Share This Article
On a hot, dusty field in 216 B.C., a Roman army perished and the dream of double envelopment was born. 

BY THE END OF AN AUGUST DAY IN 216 B.C. (THE EXACT DATE REMAINS IN DISPUTE) By the end of an August day in 216 B.C., some 50,000 Italians of nearly every age (out of the morning’s estimated 80,000) lay butchered on the small plain at Cannae, caught between Hannibal’s thin pincers—so much for the pedantic adage of military theorists that absolute encirclement of the enemy by the weaker force is tactically impossible. Absolute encirclement it surely was, for Cannae, after all, was no ambush of a maniple, collapse of a cohort, or loss even of a trapped legion, but rather the wholesale slaughter of an entire consular army—over 100 men killed each minute of a single afternoon.

Cannae has exerted an almost narcotic spell on military men for two millennia. The fascination lies in the masterful diversity of Hannibal’s tactics, the harmonious concert of such incongruous forces as skirmishers, missile throwers, horsemen, and infantry, all employed in feigned retreat, more-elaborate ruse, and open assault. Both desperate defender and emboldened attacker—be he Frederick the Great or Napoleon himself—have seen an opportunity for a similar masterstroke. It might take the form of a last cast of the die by desperate, outnumbered troops for a sudden reversal of fortune; or of a total annihilation of a stubborn, crumbling foe as he was baited and trapped.

In the 20th century, when the refinement of battlefield communication and intelligence should have made both the illusion and deception that was necessary for a Cannae more unlikely, and thus the art of encirclement inherently less feasible, there nevertheless has been a series of attempts— successful and otherwise—to surround the enemy and knock him out in one decisive blow. The Schlieffen Plan, Tannenberg, the Allied attempt to encircle the Germans at Falaise-Argentan, and the German one to pinch off the Kursk salient (which turned into the greatest tank disaster in history): All invoked Cannae in some such way in the minds of their architects.

Among classicists, too, although the stakes have been hardly as impressive, Cannae has been no less seductive. In hundreds of scholarly articles and not a few books, arguments rage about the exact locations of the Roman and Carthaginian camps, the general deployment and maneuvers of the two armies, the failure of Hannibal to march on Rome in the battle’s aftermath, and the relative accuracy of our major ancient literary sources for the battle— Polybius, Livy, Appian, and Plutarch. Even to the later Romans, Cannae and Hannibal (Juvenal’s “one-eyed commander on his monstrous beast”) were frightening images, finally to become stock topics of declamation in schoolboys’ hackneyed models of rhetoric, the suasoriae and controversiae.

Yet for all this, little attention has ever been paid to the actual battlefield environment at Cannae, specifically to the final experience of the encircled Romans, men sentenced to death. This is all the more remarkable, and ironic, inasmuch as it was the holocaust of these unfortunates that gave Cannae its notoriety and enshrined it as the ancient world’s “masterpiece” of infantry tactics. Actually, it was an event of nastiness unrivaled in its time, an ungodly bloodbath for all involved.

The general course of the battle is relatively straightforward—though the combatants’ positions on the field remain uncertain. The earliest extant history was written some 75 years after the battle. Later accounts sometimes contradict it and each other: The Roman right is on the river in one; Hannibal’s back is to the river in another; and so on. But the sequence of events is much the same in each.

The Romans confidently narrowed their front line and massed unusually deep, more like a Creek phalanx than the fluid squares of Roman maniples. They meant to use their overwhelming numerical superiority (roughly two to one in infantry alone) to smash Hannibal’s center of suspect European allies—in truth, mostly mercenaries and opportunistic freebooters.

The use of such hammer blows was not unforeseen. Hannibal himself, more mindful of Greek than Roman battlecraft, had radically weakened his center by drawing off his superior infantry so that he could stack them to the flanks and rear. The outnumbered Athenians at Marathon, it is true, had used nearly the same tactics 275 years earlier. But there the “weakened” center was still composed of crack Athenian infantry, not unreliable allies, and even then their thin line actually collapsed under the Persian advance; they nearly lost the battle before their enhanced flanks could envelop the Medes from the sides and rear and thereby rescue the victory.

At Cannae, however, only the convex shape of the Carthaginian front (it was bowed outward toward the Romans) and a few thousand choice Africans in reserve on their flanks gave any semblance of stability to the Gauls and Spaniards stationed in the center. Their mission was to contain the advance of the Romans—who were clearly their betters in training, discipline, and armament. They were to bend but on no account break; to slow down and draw in, rather than stop outright, the legionnaires’ onslaught, thereby providing critical time for others more capable to complete the ensnarement from both sides and rear. That way, thousands of surrounded Romans would be unable even to reach the enemy.

However, the Gauls and Spaniards were not mere fodder—lesser allies useful to their African masters mainly as sacrificial bait. The presence of Hannibal and his brother Mago at their sides during the ordeal demonstrates as much, and perhaps explains their admirable stoutness as they gradually gave ground. Real plans had been drawn for their relief and eventual salvation; when the battle began, Hannibal’s other brother, Hasdrubal, with an elite Carthaginian heavy cavalry on the left wing, smashed through the lighter-armed Roman horse in a collision termed “authentically barbaric” by Polybius, our best source of the day’s events.

Hasdrubal not only wreaked havoc at the Roman rear but also soon sent reinforcements to the Numidian cavalry over on the right. That way, once the African horsemen routed their adversaries on both wings, the two cavalry contingents were free to unite behind the Roman army and attack in unison at the legionnaires’ backs. True, they were no match for Roman infantry, but, like the Cauls and Spaniards at the front, that was hardly the role assigned them by Hannibal. Rather, augmented by light-armed skirmishers (who may or may not have been lying in ambush), they were to charge and retreat, to harass with javelin, arrow, and sling at the backs of the Romans. If the legions at the rear turned to face Hasdrubal’s forces, the momentum of the Roman advance would be lost.

Soon the forward pressure of cumulative shields—perhaps 50 in depth—ceased throughout the Roman mass and progress evaporated; the beleaguered Gauls and Spaniards were saved. At that point the seasoned African infantry turned inward, completing the double envelopment. (The overconfident Romans had made this task easy by penetrating into their midst.) However, even then victory, much less virtual ruination of the enemy, on a scale far exceeding Trasimene the year before, was far from assured.

The surging Roman advance had taken few losses as their infantry hammered away at their poorly armored enemy. Livy says the Cauls and Spaniards were on the point of collapse. Consequently, though they were to feel the effects of confusion, the Romans were still a formidable army. Trapped, tired, and directionless, they were nevertheless still heavily armed, more numerous, and ultimately better soldiers than the various corps at their faces, backs, and sides. Their disintegration and destruction were not inevitable, nor perhaps even necessary. To understand the collapse we must envision how awful were conditions inside the ring: Initial incredulity, subsequent confusion, ensuing terror, and ultimate panic eroded the legionnaires’ effectiveness—and eventually their will to resist as an organized body.

For most of the men caught in the interior of the Roman mass, the first sign of impending doom was not any visible inroad by the enemy troops but simply the gradual cessation of their own advance as Hasdrubal’s attacks at the rear sapped the vitality of the march ahead. To the legionnaires, it must have been eerie: a successful onslaught mysteriously gone sour, coupled with the sudden cessation of pressure at their backs. Phantoms of recent disasters at Trebia and Trasimene must have risen in their minds. Yet, was there not also the certainty that this army was somehow different?

Mindful of past blunders and also enormous in number and flush with well-heeled magnificos, the halted legions were undoubtedly surprised, but surely not yet despondent. The inclusion of so many consuls, ex-consuls, senators, and other men of substance throughout the ranks was reassuring at first—just as prisoners in Nazi death camps were occasionally calmed by the presence of notables in their midst; having doctors, lawyers, and professors among their ranks at the gates of the crematoria made indiscriminate butchery seem improbable. However, as the ring tightened, it became clear that there was little hope of salvation for anyone: A consul, Aemilius, would soon perish heroically; Terentius, the architect of the day’s folly by his rash decision to fight where his numerical superiority mattered little, fled shamelessly back to Rome; ex-consuls, quaestors, 29 tribunes, and more than 80 of senatorial rank all perished. (After the battle, Hannibal collected their gold rings by the bushel.)

The paralysis of initial disbelief quickly gave way to universal confusion. The normal din of equipment and men in motion was amplified by shouting: The now arrogant, discordant war cries of Hannibal’s men in a dozen languages mixed with Latin screams of shock and warning as attacker pressed on defender all around the enormous circle. To imagine the cacophony, we need only recall the disaster at Trasimene and Livy’s graphic description of the acoustics among the trapped legions there. He says soldiers turned in every direction to face dissonant clamor; the tumult drowned out orders and advice. The air was full of “the groans of the wounded, the sound of blows to flesh or armor, the mingled shouting of the confident and the terrified.”

As the clamor increased, visibility diminished. Over 100,000 shuffling and stumbling pairs of feet on dry earth kicked up clouds of dust, and strong summer winds blew it directly into the faces of the Romans. All sources remark on the havoc that the sirocco—known locally as the Volturnus— brought to the Romans, and most credit Hannibal with choosing his ground on precisely that account. As in other ancient battles, the atmospherics left many unsure of the exact nature and location of the enemy. Combatants had been similarly befuddled by the ash on Pylos, the dust at Solygeia. Livy wrote of the fog at Trasimene that men failed to recognize their own standards and had no idea where their particular companies had rallied, each forced to rely on hearing rather than sight. But dust was more deleterious than fog since it also produced irritation and pain in the eyes, causing the victims to rub them constantly and thereby occasionally to drop their weapons.

Appian concluded that the Romans at Cannae simply “could not see their enemies.” The density of the throng compounded the problem: Thousands of plumed, bobbing, crested helmets, together with shields raised against missile attacks, blocked the vision—and the efficacy—of the majority in the interior. Their only unimpeded view was of the ground at their feet and the growing number of their own dead and wounded who lay there.

Adding to the confusion was the sudden change in direction, as legionnaires facing ahead in serried formation turned abruptly to the left, right, or rear. The consequences were twofold: entanglement and ensnarement of javelin, shield, arm, and leg; and a total loss of legion cohesion and esprit de corps. Soldiers retreated head-on into their own men, or backed into them, as if movement, even in the wrong direction, were preferable to stationary death. Trampling and accidental wounding—the twin banes of ancient warfare— were frequent. It was not a question of simply a flight to the rear (there was no longer any “rear”) but of a shift in any and all directions.

In a matter of minutes, fluster and agitation were replaced by outright terror—which dashed any lingering hopes of an organized breakout and ended all doubt about the eventual outcome. Into the midst of the melee, missiles of every type showered down. Their target was now encircled, static, and measured, ensuring a high percentage of hits. We should assume here that many Romans stumbled, caught without warning by an arrow, javelin, or stone zooming in through the dust. Wounded in their exposed arms, necks, and faces, they fell, tripping those still standing at their sides. More lethal lance and sword thrusts from African and European infantry at the front and on the flanks met those stationed on the fringe of the ring; while they may have had more room to maneuver with sword or javelin than those in the interior, the growing Roman pressure from within must have spitted many right onto the waiting, edged weapons of the enemy. Remember, the sheer density of the trapped legionnaires nullified their superiority in number: Most simply could not see, much less strike out at, the Africans on the perimeter.

The variegated appearance of Hannibal’s composite foreign alliance, often ridiculed in the prebattle harangues of Western generals, now added to the terror. Both Livy and Polybius remark on the incongruity of the enemy’s weapons, commenting too—as did Caesar much later—on the frightening naked torsos of the Cauls and the dazzling scarlet cloaks of the Spaniards. The battle, indeed, was turning into something most un-Western. Here were no ordered and uniform squares. Screaming hordes circled the Romans, who now must have realized there was no front, rear, left, right, beginning, or end to the fighting.

Far more damaging to Roman morale were the African reserves who poured in on the flanks. Unlike the others, they were crack infantry armed in Roman fashion with weapons and armor taken from the Italian dead at Trasimene and the Trebia. Appian points out that the Romans now had difficulty distinguishing friend from foe; and when a false Roman shield appeared before a legionnaire’s face, his momentary hesitation could prove fatal— especially since the confusion in identification was largely one-sided. There also may have been an element of the macabre as Romans recognized savage attackers clad in the garb of their own dead friends and family. Misdirection and misidentification, even without the exchange of equipment, were inevitable anyway in ancient warfare once advance ceased and the enemy appeared at the flanks and rear, so accustomed were troops to flaying at anything ahead. Consequently the legions, we are told, “considered their situation worse than it actually was.”

Fatigue increased their fright. For over an hour the Romans had pushed back the Cauls and Spaniards. Exhausted now, but still encumbered with over 50 pounds of arms and armor, they saw no reward for their efforts, save for a few enemy corpses at their feet. Worse yet, their new opponents on the flanks were fresh, casting missiles that were wind-aided and, in the dust, seemed to come out of nowhere.

The Romans, however, unlike the trapped Spartan remnants at Thermopylae or the surrounded Athenians at Amphipolis, were not to be executed sterilely through aerial bombardment alone; Hannibal’s infantry, even when the outcome was clear, was never withdrawn. The enormous size of the entrapped Roman army—larger than most cities of the time in Italy—presented a real danger, and thus continuous infantry attack was required to maintain the ensnarement and finish the business before nightfall, which customarily marked the end of ancient battles. The stiff resistance of the well-trained Italians who made up the Roman militia—mostly small farmers drawn from hardy rural stock—angered the African infantry, who were faced with an oddity in their experience: doomed men who would not surrender and wished to kill savagely before they fell. After all, they were desperate troops on the defensive, aware that there were no others left to halt Hannibal’s advance on Rome. There was, too, a racial element in their final frenzy: Most had a deeply rooted dislike and mistrust of Western European and African people.

Ultimately, many had to be dispatched with hand weapons, creating a furious intensity in the last minutes of the battle. For example, the surviving Roman cavalry dismounted, preferring to die on foot and thereby make the Carthaginians pay dearly for their slaughter (between 6,000 and 8,000 Carthaginians died at Cannae). Even when the exterior shell of the Roman circle cracked and the enemy poured in, the disjointed legions continued to offer stubborn opposition. Finally, isolated pockets and islands of brave men were buried under a sea of attackers.

Livy’s graphic postmortem description of the killing field gives us some idea of the brutality during the final, inevitable bloodletting. The day after the battle, the carnage, the human flotsam and jetsam, frightened even the Carthaginian victors who came to view the spectacle. Hannibal, in the tradition of ancient victorious generals (Philip of Macedon at Chaeronea in 338 B.C. comes quickly to mind), is said to have been shocked to tears by the multitude of corpses in the dirt. There were “stacks” of slain infantry, Carthaginians covered by fallen Romans and vice versa, with an occasional horse added to the pile. Because Cannae was a relatively small battlefield, and since the surrounded Roman column was kept static, we must envision an entire field completely covered with bodies, sometimes two or three deep, the ground everywhere soaked red from thousands of oozing dead. Like the scene at Zama later, pools of blood and dead flesh impeded simple movement of the victors.

At the very center of the Roman mass, the Carthaginian scavengers found an even more horrible scene: The sheer compression of struggling bodies, the raucous din, and the inability of half-blinded legionnaires to either confront the enemy or find escape had driven a few absolutely mad. They “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.” Of all the manifold opportunities for death on the ancient battlefield, suicide through asphyxiation seems the most bizarre.

The majority, however, chose neither suicide nor flight and must have fought back savagely to the very end. Unlike Trasimene, which lasted three hours, Cannae took all afternoon. We envision, then, the final standoff as lengthy and exhausting. Without modern weapons of destruction—gas, bullets, bombs, and artillery—the killing required a horrendous effort of muscular strength to drive or throw iron into the armor-clad bodies of thousands, who were themselves desperate to blunt the attack. Croups of two and three slashed away furiously with their swords until they fell—and even then the struggle went on as they wrestled on the ground. The idea of ancient battle as orderly mass dueling with measured blows is absurd. Most often it was a mad free-for-all of pushing, grabbing, flaying away with sword, lance, broken remnant, bare hand, or tooth. Livy tells of a Numidian brought out alive from beneath the corpse of a dead adversary, his ears and nose gnawed away by the raging Roman infantryman who had lost the use of everything but his teeth— and his will to resist.

Some of the 50,000 doomed Romans were still alive 24 hours later. Given the constraints of time and space, Hannibal’s men must have sought to inflict a quick, crippling wound and then to move on to the next victim, confident that the wounded could be easily polished off the next day. The chill of morning revived a number of these, near death, bloodstained, and thirsty. Many of them were suffering from grotesque compound fractures induced by the trampling and stampeding. Some had painful (and nearly always fatal) penetration wounds to the neck and face from incoming projectiles, or less serious, deep lacerations from slashing sword and lance. Others, Livy tells us, had simply been hamstrung by a quick sword cut to the back of the leg, which was a much easier task than direct frontal assault. Like crippled insects, they too thrashed about helplessly, baring their throats to the victors. All were similarly dispatched as Hannibal gave free rein to his men to execute the wounded while they scavenged and rummaged among the dead. The August sun, remember, made it necessary to act quickly if the bloating bodies were to be stripped and collected for burial.

Thousands of Romans fled or were taken prisoner—nearly all of them horsemen, light-armed skirmishers, and some infantry who escaped the encirclement. The struggle, of course, did not end here—it raged on and off for another 14 years. Finally, young Scipio and the orphans of the Trebia, Trasimene, and Cannae brought the war home to Africa and caught Hannibal’s infantry in a ring of their own at Zama (202 B.C.), mercilessly butchering 20,000 Carthaginians in the process and ending for good Hannibal Barca’s ideas of conquest. MHQ

VICTOR DAVIS HANSON is a professor of classics at California State University at Fresno and was an MHQ contributing editor. 


This article originally appeared in the Summer 1990 issue (Vol. 2, No. 4) of MHQ—The Quarterly Journal of Military History with the headline: Cannae


Want to have the lavishly illustrated, premium-quality print edition of MHQ delivered directly to you four times a year? Subscribe now at special savings!