Which battle can we consider to have been the most important turning point in the Greco-Persian wars

The Greco-Persian wars of 498 BC to 448 BC were a clash between two completely contrasting cultures. The Achaemenid Persian Empire stretched from the subdued Ionian Greek city states on the shore of Asia Minor to the Indus River bordering India and was ruled zealously by God-Kings in Persepolis, the capital of the Empire. On the contrary, the land we now know as Greece was a collection of small, aggressive city states that managed to put aside their many differences, and, against the overwhelming odds stacked against them, turned back the might of the Persian army from within their own territories.

The Hellenic city – states that fought against the Persian Empire certainly defied the odds, but when and how did the tide change? Most Historians agree on a number of battles between 490 and 479 BC, but opinion is still somewhat divided as to which battle provided the exact turning point from the initial dominance of Persia to the Hellenic city states gaining the upper hand. In this essay, I will investigate the four key battles of Marathon, Thermopylae, Salamis and Plataea and will attempt to explain their significance within the war1.

The battle of Marathon took place in the late summer of 490 BC, during the first Invasion of Greece, organised by King Darius I. When Persian troops landed at Marathon en masse, the Athenians debated whether to stay behind their walls or come out and show their faces to the enemy. After much debate in Athens, the citizens voted according to Miltiades (Burn, 1962: 241) and decided to front up to the enemy where they had landed at the bay of Marathon.

What is also interesting is that (aside from the 600 Plataeans) Athens stood alone in this battle. Sparta was willing to send troops to support the Athenians in battle, but Pheidippides returned from Sparta saying that the Spartans could not march until the full moon had risen – signalling the end of the religious festival of Carneia (Burn, 1962: 240. This would taken at least a week (Holland, 2006: 187-90), and so the Athenians would have had to have stayed put at Marathon. When they arrived, the Athenians secured the two exits towards Athens from the battlefield before the Persians could, thus protecting Athens (Burn, 1962: 243).

The Athenians were joined by a small contingent of allied Hoplites from the polis of Plataea, which took the Greek troop number to 10 or 11,000 (Lazenby, 1993: 54). Depending on interpretation, the Athenian camp was protected on its flanks either by woods or by tree trunks cut down, so as to protect them from Persian cavalry (Burn, 1962: 243; Lazenby, 1993: 56). Whatever the correct interpretation, it should be noted that the Athenians knowledge of the battlefield and the planning of their position acted as a force multiplier.

For the Persians, Marathon was chosen specifically for its suitability for cavalry; something other potential landing sites didn’t offer (Burn, 1962: 245). The location of Marathon meant that if Datis had achieved victory there, he could have quickly marched onto Athens. It was also an ideal location to send ships around the Cape of Sounion (Burn, 1962: 245) and attack Athens while it was undefended.

For five days, both Armies were kept confined to their camps.on either side of the battlefield. The Greek camp were happy to wait for the Spartan reinforcements to start marching, while Datis was waiting for a signal from Hippias’ contacts within Athens to inform them that Athens was indefensible (Burn, 1962: 246). After 5 days, the stale-mate ended.

Datis attempted to withdraw his cavalry during the night in order to encircle and attack the undefended Athens; however, this attempt at deception didn’t go un-noticed to the Athenians (Burn, 1962: 247). This theory agrees with Herodotus’ suggestion that the Persians tried to sail around Sounion and attack Athens after the battle had been lost (VI, 115). Greek hoplites were very vulnerable to attacks on their flanks and their rear, because of the nature of their static line. Since the cavalry were easily able to do this, and they had been removed from the field of battle, it was a tactical relief to the Athenians.

Another theory suggests that the Persians simply moved their lines towards the Athenians, implying that they attacked first. However, the Athenians thought the Persians were attacking, not merely advancing and decided to attack them head on, thus taking the tactical offensive – which would still fit with the general agreement that the Athenians attacked first (Burn, 1962: 248). The view of the Athenians making the first strategical and tactical advance is supported by Herodotus account of the Council of War (VI, 110), which shows Miltiades winning the vote of Callimachus and adopting an aggressive stance against the Persians (Burn, 1962: 246). If the Persian army hadn’t actually advanced towards the Athenian line in the first place, then the Athenian advance would have definitely taken the Persian army by surprise (Burn, 1962: 249). Therefore the Persian troops would have been fairly disorganised and mentally unready for battle.

The Persian front line was 10,000 strong anyway; therefore the Athenian line had to spread out just to meet it (Lloyd, 1974: 191)! As a consequence, the hoplite ranks within the centre (commanded by the future statesmen Themistocles and Aristides) of the line were only four men deep, yet the Plataeans (led by Arimnestus) on the left and Callimachus of Aphidna, the polemarch, on the right had the normal compliment of an eight man deep phalanx. Burn suggests that the flanks were made stronger to prevent being enveloped on the flank (1962: 248-9).

Another benefit of this tactic, which may or may not have occurred to Miltiades, was that if the flanks broke through their Persian counterparts, then this would prevent the enemy from fleeing outwards and would allow them to surround the enemy. As it happens, the Persian flanks consisted of lighter armed archers (who were easily crushed), yet the Persian centre contained better troops – the native Persians and the Sakai, who were the most trusted contingents in Persian battle (Burn, 1962: 250).

As the Greeks chose to charge the Persians (Burn, 1962: 248), their archers were only able to let off one or two volleys of arrows before they were impaled upon their rushing Athenians spears. The slaughter upon impact would have been terrible. As it happens though, the Persians on the centre managed to push the Athenians back (due to the weak Athenian centre). The names of slaves within the battle obituary may indicate that the Persians broke through to the Athenian camp and killed slaves skirmishing behind the hoplites or slaves within the camp itself (Burn, 1962: 250).

However, the wings of the Greek army succeeded in routing their opponents and after doing so, paused before turning inward to surround the Persians troops in the centre. Once the Persians realised they were surrounded, their centre routed as well and fled towards their ships. The subsequent pursuit by the Greeks resulted in a few casualties as they attempted to capture Persian ships, including the polemarch Callimachus (Burn, 1962: 251). Many fleeing Persians were killed by running into the Marshes to the north of the battlefield, which Pausanias informs us of (I, 15, 4).

Marathon was an impressive victory – they had ‘contained the beach-head’ (Lazenby, 1993: 75), yet they knew the Persians would come again (Lloyd, 1974: 206). Miltiades had very little to do with the victory other than spreading the line and wanting to be aggressive towards the enemy. The major reasons for victory at Marathon were twofold. The Athenians (and Plataeans) were fighting to save their freedom from Persian dominance (Lazenby, 1993: 79) but it had nothing to do with superior Greek discipline. The Athenians were quite a head-strong bunch anyway, and as long as a Persian army was properly commanded, they were disciplined fighters in their own right (Lazenby, 1993: 79).

The Greeks were also better equipped – leather cuirasses and hoplite shields offered invaluable protection to Persian arrows (contemporary depictions of a hoplite at that time show them wearing leather cuirasses, not bronze (Connolley, 1981: 54, Lloyd, 1974: 196)), whilst the longer, stronger spear of the hoplite out-manoeuvred the shorter Persian spear and easily decimated the mostly un-armoured lighter Persian infantry. It still took great courage for two poleis to stand up to the Persian Empire and to an army twice the size of them, considering what they had done to others who had resisted them (Lazenby, 1993: 79-80).

Surrender would have entailed the dissolving of the direct democracy within Athens, the instalment of a Tyranny within Athens under Hippias, a regular tribute and naval or military service to Persia (Lazenby, 1993: 80). Given that the victory at Marathon was not due to superior tactics, Generalship, discipline or (Spartan) reinforcements, but due to actions of normal citizens, the victory is remarkable. If Marathon saved Greece, then the Athenian hoplites preserved wester civilisation (Lloyd, 1974: 207)!

The battle of Thermopylae is one of the most renowned battles in history and it took place in the either August or September (more likely August according to Lazenby (1993: 119) and to Cartledge it was the 17th (2007: 142)) of 480 B.C. There is also heavy debate among the scholars of the numbers of the armies on both sides; but we can safely assume that there were anywhere between 200/250,000 Persian combatants (Holland, 2006: 237) and about 5/6,000 Greeks (Green, 1996: 140).

The Spartans suffered the same problem of sending troops during the religious festival of Carneia, as they did during Marathon, but because of the grave situation, 300 Spartiates (all of whom had children) were sent to Thermopylae along with King Leonidas and to pick up as much support along the way as possible, to supplement the lack of his own troops (Lazenby, 1993: 135).

The pass at Thermopylae was so thin, that only a single chariot could pass through it in single file (Herodotus, 7, 176), and there were a series of gates along the pass. Leonidas decided to fortify the ‘middle gate’ and fight from there (Lazenby, 1993: 130). The constrictive nature of the pass actually made the best use of their numbers – a small force could hold the pass for days against a larger army who would be squeezed and suffocated within a confined space. So even if Leonidas didn’t get the number of troops he would have liked, in reality, the total number of Greek troops should have been enough to hold the pass for a week, if there was any plan within Sparta to send more troops later (Lazenby, 1993: 136).

It is often said that after arriving at Thermopylae, Xerxes waited four days for the Greeks to withdraw in the face of his mighty army in some sort of petulant manner. This is untrue however as Xerxes was actually waiting for his infantry to arrive before attacking – his cavalry arrived four days before his infantry did and only on the fifth day did he attack. There is also the possibility that he was waiting for his fleet to win at Artemisium so he could flank the defenders (Burn, 1962: 410). Two and a half weeks later after his army left Therma for Thermopylae, Xerxes ordered his Medes and Cissians to ‘clear the pass’ (Burn, 1962: 410).

The fighting situation evolved itself much like it did at Marathon ten years earlier, with the heavier armoured hoplites in phalanx formation destroying the lighter armed and armoured Persian troops. The Spartiates were a lot tougher and actual soldiers compared to the citizen soldiers of Athens. When the Medes and Cissians retreated, Xerxes ordered his Immortals under Hydarnes to advance yet the suffered just the same as those before them. As well as the phalanx, the Spartans ‘feigned to retreat’ and turned on their pursuers, which killed many Persians and only a few Spartiates (Burn, 1962: 410). The first day of battle was a bloody one for Xerxes.

It was the same story for much of day two. The Greek forces regularly took turns on the front line so as to relieve each other from fighting (Burn, 1962: 410). Burn says that Xerxes’ officers knew of a path through the mountains, but they needed a local to guide them through it (1962: 410). Men being led up a path they had no intimate knowledge of could have proved very dangerous, especially if it was defended. Leonidas was informed of the path before the battle by local Trachinians (Burn, 1962: 410) and the Phocians offered to guard it.

It was Ephialtes who offered to guide the Persians up the long and dangerous path at night ((Burn, 1962: 413/4); they left at dusk at reached the top of the Anopea path on the mountain at dawn, where they were discovered by the Phocians (Burn, 1962: 415; Herodotus, 6, 215). As it was, the Phocians were easily brushed off by the advancing Immortals, led by Ephialtes and Hydarnes, thinking they were the target of the attack only to find the Persians marching past (upon the Persians discovering that the Phocians were not Spartans from Ephialtes) them towards the Greek encampment (Burn, 1962: 416).

Diodorus Siculus tells us of how Leonidas called a Council of War shortly after learning that his position had been compromised (6, 8, 5). We learn that either many of the Greek contingents left by their own accord, or they were sent away by Leonidas to fight another day – one would like to think it was the latter. What is certain is that Leonidas and his 300 Spartiates, along with 700 Thespians (who refused to leave), 400 Thebans and the Helots (Burn, 1962: 417; Holland, 2006: 291-3) stayed behind due to a combination of factors. Firstly, they had to protect the other Greeks who had left (Cartledge, 2007: 147; Burn, 1962: 418; Lazenby, 1993: 144/5). The retreating Greek contingents would have been cut down easily by pursuing cavalry had they been able to ride through the pass unopposed.

Secondly, Leonidas had to remain with the rear-guard to ensure it held up the Persians for as long as possible (Burn, 162: 418). One without him would have dissolved away and lead to the retreating Greeks being harassed. The longer the Persians were held up, the better. Lastly, the Spartans were a superstitious people, and there can be no doubt that Leonidas would have had the words of the Delphic Oracle running through his mind, telling him that a King would die to save his city (Burn, 1962: 418). It was neither a Spartan thing to do to retreat from battle, so it was a case of ‘alea iacta est.’

So on the third day, the Greek rear-guard ventured into the wider section of the pass in front of the Phocian wall and attempted to slay as many Persians as possible (Herodotus, 7, 223), fighting with whatever they had available (Spears, swords, fist and teeth!). It is here that Leonidas is killed and two of Xerxes brothers (Abrocomes and Hyperanthes) are killed trying to capture it (Burn, 1962: 419; Cartledge, 2007, 150; Herodotus, 7, 224). When the Immortals approached from the Anopea path, the Greeks retreated to the Kolonos hill, where they were all promptly surrounded and were slain by arrow fire, except for the Thebans who surrendered (Herodotus, 7, 225).

The Battle of Thermopylae was a definite defeat for the Greek Allies, and one that could have actually been better for them. Perhaps, as Burn, and Cartledge all suggest, a greater detachment of Hoplites guarding the Anopea path could well have stopped the Immortals (Burn, 1962: 413/7; Cartledge, 2007: 147) from enveloping the Greeks. The Phocians posted there were enough to stop ambitious sorties, on the basis that the Persians had not been informed of a proper path.

The Greeks had only managed to hold the strongest point above the Isthmus for three days, thanks in part to the somewhat reluctant nature of the Spartan Ephors to send reinforcements north (Burn, 1962: 421, Lazenby, 1993: 136); perhaps as a result of division at home (Grant, 1961: 14). So a underwhelming compromise was made. Even the smallest of troop reinforcements would have helped. The Greeks at Thermopylae only had to hold on for another ten days until Carneia ended in Sparta until proper reinforcements could be sent. Given the topography of the battlefield, this was not impossible.

Lest we look on Thermopylae with a unfavourable view, Leonidas’ actions set a precedent for the war. Herodotus states that some 20,000 Persians were killed by the Greeks at the hot springs – a tenth of the 200,000+ strong army (Herodotus, 8, 24; Cartledge, 2007: 150). Secondly, the sacrifice of Leonidas’ 300 and 700 Thespians allowed the retreating Greek contingents to withdraw safely. But perhaps the most important factor is that every soldier who fought at Salamis and Plataea afterwards would remember the actions of those men at Thermopylae and use their memory to fight against the Persians with greater courage; “free men” who died to protect their way of life was a potent inspiration. Despite defeat, hope was salvaged.

The Battle of Salamis took place in September of 480 B.C. and was the probably the major naval battle during the 2nd Invasion of Greece. It probably happened about two or three weeks after the Persians captured Athens (Lazenby, 1993: 164-7). The Greeks armies had decided to defend the Isthmus, thus requiring the Greek navy to defend it from being flanked from the sea (Holland, 2005: 302/3). Since the Greek navy was considerably smaller than the Persian fleet, they had to find a way to counter the numerical superiority, like the Greeks had nearly done at Thermopylae. Themistocles recognised this and thus the Greek fleet remained at Salamis after the evacuation of Athens (Herodotus, 8, 63).

Word reached Xerxes of a rift in the alliance, though it is possible that this may have been the first of many ruses to trick Xerxes into a battle (Holland, 2005: 310-15; Burn, 1962: 449). This results in Xerxes ordering his fleet to block the southern exit off Salamis. Herodotus tells us that that evening, Themistocles sent his personal slave Sicinnus to Xerxes with false information that he was sympathetic to Xerxes’ cause and that the Greek fleet were planning to withdraw that evening, so all Xerxes would have to do is to block the evacuation for a decisive victory that he was seeking for; thus drawing him into the straits of Salamis (Herodotus, 8, 75; Holland, 2005: 310-15). Xerxes took the bait.

The following council of war that night between the Allies was a heated one, with Peloponnesian Greeks wanting to withdraw still to the Isthmus. However, their hand is forced when the recently recalled exile Aristides arrives with Persian turncoats in tow, describing how the Persian fleet had blocked their escape (Herodotus, 8, 81-3). So while Xerxes fleet was working restlessly at night, looking for ships that were not there (and to engage them – (Holland, 2005: 318)), the Greeks were resting and preparing for the next day (Burn, 1962: 449).

In the morning on the day of the battle, Herodotus tells us of Themistocles rousing speech to the Marines who would be fighting on the ships (Herodotus, 8, 83), declaring that ‘all is at stake’ (Burn, 1962: 455). It was probable that the Greeks deployed in two lines, since the straights were too narrow for a single line of ships, whilst the Persian navy held their position in the bay, with the Phoenicians at the head of the column and the Ionians at the tail (Lazenby, 1993: 187, 174-80). There is a possibility that the Egyptian ships were sent to circumnavigate the island of Salamis to prevent an escape should it have happened (Holland, 2005: 310-5).

The battle began in the morning, soon after the Greeks had sung their paean. The Marines boarded the triremes and the ships then began to manoeuvre further into the straits in an orderly fashion. It appears the Corinthian contingent of ships under Adeimantos began to sail away northwards as the Greek line formed, which, rather than an attempt to flee (as Herodotus believed (Herodotus, 8, 94)), was either a trick, leading to believe the Persians that the Greek alliance was malfunctioning (Strauss, 2004: 152) or was a reconnaissance trip to see if the Egyptian ships had encircled them (Burn, 1962: 458). Either way, the Corinthians soon returned.

As the Greeks prepared their battle line, the Persians were suffering already. The wind was breaking up the water and this made it difficult for the Persians to turn around and prepare a front to the Greeks, meaning that they were vulnerable to ramming (Burn, 1962: 461). The Persians had no idea about the local weather, thus explaining their predicament (Strauss, 2004: 154, 171). At that point, either a Athenian boat (on the left of the battle line) or a Aeginetan (on the right) rammed a Persian vessel (Herodotus, 8, 84). It was at this point that the 378 (number given by Herodotus (Herodotus, 8, 48)) Greek ships charged towards the 6/800 strong Persian fleet.

All of the Persian line was engaged at once. The Phoenicians at the head of the Persian column were surrounded by the Athenians (Burn, 1962: 461). Like at Thermopylae, superior numbers actually became a hindrance, and they lost all formation. Herodotus’ text descends into a series of small skirmishes during this retelling of the battle (Strauss, 2004: 179; Burn, 1962: 463).

It’s not hard to guess what happened to the Persian fleet though – the damaged front line of ships would have created chaos within the lines of the ships behind them. In turn, the ships behind them were forced back, but knowing they were being watched by Xerxes (Burn, 1962: 460; Strauss, 2004: 192), they would have been keen to press forward and attack, causing even more chaos (Burn, 1962: 463). The death of Ariabagnes, the Persian admiral, early in the battle contributed heavily to the disarray of the Persian fleet (Strauss, 2004: 174).

The Greek marines (hoplites) would have boarded Persians ships and would have easily nullified the lighter Persian marine infantry (Strauss, 2004: 176). As the battered Phoenician (and Cypriot) ships fled or ran aground, it spread panic and caused other contingents like the Cilicians and Pamphylian to flee as well (Burn, 1962: 464; Diodorus Siculus, 6, 19, 1). The Aigenetans spread further misery by preventing ships at the rear from helping the ones in front and when other Persian ships fled, they blocked their escape routes (Herodotus, 8, 91; Burn, 1962: 465).

So the Persian fleet was put to flight, and they hurried back off to Phaleron, where the army was. The Greeks had managed to fight as a cohesive unit, whilst the Persian contingents were busy blaming one another for the defeat, as seen in the case of the Phoenicians (Herodotus, 8, 90; Strauss, 2004: 193). The Persians suffered from the shock of the Greek charge, the loss of their admiral and from not knowing the geography of the area (Strauss, 2004: 206).

The loss of Xerxes fleet probably shortened the length of the war, as the Greek army could now not be flanked from the sea (Lazenby, 1993: 197) and the Peloponnese was safe. However, Mardonios still remained in Greece afterwards, and as long as he was present in Greece, the Greeks could still lose the war (Burn, 1962: 411). However, in seeking a decisive battle, Xerxes gambled and lost, reducing the chances of conquering Greece dramatically (Holland, 2005: 303).

The battle of Plataea was the last battle to take place on mainland Greece between Persian and Hellenic forces. After Mardonius had sacked Athens for a second time, the Athenians requested Spartan assistance to end the Persian occupation, threatening to accept to the Persian terms of surrender if not (Herodotus, 9, 6-9). Thankfully, when the Athenians decided to issue an ultimatum, they discovered that the Spartans had already sent a task force to engage the Persians (Herodotus, 9, 10). Mardonius decided to retreat to the Medising Thebes and prepared camp there (Herodotus, 9, 15).

Disregarding servants or Helots, the Spartans sent 5000 Spartiates and 5000 other Lacedaemonian hoplites while the Athenians committed 8000 hoplites themselves (Herodotus, 9, 28). There were also sizeable contingents sent by other city states to the battle, including 5000 from Corinth and 3000 from Sicyon (Herodotus, 9, 28). Over the course of the days leading up to the actual fighting, the Greek forces were continually reinforced by hoplite contingents from other poleis (Herodotus, 9, 35), to the extent that the total amount of Greek hoplites came to 38,700 (Herodotus, 9, 29; Hignett, 1963: 435-8).

The Persian army, by modern estimates, was judged to be anywhere between 70,000 to 120,000 men (Lazenby, 1993: 227/8; Connolly, 1981: 29).When Xerxes left Greece, Mardonius was give his choice of troops and the army at Plataea probably included native Persians, Medes, Sakai, Bactrians and Indians (Lazenby, 1993: 207). He also would have had Medising Greek contingents, including some from Thebes. There were also Phocians, whom he apparently tested to find out how Hoplites reacted to missile cavalry attacks and to intimidate them (Barron, 1998: 598-9 cited by Lazenby, 1993: 220).

As the allies formed up on the high ground at Plataea, Mardonius probably knew his troops would not do well in a frontal attack on the Greek position, and accordingly avoided battle (Holland, 2005: 343-9). However, he did use his missile cavalry well and had success harrying the phalanx until his cavalry commander Masistos was killed by (a rare occurrence in Greek armies) Athenian archers during a cavalry manoeuvre (Herodotus, 9, 22; Lazenby, 1993: 222). This was a major morale boost for the Greeks and buoyed by this, they moved to a new position occupying a low ridge in front of the Asopus ridge and the Persian position (Herodotus, 9, 25).

Despite occupying this position so close to the Persians, neither side moved for 8 days. However, Mardonius had the cavalry to harass the Greeks and their supply lines, cutting of their food supply and water supply (Herodotus, 9, 39-49; Lazenby, 1993: 233; Burn, 1962: 528). Eventually, after a couple of days, Pausanias and the other Greek commanders, including Aristides, decided to fall back towards Plataea, where they could guard the pass through Mt. Cithaeron and have access to clean water (Herodotus, 9, 51). This was to be planned at night, so they could not be attacked by missile cavalry.

Unsurprisingly, the withdrawal at night was a disaster, as one would expect from mostly part time soldiers, all from different states, when ordered to retreat over night in unfamiliar terrain (Lazenby, 1993: 235)! At daybreak, the Athenian and Spartan flanks had not started to withdraw, so the Spartan left a rearguard and withdrew uphill with the Tegeans (Lazenby, 1993: 237; Burn, 1962: 531), whilst the Athenians headed towards Plataea, even though Pausanias asked them to join up with his flank (Herodotus, 9, 55). As this would have happened in front of the Persians, Mardonius must have thought he could achieve victory. He had successfully avoided battle thus far and the Greek retreat meant that morale was low, which would cause further rifts within the alliance.

What went badly for Mardonius was when he began to pursue the Greeks; his whole army followed him without orders. The Spartan rear-guard, under grave pressure from Persian cavalry managed to join the Spartan main line just in time (Lazenby, 1993: 240). As Mardonius joined the cavalry with his Persian troops, they began to fire arrows at the Spartan lines, hiding behind their shields, while Pausanias was calling for Athenians support, only to find they had been engaged by the Thebans (Holland, 2005: 350-5).

As the Greek line started taking casualties, and Pausanias was trying to find good sacrificial omens, the Tegeans finally relented and charged at the Persians. Fortunately, the omens came good for Pausanias and the Spartans joined in (Herodotus, 9, 61-2; Lazenby, 1993: 241). Despite the Persians having numerical superiority, the hoplites were heavier infantry and they easily out-matched their enemy. The Persians tried to break the phalanx in small groups but were beaten back (Lazenby, 1993: 242). Mardonius was hit in the head by a rock and died, and as a consequence of this, the Persian army on the right flank fled (Lazenby, 1993: 243).

Artabazos, the second in command, in the centre decides to retreat upon seeing this (Lazenby, 1993: 243). The Athenians on the left flank succeed in routing the Thebans (Herodotus, 9, 67) and helped by the distraction of the Theban cavalry attacking the Megarians (Lazenby, 1993: 244), which left the Megarians with 600 dead (Herodotus, 9, 69). Thus, the Persian army was fully routed. Despite Herodotus’ estimates that 257,000 Persians died (Herodotus, 9, 70), modern estimates put the number at around 10,000 men. The Greeks on the other hand only lost 159 men (Herodotus, 9, 70), while Plutarch gives around 1,360 casualties (Plutarch, Aristides, 19). Ephorus and Diodorus Siculus give 10,000 Greek deaths (Diodorus Siculus, 11, 33, 1). Herodotus mentions of the Persian camp the routers fled to which was attacked and the defenders slaughtered (Herodotus IX, 70).

The Greek victory at Plataea came down to luck and to the superiority of the hoplite, which had been proven at Marathon and Thermopylae (Lazenby, 1993: 256); the Persian light infantry could not break up a line of spears, given that they did not engage in close quarters combat normally. Moreover, Plataea was a soldier’s battle; it was the stubbornness of the hoplite that won the day (Lazenby, 1993: 257), which according to Hignett, was won by the presence of the Spartan soldiers (1963: 343; Bengston, 1969: 62).

There was little a commander like Pausanias could do other than to choose the gound. He might not have expected to fight the battle he did. As long he kept his troops in order, he had little else to do (Lazenby, 1993: 241). He had a thankless task, against an enemy that didn’t want to fight and controlling a collection of city state militias (Bengston, 1963: 62).

Mardonius failed to control his troops and it caused his downfall. Instead of harassing the enemy on the final day, they chose to engage in direct conflict (Hignett, 1963: 344). Maybe his hand was forced (Hignett, 1963: 336), but he knew he needed a great victory here to splinter the alliance and leave Greece vulnerable. Attacking was a risky option, but it could have brought him the victory he needed. With his defeat at Plataea, the defence of Greece was concluded (Lazenby, 1993: 247). Plataea was the definitive end to the Persian invasion.

There are a many reasons as to why the Greeks won; but which battle did the Greeks use these elements in their favour enough to turn the tide of the war in their favour?

Many scholars suggest the battle of Salamis to be the crucial turning point of the war – Strauss and Holland for example (2004: 294; 2005: 399), and in some ways this is correct. Had Salamis been lost by the Greeks, the defence at Corinth would have been flanked, but it was not a foregone conclusion that they would have been defeated there (Cartledge, 2007: 166) – as we have seen, the hoplite held superiority over the lighter Persian infantry. If the Greeks knew they were about to be attacked at the sides, they could have organised their battle lines accordingly. Effectively, it meant is that the Peloponnese was safe (Lazenby, 1993: 197).

As Cartledge quite rightly points out, Marathon, Salamis and Thermopylae were battles that marked important social and cultural changes – the continuation of Greek direct democracy and the belief that the Hellenic city states could beat Persia (Cartledge, 2007: 166). The military turning point was the battle of Plataea. As long as Persians remained on their soil, the Greek poleis were not safe (Bengston, 1969: 60; Lazenby, 1993: 251). With the defeat of Mardonius at Plataea, the invasion was over. Mardonius still had sufficient troops to outnumber the Greeks and to conquer them (Cartledge, 2007: 166). Mardonius was close to winning at Plataea. If the coalition splintered, Mardonius could have picked off each polis one by one. The Greeks ended the threat of conquest at Plataea, and took the offensive to Persia afterwards.