The Battle of Thermopylae (480 BCE)

In the late summer of 480 BCE the army of Xerxes came up against a narrow coastal pass in central Greece called Thermopylae, the 'Hot Gates', where a small allied Greek force under the Spartan king Leonidas barred the road south. For some days the Greeks held the defile; then a Persian column, guided by a local man over a mountain track, came down behind them, and the defenders who stayed were surrounded and killed. The King's road into central Greece was open, and within weeks Athens itself was taken and burned. Thermopylae is, without rival, the most celebrated defeat in Western memory: the heroic last stand, the '300 Spartans', the free men who chose death over the despot's yoke. That reading is Greek, and it is very old. This entry sets beside it the harder fact that governs the whole subject of these wars. Thermopylae was a Persian victory, a successful forcing of a fortified chokepoint, and it is known to us almost entirely through the people whom the King defeated there, and above all through one of them, Herodotus, writing two generations later with a purpose that shaped every line. To recover the battle from the Persian side is not to deny the Spartans their courage but to ask what the fighting actually was, why it loomed so vastly larger in Greek memory than in imperial fact, and how a court that treated defeat as unthinkable would have understood a chokepoint stormed at some cost and the enemy's forward army wiped out.

The one source, and its design

Everything of consequence about Thermopylae comes from Herodotus, whose seventh book narrates the campaign at length and gives the battle some thirty chapters (7.201-233). Later writers, Diodorus (drawing on the fourth-century Ephorus), Plutarch, Ctesias, add embroidery from the same Greek stream; none is independent of it, and the fourth-century additions, such as Leonidas' night raid on the King's tent, are demonstrably fictional.^[1] There is no Persian account of any kind. The reference-standard survey of Greco-Persian relations states the resulting dependence flatly: for the political and military history of these wars 'we are dependent on Graeco-Roman writers, with Herodotus' masterly narrative dominating the picture.'^[2]

That narrative has a shape, and it is the shape of a moral. For Herodotus the invasion is the punishment of overreaching pride and the vindication of Greek freedom, and Thermopylae is the pattern-piece of the whole: a handful of free men, disciplined by their own law, holding at bay a king who can move millions but command no true soldiers. John Hyland, whose recent study is the fullest treatment of the campaign from the Achaemenid side, shows how the account of the advance from Thessaly to Athens is built as 'a litany of escalating setbacks, bad omens, overlooked warnings', in which Xerxes' soldiers 'fall in countless numbers at Thermopylai, displaying their lack of manliness and requiring the whip to push them into combat', while 'all Xerxes's successes are ephemeral, and divine intervention is only credited in support of Athens and its allies.'^[1] The reader must hold this at every point: the facts Herodotus reports are often sound, but the frame around them is a Greek argument, and at Thermopylae the argument is at its most seductive and least neutral.

Not 300 Spartans: the allied Greek force

The first correction is to the cast. The 'three hundred Spartans' of legend were the Spartan citizen core of a much larger allied army, and they were never alone. Leonidas led a coalition contingent drawn from across central Greece and the Peloponnese: besides his 300 Spartiates, several thousand Peloponnesian hoplites, some seven hundred Thespians who died with the Spartans to the last, four hundred Thebans, Phocians posted on the mountain, and Locrians, with helots and other attendants swelling the number. Herodotus' own casualty notice at the end preserves the reality behind the slogan: he speaks of 'four thousand' of the Peloponnese who fought there, and Hyland reasonably reads the actual last-day dead as something on the order of two thousand men, including the famous 300 and the 700 Thespians, plus perioikoi, helots and a handful of others.^[1] The reduction of this allied host to '300 Spartans' is itself a piece of the later memory, a Spartan and then a pan-Hellenic compression that wrote the other cities out of the story.

The defenders' aim was limited and, from the coalition's side, provisional. Thermopylae was a delaying position, a narrow gate between mountain and sea chosen to blunt the King's advance while the Greek fleet fought the parallel action off Artemisium and the states behind decided whether to make a stand at all. Many of the Peloponnesians wanted to fall back to the Isthmus of Corinth and abandon everything north of it. Herodotus records the quarrel in Leonidas' camp, and the Spartan's decision to hold:

"To the rest of the Peloponnesians then it seemed best that they should go to the Peloponnese and hold the Isthmus in guard; but Leonidas, when the Phokians and Locrians were indignant at this opinion, gave his vote for remaining there, and for sending at the same time messengers to the several States bidding them to come up to help them, since they were but few to repel the army of the Medes." (Herodotus 7.207, trans. Macaulay)^[3]

The battle: the chokepoint and the failure of the frontal assault

The fighting followed from the ground. Thermopylae was a defile perhaps a mile long and, at its narrowest, only about thirty feet wide, hemmed between the cliffs of Mount Kallidromon and the sea, with an old wall across it that the Greeks repaired. In such a space the King's numbers could not be brought to bear, and his advantages in cavalry and archery were nullified; the fight came down to close combat on a front only a few dozen men wide, where the heavier arms and longer spears of the Greek hoplites told. For the first two days the frontal assaults failed. Herodotus makes Xerxes send in first the Medes and Kissians, then his best infantry, and reports the lesson the King is made to draw:

"They made it evident to every man, and to the king himself not least of all, that human beings are many but men are few." (Herodotus 7.210, trans. Macaulay)^[3]

When the Medes were roughly handled, the King committed the corps the Greeks called 'the Immortals', under the same Hydarnes who would lead the flanking march. They fared no better, and Herodotus fixes the reason in the terrain and the weaponry, not in any failure of courage:

"When however these also engaged in combat with the Hellenes, they gained no more success than the Median troops but the same as they, seeing that they were fighting in a place with a narrow passage, using shorter spears than the Hellenes, and not being able to take advantage of their superior numbers." (Herodotus 7.211, trans. Macaulay)^[3]

The standard modern study of the Achaemenid army draws the sober conclusion the narrative itself supports: at Thermopylae the Immortals were 'deployed only when Medes and Elamites fail', and were 'equally unsuccessful at uphill attack on a restricted front.'^[4] There is nothing here that need surprise. A chokepoint neutralises numbers by design; the finest army in the world, fed piecemeal into a thirty-foot gap against braced spearmen, will be checked, as many armies since have been checked, until the position is turned.

The Immortals: a Greek name, and a critical caution

The corps the Greeks called 'the Immortals' has become, through Thermopylae, the emblematic Persian unit, and it repays a source-critical eye. Herodotus alone gives them the name and its explanation:

"Of these ten thousand chosen Persians the general was Hydarnes the son of Hydarnes; and these Persians were called 'Immortals', because, if any one of them made the number incomplete, being overcome either by death or disease, another man was chosen to his place, and they were never either more or fewer than ten thousand." (Herodotus 7.83, trans. Macaulay)^[3]

They were, on this account, the standing royal infantry of ten thousand, a professional guard kept always at full strength while ordinary levies dwindled, and the elite of the King's foot; the reliefs of the Spear-Bearers at Persepolis are usually taken to depict such guardsmen. But the name and the round number are Herodotus', not the empire's. No Persian source calls this or any unit 'the immortals'; the corps vanishes from the Greek battle narratives after Thermopylae and Plataea, so that 'discussion of their tactical function' is left 'problematic'; and the whole of Herodotus' famous army catalogue, with its forty-seven nations, is 'not to be taken at face value', its exotic contingents best read, with Briant, as 'merely a parade contingent for ideological display' rather than a fighting order of battle.^[4] The 'Immortals' are a real Persian institution seen through a Greek lens and fixed under a Greek name; the modern reader should keep the institution and hold the label lightly.

Their role at Thermopylae has been further cut down to size. Herodotus implies the whole ten thousand made the night march over the mountain, but the topography forbids it. Scholars have long doubted that so large a force could have covered the rough track in the time; a column of perhaps two thousand would have sufficed to make the crossing before dawn, rout the thousand Phocians guarding the path, and turn the pass.^[1] The decisive stroke was delivered not by a horde but by a picked detachment doing exactly what elite troops are for.

The turning of the pass

What the frontal assaults could not achieve, local knowledge did. A man of Malis named Ephialtes, hoping for a reward, told the King of a mountain path, the Anopaea, that ran around the defile and came down behind it. Xerxes sent Hydarnes and his column over it by night; the Phocian guards on the track, surprised at dawn and supposing themselves the main target, drew off to a hilltop, and the Persians passed on and descended into the Greeks' rear. Learning that he was being encircled, Leonidas sent away the bulk of the allied army and kept his own Spartans, the Thespians who refused to leave, and the Thebans, to hold the pass to the end and cover the retreat. The last stand, when it came, was murderous and, on Herodotus' telling, magnificent: the defenders fought in the widening ground beyond the wall until their spears broke and they used swords, then hands and teeth, and Leonidas fell, and over his body a great struggle raged. Herodotus names, on the Persian side, a cost felt in the King's own house:

"In this fighting fell Leonidas, having proved himself a very good man, and others also of the Spartans with him... Moreover of the Persians there fell here, besides many others of note, especially two sons of Dareios, Abrocomes and Hyperanthes." (Herodotus 7.224, trans. Macaulay)^[3]

The deaths of two of the King's half-brothers, sons of Darius, register the reality that the victory was bought with blood, including noble Persian blood, and would have required funerary honours at court. Even so, the Greek reader should notice what the narrative quietly concedes: the pass was forced, the enemy's forward army was destroyed, and the King's advance resumed on schedule.

The reference article notes, with a cool eye the Greek sources lack, that the disaster had a Greek author too: the pass fell because the Persians could turn it 'thanks to the treachery of Ephialtes', and 'the strategists responsible for this location had inexcusably overlooked this alternative detour on their reconnaissance.'^[2] The heroic stand was made possible, and then made necessary, by a Greek failure of scouting.

The whip, the count, and the manufacture of a Persian rout

Two details of Herodotus' account do the heaviest ideological work, and both dissolve under scrutiny. The first is the whip. On the final day, Herodotus has the Persian commanders driving their own men forward into the Greeks 'with whips', so that many were trampled or forced into the sea; the image of soldiers flogged into battle is one of his recurring devices for marking the difference between free Greeks and the King's slaves. Amelie Kuhrt, editing the passage, states the plain objection: 'There is no other evidence for Persian soldiers fighting under physical duress.'^[5] The same motif attaches to the crossing of the Hellespont, said to have been done 'under whips', where Kuhrt observes that a comparison with Persian ceremonial suggests the whips were far more likely used 'to regulate the mass of soldiers moving across a fairly narrow bridge' than to flog them onward.^[6] The whip at Thermopylae is a literary sign, not a report of Persian practice.

The second is the body count, and here Herodotus' arithmetic collapses on inspection. He puts the King's dead at Thermopylae at twenty thousand against the Greeks' handful, and then, in a set-piece of contempt, has Xerxes hide almost all his own corpses and invite the fleet ashore to view the enemy dead and 'see how he battles against the senseless men who hoped to overthrow the might of the king', a spectacle the assembled sailors are made to find laughable.^[7] The numbers are impossible. Hyland notes that the twenty thousand is a fantasy figure, that far fewer Persians can even have engaged in the narrow space, and that the true Persian toll may well have stayed 'well under 1,000', the number of bodies later observers claimed to have counted before assuming the rest were hidden.^[1] Herodotus' own figure of four thousand Greek dead, meanwhile, seems to be an error drawn from the total Peloponnesian roster.^[1] Once the fiction of catastrophic Persian losses is set aside, the manufactured 'spectacle of failure' collapses with it, and a very different Persian event comes into view.

The King as spectator, and the display of Leonidas

That different event is a victory staged as a victory. Herodotus reports that Xerxes watched the fighting from a throne and, in his fear for his army, 'leaped up three times from his seat'; the watching King, made the impotent witness of his own men's failure, is one of the narrative's set images. Read from the Persian side, the observation is not a Greek invention to be explained away but a genuine element of Achaemenid kingship. Royal oversight of battle transposed to the field the iconography of the royal audience, the King enthroned to receive his subjects; his visible presence was itself an act of rule, binding the outcome to his person and his god and meant to make his men fight the harder.^[1] A hillside station overlooking the pass, from which the King could watch and be seen to watch, fits the practice exactly, whatever Herodotus does with the trembling.

The treatment of Leonidas' corpse tells the same story. Herodotus presents Xerxes' order to behead the dead king and fix his head on a pole as an ugly fit of rage, an aberration from the Persians' usual respect for brave enemies. The Achaemenid and Assyrian record makes it something else: the standard, meaningful punishment of a rebel. Darius at Behistun records the impaling and the decapitation-and-display of the 'liar kings' who rose against him, and Artaphernes had inflicted the like on the rebel Histiaeus at the end of the Ionian Revolt.^[1] From within the King's ideology, mounting Leonidas' head displayed the suppression of the Lie and the punishment of a wicked opponent, one whom Xerxes could plausibly cast, given the Spartan leadership's role in the deposition of the legitimate king Demaratus (then in Xerxes' own camp) and its murder of Darius' envoys, as a follower of falsehood justly destroyed.^[1] The viewing of the piled Greek dead served the same end that Assyrian kings and Achaemenid seal-images served, the visualisation of the total defeat awaiting those who defied the King; the seal impression from Xerxes' own treasury, showing a royal hero spearing a kneeling hoplite and dragging captives by a rope, is the iconographic twin of the scene at the pass.^[1] Even the branding of the surviving Theban captives with the royal mark, which Herodotus reports as another cruelty, was in fact the ordinary Achaemenid marking of dependents and property, well attested in Babylonia and Egypt, and here a permanent record of ownership and of royal triumph.^[5]

The strategic weight: a chokepoint forced, and little more

For all its later fame, Thermopylae changed the campaign's course hardly at all, and that is the measure of the distance between the two readings. Strategically it was a delaying action that failed to delay for long: the pass was carried in a matter of days, the parallel naval action at Artemisium was broken off once the land position fell, and the King's timetable held. Hyland stresses that the losses were militarily trivial, that 'the vast majority of Xerxes's army waited out the fighting in camp' and marched on the moment the road cleared, and that the far greater exertion of these weeks was not the battle but the logistics of feeding the host across a hostile and stripped countryside as it moved on Athens, a task the imperial supply system met without a crisis.^[8] The Persian achievement in central Greece was as much diplomatic as martial, the successful exploitation of Greek disunity: Thessaly, Boeotia (Thebes above all) and much of the centre went over to the King, and the coalition that had sent a few thousand men to the pass could not hold the line behind it. Athens was evacuated and taken, its temples burned.

What Thermopylae did not do is as important as what it did. It did not decide the war, which ran on through the sea-fight at Salamis that autumn and was settled only the next summer, on land at Plataea and at Mycale. It did not open some new and irresistible phase of conquest; it forced one gate on the road to a city the King had come to punish. And it cost the empire nothing it could not spare. Set in the frame of a realm that reached from the Indus to Thrace, the forcing of Thermopylae was a successful tactical operation on the far western edge of the world, a chokepoint carried at moderate cost after a check of two days, the kind of thing the King's armies had done to walled cities and mountain passes for a generation. That such an episode became, for the losers, the hinge of their history is a fact about Greek memory, not about Persian strategy.

Why the memory dwarfs the event

The enduring power of Thermopylae belongs to the Greek imagination and to the uses it made of the stand. The compression to '300 Spartans', the epitaphs, the cult of Leonidas, the reading of the battle as liberty's sacrifice against Asiatic despotism, all of this is later work, built on Herodotus and swelling for two and a half millennia, down to modern nationalist and cinematic appropriations of 'the 300'.^[1] Its ideological engine was already running in the narrative itself, in the exiled Spartan king Demaratus' famous answer when Xerxes asks whether so few will really fight so many:

"For though free, yet they are not free in all things, for over them is set Law as a master, whom they fear much more even than thy people fear thee." (Herodotus 7.104, trans. Macaulay)^[3]

And in the laconic epitaph Herodotus preserves for the Spartan dead:

"Stranger, report this word, we pray, to the Spartans, that lying here in this spot we remain, faithfully keeping their laws." (Herodotus 7.228, trans. Macaulay)^[3]

These are magnificent, and they are propaganda in the exact sense: the self-image of free men, disciplined by their own law, set against the King's subjects, imagined as driven by fear of a man and the lash. It is worth noticing that the sharpest statement of that contrast comes from the King's own mouth in Herodotus, who has Xerxes reason that free men, unlike his own troops, cannot be made 'compelled by the lash to fight with greater numbers, being themselves fewer', so that the whole battle is arranged, in advance, as a demonstration of the Greek thesis.^[3] The genius of Dieneces' reply to the news that the Persian arrows will blot out the sun, that then 'the battle against them would be in the shade and not in the sun', is of a piece: it is the free man's contempt for the horde, and it is unforgettable, and it is a Greek line.^[3] The historian's craft has made the reader remember a Persian defeat where there was a Persian victory. That inversion is the real subject of Thermopylae for a Persian-side history: not the denial of the Spartans' valour, which was real, nor of their sacrifice, which was total, but the recognition that the event's colossal place in memory is a monument the victors' descendants built, and that the empire, which forced the pass and marched on to burn Athens, would have recorded, had it recorded anything, only a gate opened and a wicked enemy justly destroyed.