What happened at the Battle of Pydna?
In 168 BC the consul Lucius Aemilius Paullus turned King Perseus of Macedon out of his fortified line at the foot of Mount Olympus and followed him to the plain south of Pydna. On the afternoon of the day after a lunar eclipse, usually dated 22 June, a skirmish over a river turned into a full battle. The Macedonian phalanx drove the first Romans back, but on uneven ground it opened gaps, Roman cohorts pushed into them, and within about an hour the phalanx was broken. Perseus fled and later surrendered, and Rome abolished the kingdom of Macedon.5
Polybius ended the fifty-three years in which, as he saw it, Rome won mastery of the inhabited world with the fall of the Macedonian monarchy, and Mommsen called Pydna the last battle in which a civilised state met Rome in the field on equal terms. How the phalanx and the legion compared across all their meetings is set out in our comparison of the phalanx and the legion, and the wider story in our guide to the military history of ancient Rome.6
The last king of Macedon
Perseus was the son of Philip V, the king whom Titus Flamininus had beaten in 197 BC in the hills near Scotussa, the battle better known as Cynoscephalae. Plutarch writes that Philip spent his last years quietly preparing a new war, and that Perseus inherited both the kingdom and his father’s hatred of Rome.7
By Plutarch’s account the old king had laid up arms for 30,000 men, grain in his fortresses and money enough to pay 10,000 mercenaries for ten years. Diodorus, writing in the first century BC, remarks that Perseus had great treasures and would not spend them.8
Three years without a decision
Rome declared war in 171 BC. At the start Perseus reviewed his army at Citium, and Livy, here following Polybius, counts 43,000 men, the largest Macedonian army, he says, since the one Alexander took into Asia. Kromayer and Veith, who analysed the review line by line in 1907, put almost half of it in the phalanx.9
The first Roman commanders did badly. Plutarch credits Perseus with a cavalry victory over Publius Licinius in which 2,500 Romans were killed and 600 captured, and Velleius Paterculus writes that for two years the king generally had the better of the consuls and drew a large part of Greece to his side. Livy and Diodorus both describe the war as lasting four years in all.10
Aemilius Paullus takes command
For 168 BC the Romans elected Lucius Aemilius Paullus consul for the second time and gave him the Macedonian war without the usual drawing of lots. He was about sixty, as Plutarch and Diodorus both say, and Velleius adds that he was the son of the Paullus who died at Cannae. Mommsen describes him as a general of the old school, strict with himself and his men, and a man of Greek culture.11
Davide Morelli, who reconstructed the campaign in the journal Klio in 2021, puts the Roman army at a little over 35,000 men and the Macedonian at about 40,000, counting from the reinforcements Livy lists year by year. Gaetano De Sanctis, writing for the Enciclopedia Italiana in 1935, gave the consul 30,000 to 35,000 men in two full-strength legions with allies, and Perseus over 40,000, more than 20,000 of them in the phalanx.12
Deadlock on the Elpeus
Paullus found Perseus dug in on the river Elpeus at the foot of Mount Olympus, with the sea at his back and his line fortified with wooden outworks. Plutarch gives the king 4,000 horse and not much fewer than 40,000 foot, and Kromayer thinks that figure is simply the review of Citium repeated. Morelli identifies the Elpeus with the stream now called Enipeas, which runs past Litochoro about five Roman miles south of ancient Dion.13
The Roman camp was short of water until Paullus, Plutarch says, judged from the green slopes of Olympus that water ran underground and had wells dug. A frontal assault on the river line looked hopeless, and De Sanctis writes that the consul kept Perseus busy with skirmishes and feints, as if he meant to land behind him from the fleet. Zonaras, who epitomised the lost books of Cassius Dio, says the king had walled off the whole strip between Olympus and the sea and counted on the lack of water.14
Nasica goes over the mountain
One pass was left unguarded, through Perrhaebia past the Pythium. Publius Scipio Nasica and the consul’s eldest son, Fabius Maximus, volunteered to lead a force round. Plutarch says he follows the numbers Nasica gave in a letter to a king rather than the larger ones in Polybius: 3,000 Italian allies, the 5,000 men of the left wing, 120 horsemen and 200 Thracians and Cretans.15
A Cretan deserter warned Perseus, who sent 10,000 mercenaries and 2,000 Macedonians to hold the heights. Here the sources split. Polybius, as quoted in the Suda and by Plutarch, says the Romans caught them asleep. Nasica says there was a sharp fight in which he killed a Thracian with his own spear, and Livy follows that version. Either way the pass was lost. Fearing an attack from behind, or a Roman landing at Pydna from the fleet sailing along the coast, Perseus abandoned the river line and fell back to camp in front of the town.16
The plain before Pydna
Perseus chose to fight. Livy and Plutarch describe the ground in almost the same words: a plain fit for the phalanx, which needs firm and level footing, hills nearby where light troops could fall back and attack from the flank, and two rivers, the Aeson and the Leucus, low with the summer but still an obstacle to the Romans.17
Where exactly this was is still argued. Kromayer, who rode over the ground, put the battle south of modern Katerini and took the Leucus and Aeson to be the Mavroneri and its tributary the Pelikas. N. G. L. Hammond moved it in 1984 to the plain south of ancient Pydna and east of Kitros, and Morelli accepts that area but suggests that the two rivers were the Alkovitsa and the Agios Dimitrios. De Sanctis placed the Macedonian camp just south of the town on the left bank of the Leucus.18
A day of waiting
When Paullus joined Nasica and came down into the plain he found the Macedonian army already drawn up, and he halted. Nasica urged him to attack at once. Plutarch has the consul answer with a smile that many victories had taught him the mistakes of the vanquished and forbade him to fight straight off the march against a phalanx already formed. In Livy the reason is the heat: it was near noon, the men were tired and thirsty after a dusty march, and some stood leaning on their shields.19
Paullus had the front ranks keep up the appearance of a battle line while those behind marked out a camp, and then withdrew the army into it unit by unit, so that the enemy hardly noticed. Livy says Perseus was content to let the day pass. Morelli reads the whole sequence as deliberate: in his view the consul refused battle on ground he did not like and chose another field for the next day.20
The eclipse
That evening, by Livy’s account, the military tribune Gaius Sulpicius Gallus told the soldiers that the moon would be eclipsed that night from the second hour to the fourth, and that this was a natural event and no portent. When it happened the Romans thought Gallus almost divine, while the Macedonians took it as a sign of the fall of their kingdom. Polybius, in a surviving fragment, says the eclipse was popularly taken to portend the eclipse of a king, and that it encouraged the Romans and disheartened the Macedonians. In Zonaras it is Paullus himself who warns the army.21
Plutarch does not mention Gallus. In his version the Romans clashed bronze and held up torches to call the moon’s light back, and Paullus sacrificed eleven heifers to the moon as it emerged. In the morning he sacrificed twenty oxen to Hercules without a good sign, and only the twenty-first promised victory if the Romans stood on the defensive.22
The eclipse dates the battle. NASA’s Five Millennium Catalog of Lunar Eclipses lists a total eclipse of the moon on 21 June 168 BC, and Mommsen, De Sanctis and Morelli all date the battle to the next day, 22 June. Livy gives the Roman date, the night before the day before the Nones of September, and Morelli explains the gap by the Roman calendar, which was then badly out of step with the sun. Bill Thayer, the editor of the LacusCurtius texts, dissents: he argues that the 168 BC eclipse began before moonrise at Pydna and does not fit Plutarch’s moon high in the sky, and prefers a nearly total eclipse of 2 September 172 BC.23
A loose horse and a fight at the river
Neither commander wanted to attack across the river, and the battle began by accident. Livy says that about the ninth hour a horse broke loose and ran towards the far bank, Roman soldiers waded after it through water up to their knees, and Thracians from the Macedonian outposts tried to seize it. One Thracian was killed, his comrades crossed to avenge him, and more and more men joined in from both sides.24
Some writers, Livy adds, claimed that Paullus drove an unbridled horse towards the enemy on purpose, because the omens promised victory only if the Romans did not strike first. Plutarch reports that version and another, in which Thracians under one Alexander attacked Roman pack animals bringing in forage and 700 Ligurians sallied out against them. Zonaras has a third version, a Roman pack animal that fell into the river where both sides drew water. De Sanctis and Kromayer both treat the outbreak as a clash of outposts that neither king nor consul had planned.25
The phalanx comes on
Nasica, sent forward to look, saw the Macedonian army coming out in order. Plutarch preserves his description: first tall Thracians in black tunics with white shields and greaves, swinging heavy iron weapons on their shoulders, then mercenaries of every kind with Paeonians among them, then picked Macedonians in gilded armour and fresh scarlet cloaks, and behind them the phalanx of the Bronze Shields pouring out of the camp. Livy names the two halves of the phalanx, the White Shields and the Bronze Shields.26
The Macedonians advanced so fast that, Plutarch says, the first men were killed two furlongs from the Roman camp, and Kromayer works that out at about 350 metres. Polybius reports that Paullus had never seen a phalanx before and often said afterwards that he had seen nothing more terrible. Plutarch and Livy both tell how, hiding his fear, he rode along the line without helmet or breastplate.27
The first Romans to meet the pikes were the Paeligni. Their commander, Salvius in Plutarch and Salius in Livy, seized a standard and threw it among the enemy, and his men rushed in to recover it. The Romans tried to hack through the long spears, push them aside with their shields or even grasp them with their hands, but the Macedonians drove them through shield and armour. The front ranks were cut down and the rest fell back towards a hill called Olocrus, and Paullus, by Posidonius’ account, tore his cloak.28
Into the gaps
What saved the Romans was the ground. Plutarch and Livy explain it the same way: the line was very long and the plain uneven, so the phalanx could not keep its shields locked. Men climbing higher ground separated from those below, the fast from the slow, those pressing forward from those held back, and gaps opened along the front. Paullus ordered his units to make for the gaps wherever they appeared and to fight many separate battles instead of one.29
Livy gives the deployment that followed. Paullus led the first legion into the space between the Macedonian targeteers and the phalanx, the consular Lucius Albinus took the second legion against the White Shields in the centre, and on the right, where the fight had begun at the river, elephants and Latin allies drove back the Macedonian left. Livy thought the elephants did little, but the allies behind them broke that wing first. Once the Romans were inside the formation the pikemen, with only small daggers and light shields, could not stand against the legionaries’ long shields and heavy swords.30
Livy’s summary has become the classic verdict on the phalanx: irresistible while it stayed compact with its spears levelled, but once forced to turn its long, heavy pikes it fell into confusion, and an attack on its flank or rear threw it into irretrievable disorder. Had the Romans charged it head on all along the line, he adds, they would have impaled themselves as the Paeligni did. Kromayer saw the same pattern as at Cynoscephalae, where an uneven advance had also carried the seeds of defeat.31
Frontinus, writing under the emperors, tells it as a planned trick: Paullus fell back on purpose into broken country and then sent his cavalry from the left wing galloping along the front of the phalanx to snap its pikes. Kromayer thought this an exaggeration that might still preserve a real flank attack by the Roman horse. Morelli goes further than anyone in crediting the consul, calling Pydna one of the masterpieces of Roman tactics, fought on ground and in a deployment that Paullus had chosen.32
Rout
The Macedonian cavalry hardly fought. Livy writes that it left the field almost untouched, the king first in flight with the sacred squadron on the road to Pella, followed by Cotys and his Odrysian horse. Polybius, as Plutarch quotes him, says Perseus rode off to the town at the very start, pretending he was going to sacrifice to Heracles. Plutarch also gives the defence of a Posidonius who claimed to have been there: the king had been kicked by a horse the day before, still joined the phalanx without a breastplate, and was bruised by an iron javelin.33
Left alone, the infantry was destroyed. Plutarch says the 3,000 picked Macedonians who kept their ranks were cut down to a man. Livy describes fugitives running into the sea and begging the crews of the Roman ships for their lives, only to be killed from the boats, and those who swam back being trampled by elephants on the shore. Zonaras says that only nightfall saved the rest.34
Plutarch says the Romans began fighting at the ninth hour, about three in the afternoon, won within an hour and pursued for 120 furlongs until dark. Kromayer measured about 17 kilometres from his battlefield to Pydna and took the figure for a slight overestimate. That night the younger of the consul’s two sons in the army, the future Scipio Aemilianus, was missing until he rode in late from the pursuit.35
Counting the dead
Livy says the Macedonians had never lost so many men in any battle with the Romans: 20,000 killed, 6,000 who fled into Pydna and were taken there, and 5,000 caught straggling across the country. He puts the Roman dead at not more than 100, most of them Paeligni, with many more wounded. Plutarch gives over 25,000 Macedonians killed, and for the Romans 100 according to Posidonius or 80 according to Nasica.36
These numbers are less independent than they look. De Sanctis explains that three contemporaries wrote about the battle: Polybius, who was not present, Nasica, in his letter, and Posidonius. Livy’s account is in substance Polybius’ version, while Plutarch mixes Polybius with Nasica and Posidonius. Polybius’ own figures are lost, and Morelli notes that Livy used neither Nasica nor Posidonius. Mommsen took Livy’s total, 20,000 dead and 11,000 prisoners, and De Sanctis rounded it to over 20,000 dead against about 100 Roman.37
The end of the kingdom
Perseus reached Pella in the night and killed two of his treasurers with his own hand when they rebuked him. He went on by Amphipolis to the temple of the Dioscuri on Samothrace, failed to escape on a Cretan’s boat and surrendered to the praetor Gnaeus Octavius. Velleius says he was persuaded rather than forced to give himself up, and Zonaras that he was brought to Amphipolis, where Paullus received him at his table.38
Plutarch and Mommsen both write that all Macedonia submitted within two days. When ten commissioners arrived from Rome, Paullus announced the settlement to the Macedonian leaders at Amphipolis: the Macedonians were to be free, to keep their cities and laws and to pay Rome half the tax they had paid their kings, and the country was split into four districts. Plutarch puts the tribute at 100 talents, less than half the old sum, and De Sanctis notes that Pydna itself fell into the third district, the one governed from Pella.39
Polybius made 168 BC the end point of his main history, the moment when the fifty-three years of Roman growth were complete and everyone accepted that they must obey Rome. The settlement did not last. De Sanctis records that in 148 BC a pretender, Andriscus, raised Macedonia against Rome, was defeated near Pydna by Quintus Caecilius Metellus, and that Macedonia was then made a Roman province.40
The pillar at Delphi
At Delphi Paullus found a tall square pillar of white marble built to carry a golden statue of Perseus, and Plutarch says he ordered his own statue placed on it instead, since the conquered should make room for the conquerors. The Archaeological Museum of Delphi, which now holds the monument, calls its frieze the earliest relief illustrating a historical narrative, with scenes from the Battle of Pydna.41
Michael Taylor, who studied the 29 figures of the frieze in 2016, concludes that the scene shows the Romans everywhere triumphant and may have been meant to recall the pursuit and slaughter of Perseus’ routed army. Velleius writes that Paullus led the greatest of kings in a triumph that outshone all before it, and puts the money he paid into the treasury at 200 million sesterces. Zonaras says Perseus marched in the procession in chains with his diadem on. He was held at Alba on the Fucine lake, where Mommsen has him die a state prisoner a few years later and Zonaras says he took his own life.42
