By late in the afternoon of August 2, 216 B.C., no room remained to fight and little more in which to die. Given the crashing press of their exhausted fellow soldiers, the Roman legionaries could not retreat, advance, or even find much area to wield their swords. Frenzied Iberians in white tunics and half-naked Gauls were in their faces. Veteran African mercenaries suddenly appeared at the flanks. From their rear arose cries that Celtic, Iberian, and Numidian horsemen had cut off any hope of escape. Thousands of Hannibal’s hired men—a who’s who of the old tribal enemies of Rome—were everywhere. Nowhere were there enough Roman cavalry and reinforcements. A vast mass of 70,000 brave souls was encircled in a small plain in southwestern Italy by a poorly organized but brilliantly led invading army half its size.
Confusion and terror only grew greater as dusk neared, as each Roman pushed blindly and was shoved into the enemy at all sides. Stacked in rows to the depth of thirty-five and more, the size of the unwieldy mass began to ensure its destruction. A marvelous army designed for fluidity and flexibility was unaccustomedly caught fast in an immovable column. The men of Rome had never before marched out to a single battle in Italy in such huge numbers—and would never do so again. And not until a similar disaster at Adrianople (A.D. 378) six centuries later did the Roman army deploy itself to such an unwieldy depth, making it an easy target for missiles and preventing the great majority of its soldiers from ever reaching the enemy.
The sight of the mass fighting must have been as spectacular as it was soon sickening. Unlike the Romans, Hannibal’s men were a heterogeneous-looking bunch. In the center the backpedaling Celts and Gauls, as was their custom, fought stripped to the waist (“naked,” Polybius says), probably armed only with heavy wooden shields and clumsy swords that were virtually pointless and were only effective in sweeping, slashing blows that left the attacker wide open to quick counterjabs. A few may have had javelins or spears. Their white, muscular physiques and great size were favorite topics of Roman historians, who were quick to imply that smaller tanned Italian legionaries used training, order, and discipline to butcher such wild tribesmen in the thousands. For the next two centuries commanders like Marius and Caesar would wipe out entire armies of just such brave and physically superior warriors. We think of French slaughter in terms of Agincourt or Verdun, but the true holocaust occurred in the mostly unknown battles of the two-century encounter with the Romans, who cut down more Gauls than at any time before or after. Roman steel, not disease or hunger, doomed an autonomous ancient France, whose manhood was systematically destroyed in battle as no other people would be in the entire history of Western colonial subjugation. Caesar’s final annexation of Gaul made the nineteenth-century American fighting on the frontier look like child’s play—a million killed, a million enslaved, Plutarch recorded, in the last decades alone of that brutal two-century conquest.
Hannibal may have put these brave Gauls in the center to incur the Romans’ fury and thus draw them farther into the encirclement. Livy remarks that they were the most terrifying of all Hannibal’s troops to look upon. In the classical world the stereotype of utter uncivilized savagery was a white skin, long greasy blond—or worse, red—hair, and a flowing unkempt beard. Four thousand of them were sliced to pieces by the methodical Italians. Alongside them at the vortex marched hired Spaniards—ostentatious infantrymen with iron helmets, heavy javelins, and dazzling white cloaks bordered with crimson, which, like the nakedness of their pale Gallic allies, would soon only highlight the bloodletting. Unlike the Gauls, the Spanish also wielded the short double-edge sword—copied and improved upon by the Romans as the gladius—lethal as a slashing and stabbing weapon. Stationed next to the Gauls, they were cut down mercilessly—though Polybius says hundreds, not thousands, of these better-armed and protected warriors fell.
At the front of the oncoming Roman mass, the fighting soon degenerated into swordplay and hand-to-hand pushing, biting, and clawing. Only the steady feigned withdrawal of the Gauls and Spanish and the impending encirclement at the flanks saved these sacrificial tribal contingents from utter annihilation. Livy and Polybius both focus on the doom of the surrounded Roman legions, but more than 5,000 Spaniards and Gauls must have suffered ghastly wounds before being trampled to death by the legionary steamroller. How Hannibal and his brother Mago survived the slaughter we are not told; but both stood gallantly among the Gallic and Spanish front ranks, ensuring that their retreating pawns not break before the trap was set.
Hannibal’s best were his African mercenaries stationed on the flanks and ordered to turn about and hit the legionaries as they rushed by, heedless in their bloodlust. These were grim professional soldiers who had battled a score of North African tribes, fought Europeans during their march from Spain, and on occasion turned on their own Carthaginian masters when pay was not forthcoming. Centuries later their legendary toughness impressed the novelist Gustave Flaubert, whose novel Salammbô has as its backdrop one of their numerous bloody revolts. At Cannae they probably first pelted the outer ranks of the legions with javelins and then cut their way in through the Roman flanks, since legionaries could scarcely turn sideways on the run to meet this new and unexpected threat.
Although they were not used to the Roman equipment—the Africans more often fought Macedonian-style as phalangites with two-handed pikes—they were veteran killers, and far more experienced than the adolescents who filled the Roman ranks, which were depleted by the thousands butchered earlier at Trebia and Lake Trasimene. Moreover, the African heavy infantrymen on the flanks were stationary and fresh, the oncoming Romans exhausted from killing and pressing the Gauls and Spaniards. The former were staring intently on their prey, the latter oblivious to their danger. Within seconds the killers became the killed, and it is a wonder that even 1,000 Africans were lost during the entire afternoon— a mere fiftieth of the Roman total. The collision of African infantry with the Roman flanks must have been horrendous, as dense files of shuffling legionaries were suddenly hacked and ripped apart on their vulnerable sides, without opportunity or room to halt and face their attackers. Roman infantrymen were superbly protected at their front, and adequately from their rear; but their sides were relatively bare—exposed arms behind the shield, less body armor below the shoulder, and the ears, neck, and portions of the side of the head without cover.
Who could distinguish friend from foe, as Africans and Italians sliced away at each other, wearing similar breastplates, crested helmets, and oblong Roman shields? Polybius claimed that when the Africans hit the Romans broadside, order was lost for good and the mass rent beyond repair. The rear flanks and base of the Roman column were still unenclosed, and here the other great failure of the Roman army became manifest: besides its poor generalship, there were far too few Roman horsemen. Most of the mounted troops present were vastly inferior to the some 2,000 Numidian light cavalry on the right flank, men who had been on their horses since childhood, who could throw javelins with deadly accuracy at a gallop and slash away with swords and battle-axes at close quarters as easily mounted as on foot. On the Carthaginian left wing a horde of 8,000 Spanish and Gallic horsemen—with spears, swords, and heavy wooden shields—likewise tore apart the Roman cavalry. Hannibal had arrayed 10,000 skilled horsemen on the two wings against 6,000 poorly trained mounted Italians. After driving off the enemy cavalry, the Numidian and European horsemen turned to slaughtering the enclosed infantry from the rear.
The presence of some 10,000 fresh horsemen at the base of the Roman column, and 20,000 Africans on the flanks, with the dust in the Romans’ faces, the screaming of dying Gauls and Spaniards, and the sheer difficulty of distinguishing friend from foe, made the tiny summer battlefield a confused slaughterhouse. Three hours earlier the Roman army had marched out as a foreboding mass of iron, bronze, and wood, rank after rank of crested helmets, huge shields, and deadly javelins in a solemn procession of undisguised pride against Hannibal’s motley and outnumbered mercenaries. Now there was little left but a heap of broken weapons, oozing bodies, severed limbs, and thousands of the crawling half-dead.
The terror of battle seems not the mere killing of humankind, but the awful metamorphosis that turns on a massive scale flesh to pulp, clean to foul, the courageous to the weeping and defecating, in a matter of minutes. Just as Admiral Nagumo’s beautiful four carriers at Midway had been a showcase of power, grace, and undefeated energy at 10:22 A.M. on June 4, 1942, and six minutes later blazing infernos of charred bodies and melting steel, so the thousands of plumed swordsmen in perfect order were transformed nearly instantaneously from a majestic almost living organism into a gigantic lifeless mess of blood, entrails, crumpled bronze, bent iron, and cracked wood. Men and matériel that were the products of weeks of training and months at the forge were reduced in moments to flotsam and jetsam by the genius of a single man. Brilliant generalship in itself is a frightening thing—the very idea that the thought processes of a single brain of a Hannibal or Scipio can play themselves out in the destruction of thousands of young men in an afternoon.
For the next 2,000 years armchair tacticians would squabble over the mechanics of the slaughter at Cannae—seduced by the idea that a numerically inferior invader in a few hours could exterminate its enemy through simple encirclement. Clausewitz (“Concentric activity against the enemy is not appropriate for the weaker side”) and Napoleon both felt Hannibal’s trap too risky and the product more of luck than genius. For the Prussian strategist Count Alfred von Schlieffen, Cannae was not the chance butchery of thousands, but a tactician’s dream come true that was “most wonderfully fought” and planned to the last detail—the essence of what military erudition combined with fighting spirit might accomplish. Schlieffen, who in his own time foresaw a Germany besieged by more numerous enemies, found it reassuring that the intellect of one man could nullify the training, expertise, and sheer numerical superiority of thousands. Indeed, Schlieffen would write an entire book, aptly entitled Cannae, on the Prussian army’s bold and repeated attempts to achieve Hannibalic encirclement on a massive scale. The great German invasion that ended at the Marne (September 1914) and the battle of Tannenberg (August 1914) were both efforts to entrap and surround entire armies, and thus invoked the mythical idea of Cannae—without real appreciation that tactical encirclement, ancient and modern, need not lead to strategic victory. Yet rarely does any great captain encounter an enemy deployed so absurdly as the legions in August 216 B.C. The Romans, who might have outflanked Hannibal’s outnumbered line by two miles, instead presented a front that was roughly the same size—and far more inflexible.
Many wounded had been hamstrung by marauding small bands, their writhing bodies left to be finished off by looters, the August sun, and Carthaginian cleanup crews the next day. Two centuries later Livy wrote that thousands of Romans were still alive on the morning of August 3, awakened from their sleep and agony by the morning cold, only to be “quickly finished off” by Hannibal’s plunderers. Roman corpses “were discovered with their heads buried in the earth. Apparently they had dug holes for themselves and then, by smothering their mouths in the dirt, had choked themselves to death” (22.51). A few thousand crawled about like crippled insects, baring their throats and begging to be put out of their misery. Livy goes on to record examples of extraordinary Roman courage discernible only through autopsy of the battlefield: a Numidian who had been brought alive out of the pile from beneath a dead Roman legionary, his ears and nose gnawed away by the raging Roman infantryman who had lost the use of everything but his teeth. The Italians, it seemed, fought desperately even when they knew their cause was hopeless—a realization that must have sunk in among most after the first minutes of battle.
Hannibal, in the ancient tradition of victorious military commanders, grandly inspected the battlefield dead. He was said to have been shocked at the carnage—even as he gave his surviving troops free rein to loot the corpses and execute the wounded. The August heat made it imperative to strip promptly the bloated bodies and torch the stinking flesh—a feat of logistics in itself just to hack away the armor from the torsos and haul away thousands of putrid corpses. No grave site near the battlefield has as yet been uncovered, nor any traces of the bones of the dead, so the bodies were probably left to rot.
The destruction of some 50,000 snared Italians in a single afternoon—more than 200 men were probably killed or wounded each minute—was in itself a vast physical challenge of slicing thousands with muscular power and iron in the age before the bullet and gas canister. Livy (22.49) remarks of the legionaries’ “refusal to budge,” and emphasizes their willingness “to die where they stood,” which only further “incensed the enemy.” There must have been at least 30,000 gallons of blood spilled on the battlefield alone; even three centuries later the satirist Juvenal dubbed Cannae the scene of “rivers of spilt blood.” The sea “turned red at Lepanto” from the blood of 30,000 butchered Turks, but the tide cleansed the site within minutes. The horrible carnage of some 50,000 to 100,000 at the final siege of Tenochtitlán was beside a lake, whose waters eventually might mitigate the stench. Given the deep columns of the Romans and Hannibal’s tactics of encirclement, Cannae became an unusually tiny battlefield, one of the smallest killing fields to have hosted such large numbers in the entire history of infantry battle. For the rest of the summer of 216 B.C. the plain of Cannae was a miasma of decaying entrails and putrid flesh and blood.
From our written sources—the Greek and Roman historians Appian, Plutarch, Polybius, and Livy—we know that the late afternoon of August 2 was one of the few ancient battles in which an entire army was destroyed after hitting the enemy head-on. In general, the complete slaughter of hoplites, phalangites, and legionaries was somewhat rare and accomplished only by flank attack, lengthy pursuit by cavalry, or ambush. At Cannae the entire Roman army advanced frontally as one unit and at the same time in unobstructed terrain, ensuring a magnificent collision of arms that would lead to either spectacular victory or horrendous defeat. Polybius called the daylight encirclement at Cannae a “murder.” Livy also thought it a slaughter, not a battle, and the ill-famed nature of the fighting explains why Cannae is one of the better-recorded battles—three detailed accounts survive—of the ancient world.
Never in the five-century history of Rome had so many infantrymen and their elected leaders been trapped on the battlefield with no certainty of escape. After the battle the thirty-one-year-old Hannibal would collect the gold rings of more than eighty consuls, ex-consuls, quaestors, tribunes, and scores of the equestrian class in a bushel. Military historians have praised Hannibal’s genius and blamed the Roman catastrophe on Rome’s bureaucratic system of electing and training its generals. In their eyes Cannae is a result of singular tactical brilliance pitted against institutionalized mediocrity. That analysis is scarcely half-true: if the Roman system of tactical leadership, with its commitment to civilian oversight and nonprofessional high command on the battlefield, was responsible for producing a succession of amateurish generals who would lose a string of battles during the Second Punic War (219–202 B.C.), it also deserves credit for ensuring that Cannae and the previous disasters at the Ticinus and Trebia Rivers and Lake Trasimene were not fatal to the Roman war effort. Cannae, like so many of these landmark battles, is the exception that proves the rule: even when Roman armies were poorly led, foolishly arranged, squabbling before battle over their proper deployment, and arrayed against a rare genius, the catastrophic outcome was not fatal to their conduct of the war.