In 202 BCE, the Roman commander Publius Cornelius Scipio defeated Hannibal at the Battle of Zama in North Africa. Hannibal had spent sixteen years campaigning in Italy, where he inflicted some of the worst defeats in Rome’s history. By the time the two commanders faced each other, Scipio had carried the war to Carthaginian territory. His victory forced Carthage to accept a peace that stripped away its overseas empire.

At Zama, Hannibal’s infantry put up a harder fight than the usual comparison with Cannae might lead readers to expect. Roman infantry held the enemy in front, and Roman-allied cavalry eventually attacked from behind, but Hannibal’s first line wounded and killed many Romans. His second line then became embroiled in fighting with his own retreating troops. The veterans he had brought from Italy resisted so firmly that Scipio had to halt and reorganize his infantry before the cavalry returned to decide the battle.

The victory depended in part on decisions Scipio had made before the armies met. He had invaded Africa, gained the support of the Numidian ruler Masinissa and compelled Hannibal to fight with a varied army short of cavalry. Those advantages did not guarantee a quick victory. The ancient narratives describe a difficult battle whose result depended on sustained infantry fighting as well as the final mounted attack.

Why Hannibal Had to Fight in Africa

Hannibal entered Italy over the Alps in 218 BCE. He defeated Roman armies at the Trebia, Lake Trasimene and Cannae, where the Roman losses in 216 BCE were especially severe. Rome nevertheless refused to surrender. Its commanders increasingly avoided major battles on Hannibal’s terms, struck at his Italian allies and continued campaigns in Spain and Sicily. The Second Punic War also turned on the loyalties of allies and the armies’ access to recruits, money and supplies, even while Hannibal remained a formidable battlefield commander.

Scipio made his reputation in Iberia after Rome had suffered serious setbacks there. In 209 BCE he captured New Carthage, an important Carthaginian stronghold, and followed that success with further victories. Carthaginian control in Iberia had largely collapsed by 206 BCE. Losing the region cost Carthage both wealth and access to recruits. Scipio, meanwhile, had the experience and political standing to argue for taking the war to Africa. The account of the campaign in Simon Hornblower’s Hannibal and Scipio (2024) stresses how bold that proposal remained while Hannibal was still in Italy.

Scipio landed in Africa in 204 BCE, but his first attempt to capture Utica failed. In 203 BCE he defeated Carthaginian and Numidian armies in attacks around their camps and at the Great Plains. Among those who lost power was Syphax, a Numidian king allied to Carthage. His defeat benefited Masinissa, the Numidian leader who had joined Scipio. Masinissa would supply cavalry at Zama. His decision to support Scipio also gave Rome a powerful local ally whose ambitions ran against those of Carthage.

Carthage summoned Hannibal home from Italy in 203 BCE. He brought experienced soldiers, although the army awaiting him in Africa was assembled from contingents with different backgrounds and loyalties. The veterans from Italy formed only part of his new army, which differed substantially from the force he had commanded at Cannae. Negotiations with Scipio initially produced peace terms, but the truce failed after Carthaginian forces seized Roman supply ships and mistreated Roman envoys. The incident is described in Polybius, Histories 15.1-4. With the fighting renewed, Carthage depended on Hannibal to win a battle and improve its position.

Where and When Was the Battle of Zama Fought?

The Battle of Zama was fought in 202 BCE in North Africa, within modern Tunisia. Archaeologists have yet to locate the battlefield securely, despite its familiar name. Polybius places Hannibal’s camp near a town called Zama and describes his later movement toward Scipio, who was encamped near Naragara. Livy also associates the encounter with the Naraggara district. Neither the place names nor the distances given in the ancient texts point conclusively to a single modern site.

The date is uncertain as well. Although 19 October 202 BCE is often given as the battle’s precise anniversary, the surviving ancient accounts do not establish that date independently in a modern calendar. We can date Zama securely to 202 BCE and probably to the later campaigning season. It may have been fought some distance from the town of Zama itself.

Modern maps of Zama therefore involve a degree of reconstruction. Ancient writers indicate fairly open ground, where cavalry could move freely and pursue a defeated enemy. Polybius also contrasts the camps’ water supplies: Scipio had water conveniently close, while Hannibal’s troops had to travel farther. These features help explain conditions faced by the armies. They do not allow us to identify a particular plain or assume that Scipio had anticipated every event when he chose his position.

Battle of Zama
The Battle of Zama, oil on panel, 1567–1578, showing Scipio’s forces engaging Hannibal’s army. Source: Art Institute of Chicago
Advertisements

Hannibal's Meeting With Scipio

Polybius and Livy both say that Hannibal met Scipio to discuss peace before the fighting. In Polybius’s version, Hannibal warned the younger Roman about the reversals of military fortune and proposed surrendering Carthage’s claims to Sicily, Sardinia and Spain. Scipio answered that Carthage had already agreed to harsher terms before breaking the truce. He saw no reason to offer a better settlement now that the war had resumed.

There may well have been a meeting. The narratives describe interpreters and a small escort, details consistent with negotiations between rival commanders. The speeches the historians give them are less secure. Greek and Roman historians frequently composed speeches to clarify decisions or illustrate character, and no transcript of this conversation survives. We can understand why the talks failed without treating the words attributed to either general as a verbatim exchange.

Hannibal was under pressure to defend a homeland now threatened by invasion. Carthage had lost allies, and its leaders looked to his army to strengthen their negotiating position. Scipio approached the talks from a stronger strategic position, though a defeat would have endangered the Roman expedition in Africa. Each had something to gain from peace, but they reached no agreement. The armies prepared to fight.

How the Two Armies Were Arranged

Scipio arranged his heavy infantry in the Roman manipular system. The hastati formed the first line, the principes the second and the triarii the third, with the light-armed velites in front. Usually the units, or maniples, in one line were positioned opposite spaces in the line ahead. At Zama, Scipio lined up those spaces into passages extending through the formation. The purpose was to give the elephants a route through the Roman front while keeping the heavy infantry organised enough to fight afterward.

Cavalry covered both Roman flanks. Gaius Laelius commanded the Italian horsemen, and Masinissa led the Numidian cavalry. According to Polybius, Masinissa had recently joined Scipio with 4,000 cavalry and 6,000 infantry, although these figures do not establish an exact total for every Roman and allied soldier at Zama. Modern estimates of the armies’ overall strength differ because the ancient numbers must be interpreted and reconciled. Scipio did, however, have more effective cavalry available than Hannibal, whose mounted support was unreliable.

Hannibal deployed three infantry lines of his own, but they differed markedly in composition. First came troops drawn from several regions, including Ligurians, Celts, Balearic Islanders and Moors. Polybius calls them mercenaries, a broad term that can obscure the difference between paid recruits, allied forces and men serving through political obligations. Behind them stood Libyan and Carthaginian infantry. At a distance in the rear Hannibal placed the veterans brought from Italy, whose years of experience against Roman armies made them his most reliable troops.

By holding the Italian veterans back, Hannibal could commit them after the Romans had already fought through the forward lines. That offered a chance to meet an exhausted enemy with relatively fresh and disciplined troops. The gaps between these formations would prove dangerous when the forward troops tried to retreat. The men in front might expect help from those behind, while the rear lines had reason to fear that fugitives would carry panic into their ranks. The problem emerged in the middle of the battle.

Polybius gives Hannibal more than eighty elephants, placed ahead of his infantry. We cannot independently confirm that total, or the frequently repeated claim that most of the animals had been caught recently and were poorly trained. Their strength made a charge dangerous to an opposing formation, but frightened elephants could also damage their own troops. Noise, thrown weapons and horses added to the difficulty of controlling them. Hannibal’s war elephants explains the animals’ wider military use and why so much about their training remains uncertain.

Scipio's Answer to the Elephants

Fighting had already begun between Numidian cavalry when Hannibal sent the elephants forward. Some were frightened by trumpets and the surrounding noise, turned toward the Carthaginian left and disturbed its allied Numidian horsemen. Masinissa seized the opportunity. Other animals continued toward the Roman front, where the light infantry threw missiles at them before withdrawing through the gaps Scipio had prepared.

The elephants did not all pass neatly through the prepared corridors. In Polybius 15.12, elephants wounded skirmishers and were wounded themselves. Some moved through the intervals in the Roman ranks. Others swerved to the right, where cavalry attacked and drove them away. Scipio’s passages limited the danger to his heavy infantry, but the episode involved casualties, confusion and animals behaving unpredictably.

Even an elephant that failed to break an infantry line could pull the soldiers out of position. Men avoiding a charging animal might crowd together, abandon their alignment or obstruct comrades behind them. Scipio gave his light troops somewhere to retreat and allowed the heavier formations to stay together. It was an effective solution at Zama, though this particular success cannot be taken to mean that elephants had lost all military value in other circumstances.

The animals’ disorder affected the cavalry battle too. Laelius attacked Carthaginian mounted troops who were already unsettled, while Masinissa pursued the Numidians fighting against him. Both forces chased their opponents from the immediate battlefield. Their pursuit, however, left Scipio without cavalry for much of the struggle that followed. His infantry would have to withstand Hannibal’s main force before its mounted allies could come back.

Advertisements

The First Infantry Clash Was Not an Easy Roman Victory

The infantry met as the elephants dispersed and fighting continued on the flanks. Polybius describes a severe contest from the outset. Hannibal’s first line resisted and inflicted heavy wounds on the Romans before it finally gave ground. Many of these soldiers had fighting experience despite their different backgrounds. Their resistance seriously tested Scipio’s leading infantry.

The retreat became a crisis for Hannibal when the men in his first line reached the second. The Libyan and Carthaginian troops behind them would not let them pass. Polybius says the men trying to escape began fighting those who barred their way, even as Roman soldiers pressed after them. Scipio then brought the principes forward to help the hastati. Amid the disorder within Hannibal’s army, the first and second infantry lines were defeated.

Hannibal’s arrangement now worked against him. The three lines differed in loyalty and experience, and the spaces between them made it difficult for one body of infantry to replace another in good order. Men fleeing from the front threatened to spread disorder to the veterans at the back. Hannibal had those veterans level their spears at the retreating soldiers, who had to escape toward the sides rather than enter the last line.

Battle Zama painting Scipio Hannibal Carthage Renaissance art
The Battle of Zama, oil on panel, c. 1470, depicting the chaos of the engagement. Source: Minneapolis Institute of Art

Hannibal's Veterans and Scipio's Rearrangement

Scipio’s infantry had now beaten two Carthaginian lines, but Hannibal’s force from Italy was still waiting. These men had remained outside the heaviest fighting and retained the advantages of experience and comparatively fresh strength. Against them the Romans faced another serious infantry engagement, even after the earlier Carthaginian formations had broken.

Scipio could not march straight into that last contest. Bodies, wounded men and abandoned weapons covered the ground between the armies. Polybius notes that the debris made orderly movement difficult. Scipio recalled troops who had pursued the retreating enemy, arranged for the wounded to be taken away and reorganised his heavy infantry. The hastati remained in the middle, with the principes and triarii brought out to lengthen the line on either side.

Scipio used the pause to restore the order his battered infantry needed for another fight. Scipio had men from different lines who had just endured a confused and bloody struggle. Bringing them together into a new front gave them a better chance against Hannibal’s strongest formation. Roman infantry had already survived one hard engagement; it now had to fight again, without simply waiting in safety for the cavalry to return.

Polybius describes the clash with Hannibal’s veterans as evenly balanced for a time. Their weapons and experience allowed them to stand against the reorganised Romans, and neither side quickly broke. The elephants’ failure and the defeat of the earlier lines had not settled the battle. It was still being fought between opposing bodies of heavy infantry when the cavalry came back.

Masinissa, Laelius and the Attack From Behind

Masinissa and Laelius had driven the opposing cavalry from the field and pursued it. When their horsemen returned, they found Hannibal’s final infantry line occupied with Scipio’s Romans in front. The returning cavalry struck the Carthaginians from behind. Pressed between mounted attackers and the Roman heavy infantry, Hannibal’s veterans could no longer sustain the defence that had held up the frontal assault.

The cavalry attack settled the fighting, but Scipio’s infantry had made that attack effective by keeping the Carthaginians engaged. If Hannibal’s men had broken through the Roman centre, the returning horsemen would have faced a different battle. Without the cavalry, the infantry struggle against Hannibal’s last line might have continued. Scipio’s cavalry advantage counted because his infantry had absorbed losses, reformed and held Hannibal’s veterans in place.

Masinissa was central to that advantage. His Numidian horsemen knew the North African campaigning environment, and his alliance with Rome denied Carthage cavalry support of the kind Hannibal had used so successfully in Italy. Masinissa also had his own territorial and political interests, which made his choice of ally consequential beyond the day’s fighting. Scipio benefited from Numidian power as well as from Roman organisation and his own commands.

Scipio Africanus painting Tiepolo Walters Museum diplomatic scene
Scipio Africanus Freeing Massiva, oil on canvas, 1719–1721, showing Scipio’s diplomacy with African allies. Source: Walters Art Museum

Zama is sometimes described as Scipio turning Hannibal’s double envelopment at Cannae against its inventor. At both battles, mounted troops attacked an enemy already fighting infantry to its front. But at Cannae Hannibal’s design drew Roman soldiers inward while his flanks and cavalry closed around them. At Zama Roman cavalry first drove its opponents away, while Scipio’s infantry fought through two Carthaginian lines, regrouped and engaged a third. The eventual attack on Hannibal’s rear resembled the result at Cannae more closely than the process by which it was achieved.

Advertisements

What Happened to Hannibal's Army?

Hannibal escaped with a small escort and made his way to Hadrumetum, a coastal town in what is now Tunisia. Polybius estimates that some 20,000 Carthaginians died and almost the same number were captured, compared with more than 1,500 Romans killed. Such numbers come from ancient narratives, not from a documented casualty count, and should not be read as exact statistics. The scale of the defeat is nonetheless clear: Hannibal no longer possessed a field army capable of continuing the campaign.

After the cavalry attack, the Carthaginian army broke and was pursued across the open country. No surviving account securely fixes the hour when the battle ended, despite later descriptions that give a mid-afternoon conclusion. Hannibal’s escape allowed him to return to Carthaginian politics and offer advice. It could not replace the army whose victory had been expected to win his city better peace terms.

Polybius still treats Hannibal as an able commander. In his assessment of the general in Histories 15.15-16, he regards Hannibal’s deployment as carefully judged and gives weight to the effectiveness of Roman arms and military organisation. For Polybius, the defeat does not demonstrate that Hannibal had suddenly forgotten how to command. Polybius’s explanation gives substantial weight to the forces available to the two commanders and the way the battle unfolded.

The Peace of 201 BCE

Carthage’s army had been defeated in 202 BCE, but the Second Punic War ended formally with a settlement in 201 BCE. Zama left Carthage unable to continue effective resistance. The city kept its local laws and a defined part of its African territory, while surrendering its overseas possessions and much of the freedom to make war independently.

Under the treaty, Carthage handed over its war elephants and all but ten warships. It also agreed to pay Rome 10,000 talents of silver over fifty years, give hostages and obtain Roman permission before undertaking warfare. Masinissa’s territorial claims received recognition, storing up further disputes between Carthage and its Numidian neighbour. Polybius 15.18 records these terms, which also appear in Livy’s account.

Carthage continued to exist and, in the decades after Zama, regained substantial economic strength. It was not destroyed in 202 BCE. Nor did Roman victory immediately deliver mastery over every part of the Mediterranean: expansion into the Greek east involved later wars. The settlement nevertheless deprived Carthage of its position as an independent overseas military rival and gave Rome a much stronger hold on the western Mediterranean.

Hannibal’s career continued after the treaty. He remained involved in Carthaginian affairs before leaving his homeland and living for years in exile. His eventual death took place far from Carthage. The later stories surrounding Hannibal’s death concern Rome’s dealings with its old enemy after Zama.

Scipio Africanus painting George Smith Victoria Albert Museum
Scipio Africanus receiving his son, oil on canvas, 1832, showing a later episode from Livy’s History of Rome. Source: Victoria and Albert Museum

What Can the Ancient Accounts Actually Tell Us?

Our fullest surviving description comes from Polybius, who wrote in the second century BCE, several decades after Zama. He was relatively close to the period, although his admiration for Roman institutions and his links to the Scipionic circle matter when weighing his account. Livy wrote under Augustus, much later, using older histories to shape a narrative for Roman readers. Appian wrote later still; his version adds details, including a personal clash between Scipio and Hannibal, that cannot safely be treated as the record of an eyewitness.

Across these accounts, the principal stages of the battle are reasonably consistent: negotiations failed, elephants charged in confusion, Roman-allied cavalry drove off its opponents, the infantry fought bitterly and cavalry eventually attacked Hannibal’s rear. The exact day, location, numbers of troops, speeches and casualty estimates are much harder to establish. No securely identified and excavated battlefield is available to confirm each episode independently of the written narratives.

Scipio won at Zama because his wider campaign had brought Hannibal into a difficult position, and because his soldiers held together when the battle became dangerous. The Numidian alliance gave Rome the cavalry advantage, Scipio’s passages helped protect his infantry from the elephants, and his redeployment enabled the Romans to confront Hannibal’s Italian veterans. Only after that sustained struggle did Masinissa and Laelius return to strike the Carthaginian rear. The final cavalry attack succeeded because the Roman infantry had survived long enough to keep Hannibal’s veterans engaged.