SPEAKER_00 0:00

The Battle of Thermopylae, 480 BC. Thermopylae is a place before it is a battle. The name means hot gates, from sulfur springs that still rise along the inner edge of the Malian Gulf in central Greece. In the early 5th century BC, the pass was a coastal shelf, between the steep face of Mount Calidromus and the sea. Later centuries of silt from the Spircaeus and neighboring torrents have pushed the shoreline outward, so that a modern visitor stands on ground that was water in 480. The old track ran close under the mountain. At its narrowest useful point, the fighting corridor was only a few meters across, pinched further by a stone wall the Phocians had once built against Thessalian raids, and that the Greeks of 480 repaired. A battle here was not an open field collision, it was a problem in width. The year was the second great Persian invasion of the Greek mainland. Darius I had sent an expedition that ended at Marathon in 490. His son Xerxes I came ten years later with a combined land and sea force intended to do what the first attempt had not: bring the city-states of southern Greece into the tribute system of the Achaemenid Empire. The land army marched from Asia through Thrace and Macedon, then down the coastal route into Thessaly. The fleet moved in parallel along the shore. Greek cities that lay on that road submitted or negotiated. Others formed a league whose military leadership, by old prestige and by the weight of its infantry, fell to Sparta. Sparta in this period was a dual monarchy. The king who marched north was Leonidas of the Agiad House. He took with him the regular royal guard of 300 Spartians, men of full citizen status. They were not the whole Spartan army. The main levy was held back, in part because the campaign collided with the Carnea, a Spartan festival during which the state preferred not to send its full force abroad, and in part because the larger war was expected to be decided later, farther south, if the north could only be delayed. The 300 were therefore a holding detachment with a king at its head, not a national levy. They did not stand alone. Herodotus, who wrote two generations later, and remains the principal narrative source, lists a coalition that modern totals usually put near 7,000 hoplites at the opening of the fighting. Peloponnesians besides the Spartans, Phocians and Locrians from the neighborhood of the pass, Thebans and Thespians from Boeotia, and smaller contingents whose exact figures shift from manuscript to manuscript. Behind or beside the citizen infantry were helots from Laconia and other attendants who do not enter the famous count. The Greek force at Thermopylae was a federal sample, not a single banner. Opposite them Herodotus sets a Persian host so large that it drinks rivers dry. His grand total, built from unit lists and from a method of counting that he describes with care, runs into the millions. Almost no modern historian accepts that figure. The Empire could raise a very large army by the standards of the age. It could not feed two million men and their animals on a single Balkan road in late summer. Estimates now commonly fall in a wide band from perhaps a hundred thousand combatants, to two or three times that, with a long tail of servants, drivers, and camp followers. For the purpose of understanding the pass, the exact census matters less than the shape of the problem. A wide army entering a slot a few wagons wide ceases to be a wide army, it becomes a column. Xerxes did not attack on the first afternoon of arrival. The sources give him days of delay after the Persians came in sight of the position. The delay has been explained as reconnaissance, as an expectation that the sight of the host would produce submission, and as the simple need to close up a marching force and bring the fleet into some relation with the army. Off the north of Euboia, at Artemisium, a Greek fleet was trying to hold the sea flank so that the Persians could not land behind the pass. Thermopylae and Artemisium were one operation in two elements. What happened on the water is another essay. On land, the first assaults were frontal. A hoplite of this period fought as a heavy infantryman, bronze or composite helmet, cuirass of bronze or layered linen, greaves, a large round shield, a thrusting spear and a short sword. The formation was a shield wall several ranks deep. In a narrow way, the rear ranks were not wasted. They replaced the tired and the wounded, and kept the pressure of weight behind the front. Persian infantry was more mixed. The empire drew on many peoples, some carried large bows and light shields, some, including the corps later called immortals, in Greek writing, were better armored and used spear and bow together. Cavalry, the Persian arm that decided so many open battles, had little room to form. The first days at Thermopylae were therefore an infantry argument in a corridor. The fighting is described as lasting two full days of frontal pressure before the third day changed its geometry. Greek accounts emphasize rotation of contingents at the wall and in the throat of the pass, with the greater reach of the hoplite spear in close order, and the inability of archery to break a shielded line close to contact when it chose. Persian accounts in the strict sense do not survive as independent narratives from 480. What we have is Greek storytelling over a defeated invasion, written when the memory was still political. That does not make the outline false. It does mean that speeches, omens, and exact body counts should be read as literature sitting on top of a military fact. The pass held while it remained only a front. Above the pass, a mountain track, the Anapea, ran along the shoulder of Calidromus and came down behind the Greek position. Local people knew it. The Phocians were posted on that path to close it. On the second night of the battle, or in the dark that ended the second day, a force of Persian infantry, Herodotus says the immortals under Hydarnes, was guided onto the track by a man of malice named Ephialtes. The name later became a common noun for a nightmare and a byword for treachery. The act itself was the ordinary sale of a local secret to an invader who could pay. Mountain paths decide more campaigns than oracles do. At dawn the Phocians on the heights heard the advance, they armed and formed, expecting, the story goes, to be the object of the attack. The Persians, whose object was the rear of the pass, discharged a flight of arrows and went past them along the path. Whether the Phocians could have held a true blocking position is a question the sources do not settle. News of the turning movement reached Leonidas in the pass below. What followed is the part of the story most thickened by later telling. The outline in Herodotus is that the king dismissed the greater part of the Allied army and remained with the Spartans, the Thespians, who chose to stay, and a body of Thebans, whose willingness is disputed in the same pages that record their presence. Motives are supplied, an oracle that Sparta must lose a king or lose the city, the law that Aspartiate did not retreat, the need for a rear guard, while the rest of the League army escaped to fight again. All of these may have been spoken at the time, or some of them may have been fitted afterwards to a decision that was simply tactical. A pass that is about to be taken in front and rear is no longer a pass, it is a pocket. Keeping 7,000 men in a pocket to be counted at leisure would have served Xerxes better than it served Greece. Sending most of them south preserved a field army. The last morning is described as a forward movement. The remaining Greeks left the wall and the tightest constriction and advanced into a slightly wider stretch to meet the frontal attack, then fell back to a rise when the turning column began to arrive from behind. Leonidas was killed in that fighting. The struggle over his body is written in the same register as other Greek battle pieces. Repeated charges, recovery of the fallen commander, exhaustion of spears, a last compact on a hillock. The Thebans, in Herodotus, separated and surrendered. The Spartans and Thespians were killed. A stone lion was later shown near the spot. Epitaphs were composed, including the two-line address to the passerby, that tells him to report to the Lacedaemonians that the dead kept their word. The couplet is short enough to be remembered, and general enough to outlive the logistics. Xerxes had to pass. The road into Locris, Phocus, and Boeotia lay open. The fleet at Artemisium, once the land position was gone, could not stay where it was. Athens was evacuated to Salamis and the islands. The Persians took the empty city and burned the Acropolis. None of that is Thermopylae, but Thermopylae is why those things happened when they did, rather than later. The holding action did not stop the invasion. It used up a few days, cost the Persians men and time, and gave the southern cities a clearer calendar. The war's naval turning came at Salamis in the same season. The land turning came the next year at Plataea. Thermopylae belongs to the opening movement, not to the decision. The pass as a landform deserves a slower look than the last morning. The Malian Gulf in 480 was a deeper bite of sea than the map shows now. The coastal strip was interrupted by ridges of the mountain coming close to the water and by marshes fed by the springs. Travelers of the classical period speak of three gates, successive narrowings rather than a single gate. The middle gate, where the folk and wall stood, is the usual site assigned to the main fighting. Hot springs made the air smell of sulfur and left stains on the rock. In winter the road was a local necessity. In summer it was the way an army from the north entered the waste of Greece without climbing the main mass of Oeda. To choose Thermopylae was to choose the cheapest defensible ground on that road, not a mystic landscape. The wall itself was old when Leonidas repaired it. Phocus and Thessaly had a long habit of raiding one another. A barrier in the past was a border work before it was a monument. In 480 it served as a backstop and a night position. Men slept behind it and fought in front of it. After the battle it ceased to matter. Later ages rebuilt walls in the same corridor for other wars, including a famous defense against the Gauls in the 3rd century BC and works in Roman and later times. That is a fact about contours, numbers again, because sleep is compatible with arithmetic. 300 Spartiates is a round and official figure. It does not include the helots who went with them, and it may or may not include a small number of men who were sent away before the last day as messengers. 700 Thespians is the usual figure for the Boeotian town that stayed in full. Thespii was not a large place. If the figure is near the truth, the town put a very high share of its citizen infantry into a position it did not expect them to leave. Four hundred Thebans are named, Thebes later Medized, that is, came to terms with Persia, and the tradition about those four hundred is accordingly sour. The other allies who left on the third morning are not therefore to be written as cowards. They were the army that still had a war to fight. On the Persian side, the named commander in the past is Xerxes himself, sitting on a throne or a seat with a view, in the Greek telling, with Hydarnes as the officer of the flanking column. Unit, names, and Herodotus follow the ethnic list of the Imperial Levy. Medes and Sissians in the first attacks, then the immortals. Whether those labels match the administrative names used in Persian would be a separate study. What the labels do on the page is give the reader a sequence. Lighter or less heavily armed troops first, household troops later, a night march last. That sequence is believable, even if every ethnic caption is not. Source criticism is part of the subject. Herodotus of Halicarnassus collected stories from Greeks who had lived through the war or whose fathers had. He names informants at times and admits uncertainty at others. He likes oracles, dreams, and the moral pattern of greatness drawing down reversal. He also likes measurements, lists, and the look of ground. Later writers, Diodorus, Justin, and the Anthologist of Sayings, add speeches and sharpen the contrast between luxury and discipline. Inscriptions in the later tourist landscape at the pass adds still more. Archaeology in the corridor has to contend with the chained shoreline and with the many later uses of the same road. It can confirm that this was a traveled pinch point. It cannot film the third morning. A reader who wants the battle only as an emblem will find it everywhere. Civic sacrifice, the few against the many, the law that does not yield. Those uses began early. They are not false to the feelings of some who fought. They are incomplete as a description of what happened. What happened is that a league posted a blocking force on the best narrow ground on the invasion road, that the block worked for two days against a frontal entry, that a local path existed and was used, and that the remnant of the blocking force was destroyed on the third day, while the rest of the army left. The emblem grows in the space where the column of 7,000 becomes a story about 300. The 300 were real, so were the others. Climate and season belong in the file. The fighting is usually placed in late summer, August, or early September of 480, when the Greek harvest was in, and a festival calendar was crowded. Heat in the past would have been considerable. Water from the springs is warm and mineral. Drinking water for thousands of men and for horses had to come from streams that a large army could foul in a day. Dust, smell, and the noise of a confined fight are not decorative details. They are the medium in which two days of work were done. Night on the mountain path would have been cold enough by comparison, and dark in the way of an unlit ridge with a drop on one hand. After the dead were buried or left, the war moved on. Greek tradition records Xerxes walking among the bodies and looking for Leonidas. It records mutilation of the king's corpse, a detail meant to mark the invader as standing outside Hellenic custom. It records the later recovery of bones and the institution of memorial games. At Sparta, the names of the 300 were kept. At Thespiae, the loss is thought to have marked the town for a generation. None of this changes the military result of the week. The result was an open road, and a delay measured in days. If the pass is walked in the mind once more, from north to south, the battle becomes a short list of positions. The outer approaches where Persian scouts first saw men combing their hair in a Spartan manner, the repaired wall, the throat where a shield line could not be outflanked by horse, the hidden track along Caledromus, the hillock of the last compact, the road south that the dismissed allies took. Between those points there is room for a long telling that never raises its voice, how a hoplight puts on grieves sitting down because they are easier that way, how a bowshot in a narrow lane strikes shields more often than men, how a guide on a night path measures progress by the feel of the slope rather than by landmarks. That is the scale at which Thermopylee is useful for sleep. It is a small piece of ground with a known use, a week in a long war, and a set of lists that can be read slowly until the lists themselves become a kind of breathing.