian army playing its crucial role in both the ‘merged’ affairs.
With the last king of the dynasty of Khammurabi (about 2098 B.C.) a period of darkness falls upon the history of the land between the rivers. A new dynasty of the Babylonian kings’ list begins with a certain Anmanu, and continues with ten other kings whose names are anything but suggestive of Babylonian origin. The regnal years of the eleven reach the respectable number of three hundred and sixty-eight. The problem of their origin is complicated with that of deciphering the word (Uru-azagga?) descriptive of them in the kings’ list. Some think that it points to a quarter of the city of Babylon. Others, reading it Uru-ku, see in it the name of the ancient city of Uruk. The length of the reigns of the several kings is above the average, and suggests peace and prosperity under their rule. It is certainly strange in that case that no memorials of them have as yet been discovered, — a fact that lends some plausibility to the theory maintained by Hommel that this dynasty was contemporaneous with that of Khammurabi and never attained significance.
The third dynasty, as recorded on the kings’ list, consists of thirty-six kings, who reigned five hundred seventy-six years and nine months (about 1717-1140 B.C.). About these kings information, while quite extensive, is yet so fragmentary as to render exact and organized presentation of their history exceedingly difficult. The kings’ list is badly broken in the middle of the dynasty, so that only the first six and the last eleven or twelve of the names are intact, leaving thirteen or fourteen to be otherwise supplied and the order of succession to be determined from imperfect and inconclusive data. Only one royal inscription of some length exists, that of a certain Agum-kakrime who does not appear on the dynastic list. The tablets found at Nippur by the University of Pennsylvania’s expedition have added several names to the list and thrown new light upon the history of the dynasty. The fragments of the so-called “Synchronistic History” (sect. 30) cover, in part, the relations of the Babylonian and Assyrian kings of this age, and the recently discovered royal Egyptian archives known as the Tel-el-Amarna tablets contain letters from and to several of them. From these materials it is possible to obtain the names of all but three or four of the missing thirteen or fourteen kings, and to reach something like a general knowledge of the whole period and some details of single reigns and epochs. Yet it is evident that the absence of some royal names not only makes the order of succession in the dark period uncertain, but throws its chronology into disorder. Nor is the material sufficient to remove the whole age from the region of indefiniteness as to the aims and achievements of the dynasty, or to make possible a grouping into epochs of development which may be above criticism. With these considerations in mind it is possible roughly to divide the period into four epochs: first, the beginnings of Kassite rule; second, the appearance of Assyria as a possible rival of Kassite Babylonia; third, the culmination of the dynasty and the struggle with Assyria; fourth, the decline and disappearance of the Kassites.
Merely a glance at the names in the dynastic list is evidence that a majority of them are of a non-Babylonian character. The royal inscriptions prove beyond doubt that the dynasty as a whole was foreign, and its domination the result of invasion by a people called Kashhus, or, to use a more conventional name, the Kassites. They belonged to the eastern mountains, occupying the high valleys from the borders of Elam northward, living partly from the scanty products of the soil and partly by plundering travellers and making descents upon the western plain. The few fragments of their language which survive are not sufficient to indicate its affinity either to the Elamite or the Median, and at present all that can be said is that they formed a greater or lesser division of that congeries of mountain peoples which, without unity or common name and language, surged back and forth over the mountain wall stretching from the Caspian Sea to the Persian gulf. Their home seems to have been in the vicinity of those few mountain passes which lead from the valley up to the table-land. Hence they were brought into closer relations with the trade and commerce which from time immemorial had used these passes, and thereby they were early made aware of the civilization and wealth of Babylonia.
Whether driven by the impulse to conquest, begotten of a growing knowledge of Babylonian weakness, or by the pressure of peoples behind and about them, the Kassites appear at an early day to have figured in the annals of the Babylonian kingdom. In the ninth year of Samsuiluna, of the first dynasty, they were invading the land. This doubtless isolated invasion was repeated in the following years until by the beginning of the seventeenth century B.C., they seem to have gained the upper hand in Babylonia. Their earlier field of operations seems to have been in the south, near the mouth of the rivers. Here was Karduniash, the home of the Kassites in Babylonia, a name subsequently extended over all the land. It is not improbable that a Kassite tribe settled here in the last days of the second dynasty, and, assimilated to the civilization of the land, was later reinforced by larger bands of the same people displaced from the original home of the Kassites by pressure from behind, and that the combined forces found it easy to overspread and gain possession of the whole country. Such a supposition is in harmony with the evident predilection of the Kassites for southern Babylonia, as well as with their maintenance of authority over the regions in which they originally had their home. It also explains how, very soon after they came to power, they were hardly to be distinguished from the Semitic Babylonians over whom they ruled. They employed the royal titles, worshipped at the ancient shrines, served the native gods, and wrote their inscriptions in the Babylonian language.
Of the six kings whose names appear first on the dynastic list nothing of historical importance is known. The gap that ensues in that list, covering thirteen or fourteen names, is filled up from sources to which reference has already been made. Agumkakrime (sect. 103), whose inscription of three hundred and thirty-eight lines is the most important Kassite document as yet discovered, probably stands near the early kings, is perhaps the seventh in order (about 1600 B.C.). This inscription, preserved in an Assyrian copy, was originally deposited in the temple at Babylon, and describes the royal achievements on behalf of the god Marduk and his divine spouse Zarpanit. The king first proclaims his own glory by reciting his genealogy, his relation to the gods and his royal titles:
I am Agumkakrime, the son of Tashshigurumash; the illustrious descendant of god Shuqamuna; called by Anu and Bel, Ea and Marduk, Sin and Shamash; the powerful hero of Ishtar, the warrior among the goddesses.
I am a king of wisdom and prudence; a king who grants hearing and pardon; the son of Tashshigurumash; the descendant of Abirumash, the crafty warrior; the first son among the numerous family of the great Agum; an illustrious, royal scion who holds the reins of the nation (and is) a mighty shepherd…
I am king of the country of Kashshu and of the Akkadians; king of the wide country of Babylon, who settles the numerous people in Ashnunak; the King of Padan and Alman; the King of Gutium, a foolish nation; (a king) who makes obedient to him the four regions, and has always been a favorite of the great gods (I. 1-42).
Agumkakrime found, on taking the throne, that the images of Marduk and Zarpanit, chief deities of the city, had been removed from the temple to the land of Khani, a region not yet definitely located, but presumably in northern Mesopotamia, and possibly on the head-waters of the Euphrates. This removal took place probably in connection with an invasion of peoples from that distant region, who were subsequently driven out; and it sheds light on the weakened and disordered condition of the land at the time of the appearance of the Kassites. These images were recovered by the king, either through an embassy or by force of arms. The inscription is indefinite on the point, but the wealth of the king as intimated in the latter part of the inscription would suggest that he was at least able to compel the surrender of them. On being recovered they were replaced in their temple, which was renovated and splendidly furnished for their reception. Gold and precious stones and woods were employed in lavish profusion for the adornment of the persons of the divine pair and the decoration of their abode. Their priesthoods were revived, the service re-established, and endowments provided for the temple.
In the countries enumerated by Agumkakrime as under his sway no mention is made of a people who were soon to exercise a commanding influence upon the history of the Kassite dynasty. The people of Assyria, however, although, even before that time, having a local habitation and rulers, the names of some of whom have come down in tradition, could hardly have been independent of a king who claimed authority over the land of the Keissites and the Guti, Padan, and Alman, — districts which lie in the region of the middle and upper Tigris, or on the slopes of the eastern mountains (Delitzsch, Peradies, p. 205). According to the report of the Synchronistic History, about a century and a half later Assyria was capable of treating with Babylonia on equal terms, but, even if the opening passages of that document (some eleven lines) had been preserved, they would hardly have indicated such relations at a much earlier date. The sudden rise of Assyria, therefore, is reasonably explained as connected with the greater movement which made the Kassites supreme in Babylonia.
The people who established the kingdom of Assyria exhibit, in language and customs and even in physical characteristics, a close likeness to the Babylonians. They were, therefore, not only a Semitic people, but, apparently, also of Semitic-BabyIonian stock. The most natural explanation of this fact is that they were originally a Babylonian colony. They seem, however, to be of even purer Semitic blood than their Babylonian ancestors, and some scholars have preferred to see in them an independent offshoot from the original Semitic migration into the Mesopotamian valley (sect. 51). If that be so, they must have come very early under Babylonian influence which dominated the essential elements of their civilization and its growth down to their latest days. The earliest centre of their organization was the city of Assur on the west bank of the middle Tigris (lat. n. 35° 30′), where a line of low hills begins to run southward along the river. Perched on the outlying northern spur of these hills, and by them sheltered from the nomads of the steppe and protected by the broad river in front from the raids of mountaineers of the east, the city was an outpost of Babylonian civilization and a station on the natural road of trade with the lands of the upper Tigris. A fertile stretch of alluvial soil in the vicinity supplied the necessary agricultural basis of life, while, a few miles to the north, bitumen springs furnished, as on the Euphrates, an article of commerce and an indispensable element of building (Layard, Nineveh and its Remains, II. chap. xii.). The god of the city was Ashur, “the good one,” and from him the city received its name (Jastrow, Rel. of Bab. and Assyria, p. 196).
The early rulers of the city of Assur were patesis (sect. 75), viceroys of Babylonian rulers. Some of their names have come down in tradition, as, for example, those of Ishme Dagan and his son, Shamshi Adad, who lived according to Tiglathpileser I. about seven hundred years before himself (that is, about 1840-1800 B.C.). Later kings of Assyria also refer to other rulers of the early age to whom they give the royal title, but of whom nothing further is known. The first mention of Assur is in a letter of king Khammurabi of the first dynasty of Babylon, who seems to intimate that the city was a part of the Babylonian Empire (King, Let. and Inscr. of H., III. p. 3). In the darkness that covers these beginnings, the viceroys became independent of Babylonia and extended their authority up the Tigris to Kalkhi, Arbela, and Nineveh, cities to be in the future centres of the Assyrian Empire. The kingdom of Assyria took form and gathered power.
The physical characteristics of this region could not but shape the activities of those who lived within its borders. It is the northeastern corner of Mesopotamia. The mountains rise in the rear; the Tigris and Mesopotamia are in front. The chief cities of Assyria, with the sole exception of Assur, lie to the east of the great river and on the narrow shelf between it and the northeastern mountain ranges. They who live there must needs find nature less friendly to them than to their brethren of the south. Agriculture does not richly reward their labors. They learn, by struggling with the wild beasts of the hills and the fierce men of the mountains, the thirst for battle and the joy of victory. And as they grow too numerous for their borders, the prospect, barred to the east and north, opens invitingly towards the west and southwest. Thus the Assyrian found in his surroundings the encouragement to devote himself to war and to the chase rather than to the peaceful pursuits of agriculture; the preparation for military achievement on a scale hitherto unrealized.
It is not difficult to conceive how the Kassite conquest of Babylonia profoundly influenced the development of Assyria. The city of Assur, protected from the inroads of the eastern invaders by its position on the west bank of the Tigris, became, at the same time, the refuge of those Babylonians who fled before the conquerors as they overspread the land. The Assyrian community was thus enabled to throw off the yoke of allegiance to the mother country, now in possession of foreigners, and to establish itself as an independent kingdom. Its patesis became kings, and began to cherish ambitions of recovering the home-land from the grasp of the enemy, and of extending their sway over the upper Tigris and beyond. It is not unlikely that this latter endeavor was at least partially successful during the early period of the Kassite rule. It is certainly significant that Agumkakrime does not mention Assyria among the districts under his sway and if, as has been remarked (sect. 108), his sphere of influence seems to include it, his successors were soon to learn that a new power must be reckoned with, in settling the question of supremacy on the middle Tigris.
The darkness of the Greek Dark Age – and indeed the meagreness of all our sources up until that point – are suddenly replaced by an overload of information, beginning as a trickle in the early eighth century and turning into a flood by the following century. The Dark Age produced the occasional warrior grave with sword, spear and perhaps shield. There was almost no figurative art. After 800 we start to get votive statues of bronze warriors. The Late Geometric period (after c. 750) produced battle scenes on vases, showing the equipment used, and this tradition in art continued throughout the rest of the period covered by this book, albeit sometimes fitfully. Finally, and most importantly of all, archaeology has produced much more evidence of military equipment, especially in the weapons and pieces of armour found in Greece’s sanctuaries, most notably Olympia. There are even some literary fragments to help us, although no major surviving historical works until the end of the period. Here we examine the emergence from darkness and follows the development of the equipment that changed the face of Greek combat from the skirmishing of individuals and small groups into the hoplite phalanx of the city-state.
The phalanx is a body of men in close order, standing shoulder to shoulder, and closing on the enemy with the thrusting spear. They are called hoplites by ancient Greek authors, after hopla, meaning ‘arms’ or after the name of the large round shield they used, which was sometimes called the hoplon. This is more usually – and confusingly – called the aspis. Missile weapons became much less common, while the soldiers became more heavily armoured. It was a gradual process and, as Snodgrass (1965a, passim) has shown, all the pieces of equipment for hoplite warfare made their appearance before hoplite warfare itself was known to be in existence, in around 675 to 650 BC.
Our earliest evidence for what warfare was like, rather than just what the equipment was, comes in the depictions on Late Geometric vases of Athens dating from c. 760 and later. Here there is no evidence for the hoplite phalanx, but all the signs of skirmishing warfare, which was assumed to be the norm throughout the Dark Age. Warriors are depicted in small groups or individual combats, carrying a variety of shield types and weapons. The composite bow is seen as often as the spear, and swords are also frequently depicted (Ahlberg 1971, p. 44). On the Corinthian aryballos (vase) we can see a hoplite in armour with a hoplite shield and spear, but he also carries another spear for throwing and is certainly not fighting in a phalanx with other hoplites. He is supported by an archer, and by one other whose weapon is unclear. His opponents also carry more than one spear, have helmets and perhaps body armour, but carry Dipylon shields (see below), and appear to be throwing spears rather than fighting in a phalanx.
As we shall see from the evidence outlined below, although all the equipment worn by later seventh-century hoplites was available from c. 700 or a little before, the hoplite phalanx did not appear in art until the Chigi Vase of c. 650. Here the men are shown fighting together in ranks with the thrusting spear, in a body known as the phalanx, which in later times was usually eight men deep. The heavy shield and the bronze armour, which some warriors had adopted for personal protection in the later eighth century, made it harder to run in any sort of skirmishing warfare, and it gradually became the rule to stand by your fellow warriors to avoid becoming isolated. This procedure then gradually developed into the phalanx, which became the mainstay of Greek warfare for the next 250 years, although there were subtle changes. Even on the Chigi Vase there is evidence of second spears (for throwing) still being carried.
Literary fragments also tell us about hoplite warfare. Tyrtaeus, a Spartan writing in the later seventh century, says: ‘Everyone should close up to his man with his great spear or sword and wound and kill his enemy. Standing leg to leg, resting shield against shield, crest beside crest, and helmet to helmet having drawn near, let him fight his man with his sword or great spear. And you, O light-armed fighters, crouching behind the shields on either side, hurl your great boulders’ (Tyrtaeus, frag. 11, in Sage 1996, pp. 28–9). The mention of light-armed troops is interesting. We have seen how common archery seems to have been, at least at Athens in the later eighth century, from the depictions on Geometric vases. The slow-moving phalanx would have been particularly vulnerable to missiles, but there is a great lack of evidence for archery and slingers in the seventh and sixth centuries. It seems that there was a continuing doubt about the suitability of killing an opponent from a distance, which we noted in Homer’s Iliad. It was not regarded as brave or heroic and, to help stamp it out, some city-states even made formal agreements not to use such weapons (Hanson 1999, p. 38). The idea that only close, hand-to-hand combat was ‘civilised’ probably led to the eventual abandonment of the second spear, which could be thrown, in the later seventh century. Hoplite warfare then settled down into an almost ritual battle between city-states, with unchanging equipment and tactics until the later sixth century.
There is no evidence for Greeks using bronze helmets following the Tiryns example of c. 1050 until the middle of the eighth century. At that point, bronze was imported into Greece again with the widening of trade, and bronze artefacts began to reappear in some numbers. Indeed, it seems that the earliest Greek ‘colonies’ at Pithecusae in Italy and Al Mina in Syria were trading posts founded particularly to import metals into Greece (Boardman 1980b, pp. 43–5). A series of bronze votive statuettes, mostly from Olympia, which show otherwise naked warriors wearing bronze waist belts and helmets of a simple cap form, is our earliest evidence (Snodgrass 1967, p. 42). After c. 750 we also have the art of the Dipylon Master on vases from Athens, and his contemporaries, showing warriors in combat or funeral processions (Ahlberg 1971, passim). Some of these have helmet crests, but the style of the art makes it difficult to interpret the helmets (Snodgrass 1965a, pp. 5–6). Securely dated to c. 725, however, is our earliest actual example of a post-Dark Age helmet, found in a tomb at Argos with a bronze cuirass and possibly greaves (Courbin 1957, passim). The helmet is known as a Kegelhelm, or Kegel helmet, from the German for cone, and it is clear that this is the sort of helmet portrayed on some of the earliest bronze statuettes.
The helmet is constructed in five pieces plus a large curving crest. There is a cap for the top of the head to which are attached a forehead guard, back of the head/neck guard, and two cheek pieces. All these have repoussé lines at the edges for extra strength. Other examples of Kegel helmets have been found at Olympia; these have a simple pointed cap, an attached crescent-shaped crest, or a curving forward crest made as one with the skull of the helmet (Dezso 1998, p. 31). These crest forms are clearly Assyrian and feature on friezes from Assyrian palaces (Snodgrass 1967, p. 43; Dezso 1998, pp. 31–3). The helmets are definitely of Greek manufacture, however, probably made after seeing Assyrian helmets in Cyprus or Asia Minor as trade routes opened up. An Assyrian helmet is also known from a grave at Argos (Dezso 1998). No Assyrian helmets are made in five pieces, and by this period they were starting to manufacture their helmets in iron anyway (Dezso 1998, p. 9). If we examine the other Kegelhelms we see that some examples, like the one in Budapest Museum of Fine Arts (Dezso 1998, figs 16–17), have a repoussé ridge halfway down the neck guard for extra strength, and a lower cap and broader cheek pieces. The plates on this helmet also overlap by more than the Argos helmet to give greater strength to the welded joints. This is obviously a late example, dating to perhaps c. 675, giving the Kegelhelm a lifespan of only about fifty years. The Argos helmet, dating from c. 725, is of much poorer quality and its high crest sits on a tube made of cast bronze, making it top heavy and precarious (Snodgrass 1965, pp. 14–15).
Returning to the crests on these helmets, Dezso (1998, pp. 35–6) lists a few helmets from Olympia with crescent crests, but from which the crest part is actually missing, so the final form must remain uncertain. Three examples with a forward-curving crest have been found in Italy but there must have also been Greek examples, as they are portrayed on some of the bronze statuettes and on a terracotta model shield from Tiryns dating from c. 700. Dezso (1998) shows that this crest, too, is derived from Assyria. The Kegel helmet had such a short life because its multi-piece construction made it very weak compared with other helmets that were being developed in Greece. The helmets were prone to break apart at the joins, as can be seen from the many ‘parts’ of helmets that have been retrieved from Olympia. The crests, especially when cast, also made the helmets difficult to wear; they must have had chin straps to keep them on. Crests were an important part of psychological warfare. They made you look and feel taller and more threatening to an opponent and, later on, there is also some evidence for their being used as badges of rank.
The first improvement on the Kegel helmet was the so-called Illyrian helmet, which was clearly derived from the Kegelhelm. It had a very similar outline, but was made initially in just two halves. These were joined from front to back and reinforced by two parallel repoussé lines, between which a crest would have sat. It is this design of helmet which probably led to the common fore-and-aft crest we find on most later Greek helmets, and to the phasing out of the more precarious ‘Assyrian’ crests, although raised crests did persist for some time. The Illyrian helmet was open-faced, like the Kegel, with cheek pieces that became more elongated over time, and a neck guard that got lower and more protective. The lengthened cheek pieces helped to protect the throat area, which was particularly vulnerable to spear thrusts above the breastplate; and a lower neck guard would protect against slashing swords. This helmet would probably not have needed a chin strap, as it had a lower centre of gravity. From the evidence we have today, it would seem that it was developed towards the end of the eighth century in the Peloponnese. It must have followed on from the Kegelhelm and cannot be placed much before 700 (Jackson 1999, p. 161). Later helmets were beaten out of a single piece of bronze for even greater strength and, by the end of the sixth century, some examples have cut-outs for the ears and even hinged cheek pieces to make a more comfortable design (Olympia Bericht VIII, pp. 66–71). Towards the end of the sixth century, the phalanx was becoming more lightly armed and more mobile. It seems that more commands were being used in the phalanx and, for that, the hoplites had to have their ears uncovered to be able to hear. The Illyrian seems to have been a popular helmet in the seventh and sixth centuries, especially in the Peloponnese. It may even have been a Spartan development. By the time of the Persian Wars it was in only occasional use in southern Greece, but remained very popular in northern Greece, Macedonia and Illyria: hence the name by which it is known today (Snodgrass 1965, p. 20).
Meanwhile, Crete and the islands developed their own ‘Insular’ helmet, which consisted of a simple cap brought down at the back and sides leaving an open face, and surmounted by a very tall crest. This was not cast like the Argos crest but made of two thin sheets of bronze, so it was much lighter. This crest was often adopted by Corinthian helmet wearers (see below). These helmets were also made in two halves, like most Illyrian ones, and are best represented by some miniature votive helmets found at Praisos in Greece. There is also a fragmentary, full-sized example now in Hamburg Museum, and published by Hoffman and Raubitschek (1972, pp. 5–6). This too appears to have had some Assyrian influence. Like the Argos helmet, it must have been somewhat top-heavy to wear, and this sort of crest seems to have disappeared soon after 650, perhaps because of the adoption of the hoplite phalanx.
Also popular in the islands and on the coast of Asia Minor was the ‘Ionian’ helmet, which features in many helmet-shaped vases of the seventh century and later. It had an offset neck guard and separate hinged cheek pieces with scalloped edges (Snodgrass 1967, p. 65 and plate 25). Some examples of the cheek pieces have been found at Olympia, and examples are shown on the Siphnian treasury frieze at Delphi (Olympia Bericht VII). Similar cheek pieces were worn with Assyrian-style conical helmets on the island of Cyprus (Pflug 1989, p. 10; Dezso 1998), and it is tempting to think of them as a Mycenaean survival, although there is no evidence for the scalloped cheek piece between the Tiryns helmet of c. 1050 and Cypriot helmets of c. 700. Cyprus was heavily influenced by Assyria in the ninth and eighth centuries, and by 709 was incorporated into the Assyrian Empire (Boardman 1980b, pp. 38, 44), although the population remained largely Greek.
By far the most popular helmet in Greece was the one known today as the Corinthian. There was a helmet known to the ancient Greeks as Corinthian (Herodotus IV, 180), and since the helmet about to be described appeared first and most frequently on Corinthian vases, most scholars agree that it is one and the same with Herodotus’s Corinthian helmet (Snodgrass 1967, p. 51). This is the most commonly found helmet in Greece, and lasted from the late eighth century well into the fifth century. Apart from Greece and Magna Graecia (Italy and Sicily), examples have been found on the Black Sea coast, in North Illyria and in southern Spain. Its popularity stems from its almost complete protection for the head and its (usually) one-piece construction, which made it very strong. The Corinthian helmet is a much more competent piece of work than the Kegelhelm or Illyrian types. It covers the entire head, leaving just a T-shaped slot for the eyes, nose and mouth. A problem with this, of course, is that it must have made seeing, and especially hearing, difficult. The introduction of this helmet may have persuaded soldiers to keep close together so they knew what was going on, which in turn helped lead to the close combat of the phalanx. The helmet first appears on early bronze statues (Snodgrass 1967, plate 15) and votive miniature helmets from Praisos, and it is clear that the earliest examples of the type date to c. 700 or shortly before. The earliest full-sized examples known make few concessions to anatomy or to comfort (Connolly 1998, p. 61). They are taller than they are broad, with perhaps a slight neck guard. Some are made in two halves, like Illyrian helmets, and have the same crest holders (Pflug 1989, p. 14; Sekunda 2000, p. 11).
By c. 680 the helmets had become lower, with more of a neck guard, and indentations to help the helmet sit on the shoulders better (Snodgrass 1965, p. 23). This may have come about with the more frequent use of bronze body armour. The helmets also had nose guards by this time. The earliest Corinthian helmets had a row of holes around the edge for attaching a lining and the padding that would need to be worn with the helmet, but later Corinthian examples simply had the bronze edges rolled over, or thickened, for extra strength. The other designs of helmet mentioned seldom had these edge holes, and it is likely that padding and linings were glued in. There appears to be no pictorial evidence from this time for a separate ‘arming cap’ to have been worn under the helmet.
Crests on the Corinthian helmets were front to back like the Illyrian ones, although they could also be mounted on a raised stilt like the warrior with the hoplite shield on the proto-Corinthian aryballos. Lower stilts still featured in art as late as 600. Examples of these stilts have been found at Olympia (Olympia Bericht VIII, plate 31.1), and may have remained popular until the Persian Wars.
There are even some vase pictures of the mid- to late sixth century showing two stilted crests on either side of the same helmet, or other even more elaborate creations, some of which may be fanciful (Boardman 1980a; 1998). A transverse crest is also occasionally seen, such as that on the well-known bronze statue of a Spartan warrior wrapped in his cloak (Sekunda 1999, pp. 10–11). A bronze of a warrior on horseback in the British Museum also originally had a transverse crest. It has been suggested that this may have been a form of rank recognition, like the Roman centurion’s crest, but no one knows for sure. It was perhaps just a personal choice. In the course of the late seventh and sixth centuries, the Corinthian helmet developed elaborately decorated borders, and by c. 530 the whole crown was raised up with a repoussé ridge for greater strength and to allow for more padding). The cheek pieces followed the Illyrian style of becoming longer and more pointed.
The rounded form of the Corinthian helmet, especially by the later sixth century, was a great asset in the deflection of missiles and spear thrusts, and the padding underneath would have helped to absorb heavier blows. A crest raised on a stilt could be top-heavy and was also perhaps liable to be seized by an opponent in combat, which is why the lower fore-and-aft crests gradually became more popular. A crest was certainly considered essential at this time, and all helmets had crests until the fifth century.
The Cretans seem to have adopted the Corinthian helmet at an early stage, when it was still made in two halves and had no nose guard. They then went on to develop their own form throughout the next two centuries. This helmet has no indent at the shoulders, no nose guard, and usually an extra bronze forehead guard attached to the brow. The cheek pieces are usually scalloped to leave a proper gap for the mouth (Hoffman and Raubitschek 1972, plates 1–12), but occasionally have the straight front edge of the standard Corinthian helmet. The factor that particularly distinguishes the Cretan helmets from other Greek helmets is their heavy decoration.
Hoffman and Raubitschek (1972, Helmet 3, plate 11) have published details of four helmets from Afrati, one of which is plain except for an embossed and incised decorated lower border, and incised rosettes on the cheek pieces. Another has embossed human ears on the sides of the helmet, while the remaining two have very elaborate pictures of men and horses embossed around the helmet, with further detailed decoration inscribed (Hoffman and Raubitschek 1972, Helmet 4, Helmet 1 and Helmet 2, plates 1–10, 12). They also publish a helmet from Axos and some further fragments. The Axos helmet has embossed winged pegasi on either side and incised border decoration. Its forehead guard is also intact and is decorated with embossed and incised rosettes and dragon heads (Hoffman and Raubitschek 1972, plates 14–15). These helmets must have looked stunning when first made, and it is possible that the decoration was enhanced by painting. These Cretan helmets all date from just before 650 until the end of the seventh century, and the fashion probably continued through the sixth century because we have other pieces of decorated Cretan armour with later dates.
Evidence for helmet painting and decorating in mainland Greece at this time comes mainly from vase paintings. Early examples of the Corinthian helmet in art (which accounts for 90 per cent of all helmet illustrations in the seventh and sixth centuries) are generally painted in the natural bronze colour, as on the Chigi Vase of c. 650. Where there are helmets in a different colour, it is not really safe to assume that they are necessarily painted, because of the limited number of colours available – especially when dealing with black figure vases. However, from about 650 we start to come across vases where the helmets on different warriors are painted in different colours on the same vase; here I think it is safe to assume that helmet painting, or perhaps even tinning or gilding, is taking place (Boardman 1980a; 1998). By c. 530 two colours start to appear on the same helmet (Boardman 1980a). Two Illyrian helmets with gold-leaf borders around the face openings have been discovered in Macedonia, dating from c. 525. However, the fact that another helmet has been found with a gold face-mask seems to suggest that this was applied as a funereal addition, and was not present on the helmet during its practical lifetime (Vokotopoulou 1995, pp. 116–17, 122–3). From c. 530 a general lightening of armour began, and new helmets such as the Chalcidian and ‘Attic’ were introduced, along with more elaborate decorative techniques.
THE BRONZE BELL CUIRASS
Although the Greeks seem to have developed their bronze helmets through copying Assyrian examples, they did not – with some exceptions (see below) – adopt Assyrian body armour made of bronze or iron scales. The first certain evidence for body armour after the Dark Age is bronze plate armour in the form of the Argos cuirass, which was found with a crested Kegelhelm and is dated to c. 720.
Since the Greeks certainly used bronze plate armour in the Late Mycenaean period, it might be thought that they used bronze armour right through, especially given the proximity of Argos to Dendra. It showed that there was an almost complete lack of bronze in Dark Age Greece, to the extent that items such as fibulae were made in iron instead; but they began to be made in bronze again when it became available in the ninth century. In Merhart’s mammoth 1969 work, Hallstatt und Italien, the author argues that the Greek cuirass must have been descended from the cuirasses of the Urnfield culture, such as Čaka and Ducové. We certainly know that there was contact between these cultures in the Late Mycenaean period, when the Greeks adopted the Naue II sword from there. But the Urnfield-Type culture does not have such a securely dated pottery sequence as Greece, and even those finds that are in context can be only approximately dated. It seems that, with the exception of Čaka and Ducové in the early twelfth century, no European cuirass can be placed before 800 BC, making their reappearance very similar to the Greek model.
European corslets come from two separate groups called West Alpine and South East Alpine by Merhart. These two groups are quite separate in design, and both have Greek and non-Greek features. The South East Alpine Group from Austria and former Yugoslavia is represented by the cuirasses from Vorlagen, St Veit, Saint-Germain-du-Plain and a now-lost example originally in the Mecklenburg Collection (Merhart 1969, p. 162, plate III). These corslets have a clear bell shape, like the Argos cuirass, and are similarly decorated, except for the Saint-Germain-du-Plain one, which is closer to the Ducové example (Snodgrass 1971, p. 39). Unlike the Argos cuirass, though, these corslets were fastened by simple bronze rings for leather straps, or by long tubes through which wire was threaded. They almost certainly date from the sixth or fifth centuries BC and so derive their advanced shape from the Greek cuirasses, while using simpler fastenings.
The cuirasses of the West Alpine Group are from Switzerland and Italy, although the find locations of the two ‘Italian’ cuirasses are very uncertain. These have been used by Merhart and Snodgrass to suggest that the Greek bell cuirass was derived from them, despite the dissimilarities. The Greeks founded Pithecusae as a trading post/colony in the early eighth century BC (Boardman 1980b, p. 165) and Snodgrass regards this as the conduit through which knowledge of bronze cuirasses travelled. When we look at the Argos cuirass in detail, however (see below), we see that it is an advanced piece of work. Since it dates from c. 725, then the Greeks must have had simpler cuirasses earlier in the century, perhaps even before Pithecusae was founded. The West Alpine cuirasses are short and without the bell flange curving out at the bottom; they are also decorated in a completely different fashion from the others mentioned, being covered with heavy embossing and bronze studs. Their only similarity to the Argos and later Greek cuirasses is that they are generally fastened with hinges and loops. Merhart dated these cuirasses by typology to c. 800 at the earliest, but Snodgrass thinks they are too crude and dates them later, bringing them even closer to the Argos model. He still thinks that Argos and other Greek cuirasses must have been derived from them, despite the technical superiority of the former (Snodgrass 1965, p. 82; 1971, p. 35). It seems much more likely to me that the exchange was the other way around. If that is so, then when and how did they rediscover the art? Let us examine the Argos cuirass in detail.
Like the European cuirasses, the Argos cuirass is of beaten bronze, with the edges at the bottom, neck and armholes rolled over wire. The bronze is about 2mm thick, rather thicker than later examples, and the edging wire is of iron, whereas it is usually of bronze later on. Also, for extra strength, there is a repoussé line parallel to each armhole and two lines parallel to the lower edge. A further two embossed repoussé lines run around the middle of the corslet in imitation of a belt. They would have added much-needed reinforcement in this expanse of undecorated bronze. On the breastplate the repoussé lines stop 1.6mm short of the edge at the shoulders, and it is clear that the backplate overlapped the breastplate here (Courbin 1957, p. 342). A hole in the middle of each shoulder piece on the backplate slotted over a corresponding pin on the breastplate. This is a simple fastening, but the fastenings at the sides display a high level of technical achievement, the like of which has not appeared so far in any of the Central European finds. The relief lines on the backplate at the sides stop some 4cm short of the front edge, and so here it is clear that the breastplate overlapped the backplate. The right side of the breastplate has two hinge tubes which are pushed through slots in the backplate and a hinge pin is then inserted on the inside, thus protecting it from damage in battle. The left side of the breastplate has a small bronze loop on the inside of the corslet near to the lower edge, which matches a corresponding loop on the outside of the backplate. These loops would have been fastened together by a piece of bronze wire or a leather thong, and the resultant join would have been covered by the breastplate. To make the join firmer, the piece of rolled-over bronze on the bottom edge of the backplate on this side is wider and lacks the iron-wire core for the last few centimetres, so that the bottom edge of the breastplate could slide into the channel thus made. This same slot technique is used on the lower edge of the left armhole as well (Courbin 1957, p. 344).
Courbin suspects that the wearer would have needed a fellow warrior or squire to help him put on this cuirass, and I think that is quite right. The right side of the cuirass would have had to be fastened before it was put on, because of the internal pins. Another man would then have had to put it on the warrior’s torso and fasten the shoulders, while also slotting the breastplate into the grooves at the armpit and lower edge, and then tying the two bronze loops together before the two halves of the cuirass were fully closed. Such a complicated system might even have required two or three assistants to accomplish it. The corslet was lined with linen, presumably padded, which was probably glued to the inside of the armour rather than stitched through tiny holes as Courbin thought (1957, p. 350).
Such a system was not in use among contemporary European corslets, where their simple loop and wire attachments would have been very susceptible to battle damage, and so the Archaic Greeks must have learned it from elsewhere. No one else used bronze plate armour, so it must have been an indigenous invention. Going back to the Dendra panoply of 700 years earlier, we remember that it possessed a primitive hinge and very similar shoulder fastenings, and some of the pieces of armour had rolled-over edges (though not over wire). As has been mentioned, it is very unlikely that armour use continued through the Dark Age, and even less likely that knowledge of armour making survived. The simplest explanation is that the idea for bronze plate armour and its method of manufacture came from Mycenaean tombs themselves, an idea first suggested to me by Dr A.H. Jackson, formerly of Manchester University. There is much evidence that later Greeks made shrines for ‘Hero Worship’ at several Mycenaean tombs, and they would have known about the bronze panoplies worn by ancient warriors from these tombs, as well as from the memories preserved by Homer (Whitley 1988, passim). Apart from this evidence there is also the story of the ‘Bones of Theseus’, which were recovered from a presumably Mycenaean tomb by Cimon on Skyros in 476–5 (Plutarch, Life of Cimon, 10; Life of Theseus, 36). So, the reintroduction of bronze plate armour in the eighth century was almost certainly a Greek phenomenon. The fastening systems were Greek and the decoration was almost entirely of a new style, which the Central Europeans were themselves to borrow.
This decoration, apart from the repoussé strengthening lines already mentioned, consisted of the marking out of basic anatomy on the cuirass. Repoussé lines mark out the shoulder blades and breast muscles and also the lines of the ribcage, known as the omega curve. An indented line marks the linea alba (the line from the sternum downwards) and another the spinal furrow on the backplate. The only non-anatomical decoration other than the repoussé lines is a line of 3.5mm-diameter circles around the lower edge of the cuirass, incised with a compass (Courbin 1957, p. 342). In his thesis ‘Bronze Armour and the Earliest Greek Kouroi’, Hartmann shows that these anatomical features are indigenous to Greece and appear on seventh-century kouroi (statues) as a result of the Greek armour-making tradition; they do not come from Egypt, whence came the idea of stone statuary. He describes several kouroi which are carved in detail only on the head, torso and lower leg, precisely the areas where the Greek hoplite wore his armour. This decoration appears uniformly on Greek kouroi, and must have already developed into a standard convention before the first stone statuary appeared in Greece in the 660s BC. This was noticed by Hagemann in studying the kouroi of Cleobis and Biton (Hagemann 1919, p. 18) and is well illustrated by an example in Bury and Meiggs (1980, p. 125). This gives further confirmation that the Argos cuirass is Greek, not a European import, and that since it had already reached the basic anatomical conventions, it must have been pre-dated by cuirasses of a more primitive design. This idea is supported by the technical qualities of the bronze work, which show that it is not an early piece of work by its maker (Snodgrass 1965, pp. 75–6).
The evidence for cuirasses earlier than Argos (c. 725) is slight. It used to be thought that the cuirass did not appear in Greece until well into the seventh century, and it was only after the Argos find that Late Geometric pottery was examined for traces of the bell cuirass. The silhouette technique and exaggerated forms used on such pottery make it difficult to reach definite conclusions, but there are some illustrations which show a flange either side of a warrior’s waist, and it is possible that the artist was trying to depict a bell cuirass (Snodgrass 1965, p. 73, and p. 234 n. 5; but see his plate 4, especially figs b and d). Perhaps the most convincing example is on a large Late Geometric vase in the British Museum. Here there are warriors in chariots, whose torsos seem to be wearing some sort of body armour marked with separate breasts. These could well be bell cuirasses. The bell cuirass in its simplest form must certainly have been in use by 750, if not before, and it would have travelled with the first Greek colonists rather than being discovered by them. There is also the case of General Timomachus, who had a bronze cuirass called a hoplon – apparently from the later eighth century – which was later carried in procession by the Thebans (Cartledge 1977, p. 25).
If we have decided upon where the bronze cuirass came from, we have yet to decide why. As Hanson (1989, pp. 56–7) has pointed out, these cuirasses, along with the helmets and later the hoplite shield, were very heavy, and bronze armour in the Aegean heat seems very impractical. The Mycenaean warrior had occasionally worn bronze, but only chariot warriors wore large amounts of body armour and they would not have had much walking to do. The large numbers of infantrymen in the Late Mycenaean period do not seem to have adopted bronze body armour (apart from greaves), and that may have been because of the heat and weight. If our theory about eighth-century Greeks adopting bronze plate armour because Mycenaeans wore it is correct (they had heard about it in Homer), then that would explain why an armour was adopted that was otherwise perhaps unsuitable. Once adopted, bronze armour did have many advantages for the Greek warrior and later hoplite. As we have seen with the helmets, bronze – especially in the curving forms of the bell cuirass – is very effective at deflecting attacks, and the padding behind would have absorbed many blows. The warrior would have felt like one of Homer’s heroes in his shining panoply, which would have given him great confidence on the battlefield. Also it seems likely that hoplites, burdened by their shields as well as by this bronze armour, may have been able to fight only for an hour or so (Hanson 1989, pp. 55–6). This would have led to fewer casualties as both sides quickly became exhausted, and explains why much effort was put into winning the battle in the initial charge. Hoplite battles at this time were about winning and losing, collecting the dead and putting up trophies. They were not meant to lead to the annihilation of the other side, and bronze armour may have played its part in this ritualisation of hoplite warfare.
The bell cuirass remained virtually unchanged for two centuries, but it does become commoner in both finds and illustrations in the middle of the seventh century, at the point when hoplite warfare emerges (Jarva 1995, pp. 24, 27). It seems that bronze for cuirasses became easier to obtain and so more people were able to fight, although cuirass finds do not come close to the number of helmet finds. This may show that many warriors did not wear this body armour, but artistic depictions such as the Chigi Vase tend to suggest that bronze cuirasses were fairly universally worn after c. 650. The evidence from the sanctuaries shows only that helmets (and perhaps shields) were more likely to be dedicated than cuirasses. They probably had more visual appeal. Only the Argos find is securely datable, because our other finds come from sanctuaries like Olympia and Delphi, and the Afrati group from Crete. Armour captured from the enemy was dedicated at these sanctuaries, and inscriptions on the armour relating to such battles can, very occasionally, be used in dating (Jackson, in Hanson 1991, pp. 228 ff.). The dating problems stem from the fact that when a sanctuary building became packed, offerings were cleared out and thrown into pits, wells, rivers, etc., removing any stratigraphical dating material. At Olympia several pieces of armour have been found built into the bank supporting the stadium, or thrown into a nearby river (Snodgrass 1965a, pp. 73–4; 1967, p. 49).
Courbin tried to devise a typology based on the cuirass becoming shorter in height over time, with a less pronounced bell curve, and the anatomical detail on the shoulder blades becoming more elaborate. Jarva has shown these dating methods to be illusory (Jarva 1995, p. 25). Although highly decorated shoulder blades tend to be found later, there are some demonstrably late cuirasses that have shoulders of a simple form like the Argos model. One cuirass which can be dated is the ‘Crowe’ Cuirass from Olympia, and there is a companion piece almost certainly by the same armourer; these can be dated by their decoration to c. 630–610 (Hoffman and Raubitschek 1972, p. 52). These pieces are both backplates, heavily decorated with incised pictures of animals and figures like the Cretan cuirasses (see below), but the technique is simple and more closely related to Corinthian work (ibid., p. 50). The ‘Crowe’ Cuirass has added interest because of several square holes, which were originally thought to have been caused by arrows. Hoffman and Raubitschek (1972) have now shown that these holes were made with a chisel, and it seems that the cuirass was nailed up on a post, probably in a sanctuary as a battlefield trophy. The decoration shows that it belonged to an officer or a wealthy hoplite, and was presumably chosen as a trophy because of its splendid decoration.
Jarva (1995, p. 24) has shown that the height of cuirasses is not really determinate of age, but more of the size of the man wearing it. A variant of the bell with a jutting flange rather than a gentle bell curve does seem to have become more usual towards the end of the life of the bell cuirass (c. 525), but some examples with such a flange could equally be from the seventh century (Jarva 1995, p. 22, no. 3; Connolly 1998, p. 55, no. 6). The cuirass illustrated by Connolly (from the Olympia Museum) has a highly defined omega curve which is a late feature (Mallwitz and Herrman 1980, p. 93; Jarva 1995, p. 26), and semi-spirals on the breast that also appear to be a late design. Certainly a late feature is a hinged joint at the shoulder that appears on one breastplate from Olympia (Jarva 1995, p. 20, no. 8), but it is otherwise attested only for the muscle cuirass (see below) (Snodgrass 1965a, p. 75).
In Crete, evidence for bell cuirasses begins with a large group of miniature votives from Praisos. As mentioned earlier, this group includes miniature Insular, as well as Corinthian, helmets and must date from c. 700 (Bosanquet 1901–2, passim). Miniature armour would have been cheaper to dedicate than the real thing, and examples tend to be from the earlier seventh century; but there is also a preponderance of finds from Crete, where it may have been a popular custom (Jarva 1995, pp. 15–16). I know of just two miniature votives from the mainland: at Sparta and Bassae in Arcadia (Snodgrass 1965a, p. 74; Cartledge 1977, p. 14).
From Praisos, we have two complete miniatures marked with repoussé breasts and omega curves, and with the breastplate attached to the backplate at the shoulders. There are also ten other single cuirass halves of a long tubular form, as well as miniature helmets, abdominal guards and shields. No full-sized examples of this long, undecorated tubular cuirass have been found, although there is another miniature from Gortyn, also on Crete (Snodgrass 1965a, p. 74). This latter cuirass does have anatomical markings, however. The Cretans seem to have adopted the cuirass, like the helmet, at an early stage and then went ahead with their own tubular design, which reached down to the thighs for extra protection while sacrificing some freedom of movement. It may have had anatomical markings, which are absent from these miniatures (perhaps they were only painted on), and this seems likely given the amount of decoration that Cretan cuirasses had later. On a scale size, some of these tubular plates are excessively long. If this is not a distortion caused by re-creation in miniature, or a stylistic convention, then I would suggest these plain pieces must be backplates. They would reach down past the buttocks, but could be worn with a breastplate of normal length and an abdominal guard (see below). Miniature abdominal guards were found at Praisos but, since they were not attached to the cuirasses, it is difficult to connect them definitely with the tubular cuirasses. Abdominal guards are a peculiarly Cretan item, and the early use of tubular backplates might explain their origin. Only one ‘full-sized’ breastplate was found at Praisos and is about 22cm wide (it is badly crushed). Snodgrass (1965a, p. 74) suggests it might be a large votive, in which case it is the only one known. It could equally well be a proper piece of armour for a small man or even a boy. Armour for children is well known from the Middle Ages, and this could be an ancient Greek example of the practice.
Later seventh-century Cretan cuirasses come mostly from Afrati in Greece and their details have been published by Hoffman and Raubitschek (1972). Apart from the helmets already mentioned, there are nine cuirasses and sixteen abdominal guards. The absence of greaves is particularly noteworthy. Perhaps they took a while to become popular in Crete, or perhaps Cretans never really took to them. Corslet no. 1 in Hoffman and Raubitschek 1972 (plates 19–23) is the most elaborate. It has repoussé figures of lions on the breasts and a pair of griffins marching up the omega line. On either side of the griffins, a warrior wearing an Insular helmet kneels on a curved tendril; and the breasts are marked out with sea dragons. A pair of uncertain animals is below the omega curve. All this repoussé decoration is supplemented with incised lines, and the style dates it to about 660 (Hoffman and Raubitschek 1972, p. 43). The other corslets are plainer, but three have the curved repoussé lines around the breast bulges ending in lotus flowers, with another lotus flower below the neck. Corslet 8 (Hoffman and Raubitschek 1972, plates 28–9) is unusual, in that instead of a repoussé omega line the entire upper thorax is raised in high relief above the stomach region; this shows a very skilled technique. It probably dates from shortly after 600, like some of the abdominal guards (Hoffman and Raubitschek 1972, pp. 44–5). The bronze on these cuirasses is thinner than the Argos cuirass at 0.6–1mm, giving a weight of about 5kg rather than the 7–8kg that the Argos cuirass would have weighed. The edges are also rolled over bronze wire, rather than iron. Where tube slots survive on these cuirasses they show that the breastplate overlapped the backplate, but that hinge pins were now inserted on the outside of the cuirass, making the cuirass easier to put on. The warrior could certainly have taken it off by himself if he needed to, such as to aid flight.
The bell cuirass was the main form of body armour – when body armour was worn – from c. 750 to about 525 or 500, and it seems to have gained in popularity throughout that period, judging by the examples illustrated in art and indeed the dedications at sanctuaries. Examples of the cuirass in seventh-century art invariably show it in its natural bronze colour and with the simple anatomical decoration that we have described. In the sixth century there is some evidence for painting the cuirass, just as we have seen with the helmets. Many cuirasses show spiral curves on the breasts that are much longer than actual examples. Jarva (1995, p. 25) thinks we are dealing with artistic exaggeration here, but it seems quite possible that the repoussé lines were continued with painted lines to give the long spirals that we see. An example on a vase dating to c. 540 (Boardman 1980a, p. 80, no. 98) shows double lines of spirals on the breasts, one painted in red, as well as red lines below the omega curve. The white lines could be repoussé work in bronze, while the red lines were painted to enhance the effect. Another example has white dots painted along the spiral (Arias 1962, fig. 59). There are also examples on vases of bell cuirasses painted in two different colours, often with the omega line as a boundary (Boardman 1980a, p. 50, no. 57). The vase featuring the cuirass with white dots also shows another hoplite in a gold-coloured cuirass. These cuirasses are often worn by Homeric heroes on vases and remind us of Homer’s mentions of gold armour. It is possible that some wealthy hoplites gilded their helmets and armour.
Some late illustrations of ‘bell’ cuirasses show a cuirass stopping at the waist, with the hips protected instead by pteruges, a row of leather or linen flaps probably adapted from the shoulder-piece corslet that was then becoming popular (Jarva 1995, p. 29 Type II). It seems likely that these flaps were attached directly to the cuirass or its lining, rather than being part of a separate arming jack. These late bells date from around 520. An example on a tomb painting at Elmali shows a large circular pattern of scales or feathers on the chest which could be embossed or painted decoration, and also clearly shows the join at the left shoulder where the armour covers the shoulder at the left side. This would have restricted movement a lot, but the shield arm did not need to be so flexible (Boardman 1979, p. 13, no. 1.2).
Jarva also lists another cuirass (Type V) (1995, pp. 46–7), which appears to have features of both bell cuirasses and shoulder-piece corslets, generally showing the breast spirals of the former and the horizontal, decorated bands and pteruges of the latter. A well-known ‘bilingual’ vase (i.e. featuring both black and red figure work) of Ajax and Achilles at play shows one of these, but the horizontal line could be embossed or painted decoration on a normal bell. Even the Argos cuirass had parallel horizontal lines like a belt. Jarva (1995, fig. 17) features two further examples, also from the last quarter of the sixth century. This figure clearly shows a linen corslet, because it is painted white and the ‘breast spiral’ is both small and actually on the back of the hoplite; it could be the sort of woven or embroidered decoration that appears on these (see below). His other example looks more like a mixed bronze and linen cuirass. A black figure vase now in Rome (Jarva 1995, fig. 16) shows a hoplite wearing a cuirass which has breast spirals and an omega curve; but there are two or three bands below that which look much more like a composite corslet, and possible pteruges below those. Altogether, Jarva lists sixteen examples on vases of this kind of mixed corslet, which he sees as a transitional piece of armour. As I have shown, some of these can be assigned to either a bell or a shoulder-piece cuirass, whereas many others are of such poor artistic quality that it is likely that the artist is in error (Jarva 1995, p. 47). There is no archaeological evidence for a cuirass made only partly of metal at this time, which is particularly significant given the amount of armour discovered at sanctuaries dating from this period. The bell cuirass continued to be shown in art until c. 480 (Boardman 1980a, p. 164, no. 283), but these last examples are very crude and are probably cheap copies of earlier vases. It would be safe to assume that the bell cuirass was last made in about 500 and did not see action in the Persian Wars. The shoulder-piece corslet became the popular form of body armour, and bronze armour workers went on from the late sixth century to develop the muscle cuirass.
Another piece of body armour in use at this time was the abdominal or belly guard, a usually semi-circular piece of bronze hung from a waist belt. It is sometimes called the mitra, after a Homeric piece of armour, but the two items are probably not equivalent. Around about fifty of these guards are known, outnumbering the forty or so known cuirasses, but it is significant that two-thirds are from find spots certainly or probably in Crete. The remaining examples are mainly from Olympia, and it is possible that they were all dedicated by Cretans, or taken from Cretans, but the number is such that we can say that the belly guard was probably also used to a small extent by other Greeks (Jarva 1995, pp. 51 ff.). The guards we have date from about 675 for an example from Axos to c. 525 for some examples featuring engraved winged horses (Jarva 1995, pp. 54–6). There are also some miniature belly guards known from Praisos and Gortyn on Crete, and some of these may date from as early as c. 700 (Bosanquet 1901–2, passim).
Unlike cuirasses, belly guards were not made to fit the body and are not lined or padded. For extra protection they are made much thicker than other pieces of armour – generally 5mm to 7.5mm thick – and they do not normally have edges rolled over bronze wire, although this is present in some examples. The belly guards are suspended from a belt by three bronze rings and are generally 25cm wide along the top edge and 15cm deep in the middle. Jarva has noticed a change in shape from the earliest examples, which were shallow crescents. These became longer and squarer before developing into a more aesthetically pleasing true semi-circle (Jarva 1995, pp. 54–5). As with Cretan helmets and cuirasses, many of these belly guards are highly decorated with both repoussé work and engraving. The commonest decorations are facing-horse or winged-horse protomes, but there is also an example with a double-bodied panther, and one from Rethymnon decorated with four youths (Hoffman and Raubitschek, pp. 24–5). The belly guards generally have two or more lines of heavy repoussé decorating the borders.
A second type of belly guard is represented by just two examples from Olympia and one complete example from a tomb in Bulgaria, ancient Thrace (Venedikov 1976, p. 51, no. 193; Webber 2001, p. 34). This type is more of a trapezoidal shape, much longer and with a long horizontal hinge at the midpoint to allow bending. One Olympia example dates to c. 520 and the Thracian example is late fifth century (Jarva 1995, p. 57). With only two examples at Olympia, one could see this late belly guard as being used only on the fringes of the Greek world. The dates correspond with the general use of the stiff linen and composite cuirass in Greece, and the two are not compatible. I think it is much more likely that the two examples from Olympia have come from Thrace or from Italy, where the bell cuirass and its descendants continued to be used.
Returning to the Cretan pieces, the finds from Afrati are inscribed for the most part as dedications, often with the dedicator’s name. This enables us to see that in four cases we have pieces from the same panoply: two cases of helmet and belly guard, and two of cuirass and belly guard (Hoffman and Raubitschek 1972, p. 16). Even allowing for missing pieces, this evidence and the number of belly guards compared to cuirasses suggest that some hoplites chose a belly guard rather than a cuirass as an extra piece of armour. This may well have been on grounds of cost rather than for a specific combat benefit. The number of belly guards also suggests the possibility that they were sometimes worn in pairs. Belly guards no. 1 and no. 8 in Hoffman and Raubitschek (1972, p. 77) seem to have parts of the same dedicatory inscription, suggesting they were both from the same panoply. I would suggest that the plain guard hung at the back and the decorated one at the front. Such a piece of armour might seldom have proved useful, although the same could also be said of a cuirass backplate. The belly guard hardly ever appears in art. As recently as 1965 Snodgrass (1965a, p. 89) could state that it was never shown in art, just as Benton (1940– 5, p. 82) had said earlier. Jarva (1995, p. 58) has now published a photograph of a sculpture from Albania that does clearly show the belly guard. It is worn with a bell cuirass and Corinthian helmet, but is apparently from the fifth century, which is after the bell cuirass and belly guard were in use in Greece and Crete. This fact, and the complete lack of illustrations of the belly guard in mainstream Greek art, shows that it was very much restricted to Crete in the seventh and sixth centuries and to northern Greece (or north of Greece!) in the fifth. There is also some uncertain evidence in the form of a sculpture for its use in Etruria (Jarva 1995, p. 59, fig. 24). Apart from the throat, the groin and thighs formed the other vulnerable area in combat. Spear thrusts deflected downwards by the shield might go below the cuirass to injure a warrior there, and it is perhaps surprising that this belly guard was not more popular. It would have added more weight to the panoply, of course, which was perhaps not thought to justify the added protection. When pteruges were invented in the middle of the sixth century they were widely adopted, showing that the hoplite was well aware of his vulnerability in this area, but had been generally content (until the advent of pteruges) to rely on his shield for protection.
Armour for protecting the limbs was also used extensively through the Archaic period, with the greave being the most popular item. This is due to the fact that the lower legs are the hardest part to protect with a shield and, although the lower legs were not very vulnerable in hand-to-hand combat, they were vulnerable to missiles. Greaves are mentioned late in the seventh century as being specifically worn for protection against missiles (Alcaeus, frag. 54 (Diehl); see Jarva 1995, p. 85). As with helmets and body armour, the greave had been invented in Mycenaean times, but evidence for continuity of use through the Dark Age is problematical (Jarva 1995, p. 85). There are European greaves, which may date from as early as 800, but the same problems of dating exist with them as for the European cuirasses (Schauer 1982, passim).
The earliest Greek greave is almost certainly an example from Olympia, which Kunze dates to c. 750 (1991, p. 5 and plate 1; but see Jarva 1995, p. 85, who suggests it is much later and non-Greek). Although not datable by context, this is an extremely simple short greave, which was laced onto the leg in a similar fashion to Late Mycenaean greaves. The most remarkable feature about it is that it has three parallel repoussé lines at the edges, and two large circle decorations with central bosses near the top. This decoration is almost identical to the best-preserved Late Mycenaean greave from Enkomi and suggests either a possible continuity, or again the resurrection of use through the discovery of examples in Mycenaean tombs, as has been suggested for cuirasses.
The Argos grave containing the cuirass and the Kegel helmet also produced some thin fragments of bronze, which were interpreted as possible greaves. Their fragmentary state in comparison with the cuirass and helmet from the same tomb is a useful reminder of how many pairs of early greaves may have corroded away for ever. Another probable early pair of greaves comes from Kavousi in Crete, although not everyone is agreed on the interpretation, some thinking they may be part of a bronze statue (Jarva 1995, p. 65). Kunze (1991) thinks they are greaves and can be dated to the late eighth century. They are short at 23.5cm, and are decorated with studs rather than repoussé lines around the edges. They could not have fitted onto the leg without lacing and, although there are edging holes for attaching a lining, there seems to be no means of attachment to the leg. The greaves are very fragmentary, however. Apart from a pair of miniature greaves from Gortyn (Yalouris 1960, p. 51) in Crete, the only other early greaves (perhaps c. 680) are a badly crushed pair from Praisos that are only about 20cm high (Astrom 1977, p. 46, fig. 15). Astrom sees these as one greave and one lower arm guard, matching what he found in the Dendra cuirass tomb (Astrom 1977, p. 49; Schauer 1982, p. 148), whereas Snodgrass (1965a, p. 87) thinks they are early short greaves but perhaps not a matched pair. Kunze (1991, p. 6, no. 15) thinks that they are from a bronze statue and not armour at all. In support of Kunze is the fact that they are plain pieces of bronze, whereas the early short greave from Olympia and the Kavousi greaves all have decoration of some form. However, the Kavousi greaves, which I think definitely are greaves, are quite different from the early Olympia greave, and also from the ‘prototype’ greaves from Olympia (see below), which begin in the last quarter of the eighth century, if not earlier. It seems there is no standard design at this early period.
Apart from these early pieces, more than two hundred greaves have been uncovered at Olympia and recently published by Kunze (1991) where they have been sorted into types and dated stylistically. Jarva (1995) has added to this by dating using the edge perforations. Most Archaic greaves had edge perforations for the attachment of a lining, and the gap between each hole narrowed over time. In the early seventh century each hole was about 25mm apart and linings seem to have been fixed with rivets (Jarva 1995, p. 65). By the time we get to c. 525, holes were less than 2mm apart and the lining was probably sewn on to the bronze through these holes (Jarva 1995, pp. 65–72 and fig. 28). Both Jarva’s and Kunze’s methods have exceptions, but between them they help to date approximately many otherwise undatable finds from Olympia. Jarva has divided the Archaic greaves (pre-700 to c. 500) into four groups: prototype, transitional, calf-notch and spiral.
The prototype greaves started in around 700 according to Kunze, but could have begun as early as 750 according to Jarva (1995, pp. 85–6). Their main typological features are that they are short, 32 to 36cm, ending on or below the knee, although this made them much longer than the early greaves discussed above. They also often came with extra lacing holes to secure the greave to the leg. Most have some decoration, perhaps delineating the calf muscle and with a semicircle or volute at the knee, and some are a little more elaborate (Jarva 1995, pp. 86–7; Kunze 1991, plates 2–9). This greave was in use until about 650/640. Jarva’s transitional greave (1995, p. 88) has more of an angle with the front face at both bottom and top and is slightly longer, 34 to 39cm, sometimes covering the knee. Kunze (1991, p. 25) says that there is more elasticity in these greaves and that there is usually only one set of lacing holes near the top of the greave for holding them on. It is this greave type that features on the Chigi Vase as the sharp angle at the top of the greaves can be seen, as well as the fact that they do not cover the knees (Jarva 1995, p. 90). This type lasted until c. 600.
The third type of greave, Jarva’s ‘calf-notch’, was the first to cover the kneecap and generally corresponds to Kunze’s High Archaic greave. These average about 39cm in length, but vary between 36 and 44cm. The inner side of each greave is marked with a distinct notch outlining the calf muscle, and the greaves wrap around the back of each leg much more. Only a few of these had lacing holes, and most were held on just by the elasticity of the bronze. A few of the later greaves in this group did not have edge perforations for a lining, which must have been glued on instead. These greaves have the edges of the bronze rolled over (Furtwangler 1966, no. 988). The calf-notch group lasted from c. 630 to after 540, becoming more elaborate and developing into the spiral group. The spiral group saw the calf-notch develop into decorative spirals, sometimes ending in lotus heads or snake heads and with anatomically decorated kneecaps. Kunze has further divided this group, depending on a decorative feature below the kneecap, into the V variant, S variant and club variant (Kunze, 1991, p. 68; Jarva 1995, pp. 94–5). The V variant dates from c. 560–540, the S variant from c. 540–510 and the club variant from c. 540–525. All variants average about 42cm in height, and the bronze in the spiral group tends to be thinner than in earlier greaves (Kunze 1991, p. 66). Some elaborate greaves had engraved sides or decorated kneecaps (Jarva 1995, p. 95) and this decoration may have been enhanced by painting, as we surmised earlier for helmets and cuirasses. Spiral greaves are sometimes shown on vases of the period and correspond well with the dates formulated by Kunze and Jarva using style and perforations (Boardman 1979, no. 2.1; 1980a, no. 68).