Thorax1A bell-shaped hoplite thorax of the archaic period with an extended bell-type projection in the waist, for the repulse of the enemy arrows, javelins, stones etc.

By Periklis Deligiannis

[Actually, this paper is a  subchapter of my published book: The Spartan army, Athens 2007].
In the Greco-Persian Wars (490-479 BC) between the Achaemenid Persian Empire and the defensive Alliance of the city-states of South Greece, the victory of the latter at the sea Battle of Salamis (480 BC) on Xerxes’ fleet, secured the control of the sea for them. The Asiatic fleet (mainly East Phoenician) was neutralized and fell back to the eastern Aegean. However, the Persian army remained almost untouched. King Xerxes, fearing the possibility being trapped in Greece and eventually captured or killed after a possible defeat on land, withdrew “discreetly” in Asia officially considering that the objectives of his campaign had been achieved. Before he withdraws, he left his cousin Mardonios (Mardonius in the Western historiography) as head of the army in order to continue the military operations. Mardonios was a stubborn and brave man (his name means the “gallant” in ancient Iranian, originating from the word “mard” for the man or the warrior). On the other hand, in the winter of 479 BC a change occurred in the Spartan military leadership, which proved to be very important for the Greek defense against the invasion. Shortly after Salamis, the Spartan royal commissioner (regent) Kleombrotos died. His office was occupied by his son, Pausanias.

Mardonios initially tried to gain over the Athenians. But the victors of the Battle of Marathon (490 BC) would not be subdued to the losing side in that battle, and twice rejected the tempting terms that he offered them, assuring at the same time the Spartan envoys who were at Salamis Island (the military base of the Athenian army and fleet) that they would never betray their Greek compatriots. Till that moment, the Spartans were avoiding the confrontation with Mardonios’ army. But at that time, they were pressed even more intensely by their Athenian, Megarian, Plataean and Aeginetan allies whose countries were either occupied by the Persians or directly threatened by them. The Spartans had to satisfy the demand of their allies and finally sent their army led by the regent Pausanias, to face the invaders who had already occupied Attica (the territory of Athens) for the second time during the Second Persian campaign (480-479 BC). The women and children of the Athenians had long ago found refuge in Peloponnese and the small islands of the Saronic Gulf. The Athenian resistance was concentrated in the Island of Salamis, where they had defeated the enemy fleet almost a year ago.

When Mardonios was informed that the Spartans were campaigning against him, he evacuated Attica. Before leaving Athens, he destroyed what was left after the first Persian conquest of the city. Herodotus mentions the reason for the withdrawal of the Persian army from Attica. The country’s ground was not suitable for the tactics and the charge of the Persian cavalry on which Mardonius was based to achieve victory over the Spartans and the other allies. Moreover if the Alliance won the battle in Attica, the Persian army would be threatened to be isolated there and eventually be destroyed. According to Herodotus, the neighboring powerful city-state of Thebes (in Boeotia, north of Attica) was friendly to the Persians and the most important, the valley around her was appropriate for the tactics of the Iranian cavalry. On the pro-Persian role of Thebes and the other Boeotian city-states (except Plataea and Thespiae) in the Greco-Persian Wars and some arguments against Herodotus’ references, you can read my article: THE PRO-PERSIAN ROLE OF THEBES & BOEOTIA IN THE PERSIAN WARS: MYTH AND REALITY.


Περσικοι πολεμοι χάρτης

Map of the Greco-Persian wars.


Α detailed map of ancient Attica and Boeotia. The major area of Plataea is in the center (Penguin Books).

The Persian commander withdrew his army following the road of Decelea via Parnes Mountain ending at Tanagra in Boeotia. The Spartan general Pausanias led the allied Peloponnesian army to Megara where it was reinforced with the Megarian forces, and then to Eleusis where it was joined by the Athenians and the Plataeans who had landed there coming from Salamis. The reinforced Greek army directed north and following the passage of Dryos Kephalae, crossed Cithaeron Ridge, the physical boundary between Attica and Boeotia. Thus ended its march at Parasopia, the southern region of Boeotia (named after the river Asopos). More accurately, it was the Upper Parasopia, the territory of the city-state Plataea which was a member of the Greek alliance and had been destroyed by the Persians. After a year, the Plataean hoplites were again marching on the ground of their motherland.
The Greek army stopped at the Boeotian town Erythrae. Mardonios following his strategic plan, did not prevent the Greeks from passing Cithaeron Ridge because he hopped to lure them in the Boeotian plain. It seemed that he had managed to lure his opponents in a battlefield suitable for crushing them with his cavalry. The Greek army had no cavalry, because the Boeotians, Thessalians and Macedonians, the only Greek peoples who had considerable cavalry forces, were fighting on behalf of the Persians (most of them unwillingly, being already vassals of the Great King). However the Greek allies were not deceived by this suspected inertia of their enemies as Mardonios hoped that they would be, and stopped their advance south of the Asopos River. The reason for their stop was Parasopia’s topography. The ground in the north of the river was flat, suitable for cavalry action. Unlikely in the south of the river, the ground was interrupted by large cracks and hills being appropriate for the protection of the Greek army from the enemy cavalry.

It is obvious that the strategic objective of the Persian general was to give battle against the Greeks on the plain north of Asopos. The purpose of the latter was to lure the Asiatic army in a conflict south of the river, where the formidable cavalry of the Iranian peoples would be largely useless. Mardonios had been prepared for the clash with the enemy gathering his army north of the river. He had deployed the Persian warriors on the right horn of his order of battle, on the center he had lined up the other peoples of the Empire, and on the left horn his Greek vassals. The Persian army had already built in the rear of his order, near the town Skolos, close to Thebes, a large foursquare fortified camp. Each side of the camp was extended for ten ancient stadiums (about a modern mile), thus it covered an area of a square mile. Mardonios had deployed outpost units in the south of the Asopos River, between his order and the towns of Erythrae and Plataea.

The Greek allies established their camps in the northern foothills of Cithaeron, in the area south of Asopos, approximately between Erythrae and the town of Hysiae, in rough terrain providing protection from the Iranian cavalry. The forces of the individual Greek city-states were deployed on the following order of battle: the Spartans occupied the honorary right wing, the other units were installed on the center and the Athenians lined up on the left wing.

After a waiting period, Mardonios ordered Masistios, the cavalry commander, to attack the enemy. It was partly an action of reconnaissance and partly a tactical maneuver. It can be considered as a tactical move because Mardonios hoped that the Greeks would be lured into the crossing of Asopos. The latter view is compounded by the fact that the Persian cavalry strangely attacked not in its usual cautious style, but in an impetuous mode straight ahead to the armored Greek hoplites. The Megarian hoplites especially had suffered large pressure from this attack due to the unsuitability of the low ground where they were lined up. When they informed Pausanias for their difficult situation, he sent three hundred Athenian elite hoplites and archers who were nearby, to help them. The Athenian army had an archers’ corps suitable for dealing with cavalry. With their arrival, the clash became bloodier and the Athenians managed to kill Masistios himself. However, the Iranian cavalrymen continued fighting stubbornly and the conflict was generalized due to the arrival of more Greek reinforcements. When the latter rose significantly, the Persian cavalry retreated.

Mardonius’ offensive maneuver ended. The initiative passed to Pausanias who moved his army to the west, in the area of Plataea, through Hysiae and following the northern foothills of Cithaeron. Herodotus does not mention the reason for that decision. However, the fact that the area of Plataea is more flat than the first site of the Greek camp denotes that Pausanias “threw down his gauntlet” to Mardonios, calling forth for him to send his cavalry once more. I’ll deal below with the strategic probability that he intended to use the tactics of double envelopment on the Persians. Another reason for the move of the Greek army at its new (second) positions may have been the lack of water in the old site of its camp.
The Spartans were lined up again on the right horn, near Gargaphia Spring and the Athenians on the left one. Both were installed on hills. The hill of the Athenians had the characteristic name “the Tower”. The other troops of the Alliance took positions in the valley between the two hills, being again exposed to the enemy cavalry. Once Mardonios saw the movement of the Greek army, he shifted his forces to the west, always north of Asopos and opposite to his enemies.


Herodotus mentions the sizes of the opponent armies, which have been the subject of many modern debates and controversies. Mardonios had 300,000 warriors while Pausanias commanded 38,700 hoplites and 71,300 light infantry combatants, a total of 110,000 men.
The figure of Mardonios’ troops is definitely exaggerated. Most probably his alleged 300,000 men were in reality the total number of Xerxes’ land warriors and crews of his fleet at the beginning of the invasion in Greece; the 120,000 of them being the crews of the warships (mainly triremes). Thus, taking into consideration that Xerxes returned to Susa after the naval defeat at Salamis with a significant part of the land army for his own safety, some ‘natural losses’ because of diseases or other causes (desertion etc), but also the reinforcement of the Persian army with Xerxes’ Greek vassals and allies, we conclude that in Plataea Mardonios had at his disposal around 150,000 men, combatants and non-combatant auxiliaries.

α1A  Greek hoplite with a Corinthian helmet, a hoplon (Argive shield) and a kopis sword (κοπίς) leaning on his shoulder (reenactment by the British Hoplite Association).
The method by which the Achaemenid army was trying to defeat his opponent was not through tactics or some superior equipment of its warriors or by their militancy, but mainly by exploiting the vast numbers of its combatants. The Persian commanders were trying to crush the enemy using the unbearable pressure of their endless military manpower or by the storm of arrows of their countless archers (for example, battles of Thermopylae, Issus and Gaugamela). The only part of the Achaemenid army that used to fight in somehow “realistic tactics” in military terms, was the cavalry. The latter was of high quality, consisting almost entirely of Iranians, namely Persians, Medes, Sagartians, Saka, Massagetae, Sakaurakae, Bactrians, Sogdians and others. These peoples were very experienced in cavalry warfare. On the contrary, Mardonios’ infantry was mostly of low quality, largely composed of light armed tribal warriors, coming from the same Iranian nations and from many other peoples of the empire. The concentration of such large numbers of men would not be difficult for the Persians because, among other things, there was no significant cost for their light equipment (it has been estimated that some of them were armed with their own poor hunting equipment). Furthermore it is certain that many of them were not combatants but servants, supply and logistics personnel and other non-comb. Moreover an imperial army which was operating thousands of miles away from Susa and Persepolis obviously needed a large number of such non-combatants.

Some researchers have attempted to calculate the size of the Persian army based on the number of men who could have been accommodated in the Persian camp as this is described by Herodotus. As correctly noted in Wikipedia, the numbers given by these calculations are usually between 70,000 and 120,000. The British historian Lazenby calculates 70,000 men, including 10,000 cavalry, making a comparison of the Persian camp with the later Roman ones, which is in my opinion is an error. The Roman camps can not be compared with the Persian camps four or five centuries earlier for many different reasons. To name only one, the sophisticated and specialized military material of the Roman army from the common armors to the siege engines, heavy engineering equipment etc. which generally the Persian army lacked, required large storage areas. That is why Lazenby’s assessment is rather low. Furthermore this is not the first time that this in anyway very important historian, has fallen into errors concerning calculations on ancient military forces (for example on page 28 of his book “The First Punic War”, 1996).
For similar reasons, I reject Delbrück’s assessment, which was based on the distance that the Persian army marched in one day when Athens was attacked. Delbruck calculated a number of 75,000 men in total, combatants and non- combatants. Apart from the fact that such a criterion for the calculation of the Persian army is not considerably reliable (for example, it is possible that Mardonios had left much of the army in Boeotia in order to march faster with the rest), the strength of the army could never be so small, because at it has been mentioned above, the primary Persian method of prevalence against a hostile army was not the tactics or the superior equipment or anything else, but the overwhelming superiority in military force. The Persians tried to defeat their opponents mainly by crushing them with the weight of their countless Asian and Egyptian warriors or at sea with the colossal trireme fleets that they could gather.
Peter Connolly has estimated the number of the Persian army based on the size of its camp, at 120,000 men which is a more reasonable figure. However I note a probable error that all the aforementioned scholars have fallen into: they have not taken into consideration that Mardonios had at his disposal an even larger and stronger camp in the immediate area: the city of Thebes as well as some other cities of Boeotia. Thebes and some other nearby cities could serve as camps at least for most of his Greek allies who numbered a total of 50,000 as reported by Herodotus. If we add the above mentioned number of 120,000 men of the Persian camp according to Connolly’s reasonable estimate to the figure of the Greeks who could have been accommodated in Thebes and other neighboring cities, we get a total number of 170,000 or at least 150,000 Mardonios’ men which is almost the same as the number that we estimated above by another method.
Some researchers have estimated that the real combatants of Mardonios did not exceed 100,000 including the “medizing” (pro-Persian) Greeks, and this is my own assessment as well. It is generally likely that each Asiatic combatant was supported by a non-combatant auxiliary, who could be armed with a javelin or other light weapon in an emergency. Thereby from the above mentioned 150,000 men that Mardonios had gathered in the area of Plataea on my estimation, the real combatants numbered 75,000-100,000 depending on the seriousness of the conflict. I believe that in the late phases of the battle of Plataea several auxiliaries would have been hastily armed, raising the combat forces to 100,000 but many of the latter would have been of poor quality.

On the types of combatants of Mardonios’ Achaemenid army, there were three main types: the cavalry, the so-called “heavy” infantry and the light infantry. Chariots are not mentioned. We spoke about the cavalry and light infantry above (in the near future there will be a series of analytic articles on the Achaemenid Imperial army). The “heavy” infantry was composed of the Immortals (the imperial guard of the Great King) and some units of Medes and Persians. In fact, compared to the armoured Greek hoplites, those warriors were light infantry as well for many reasons (lack of armour, usually even of helmets, their tactics and other reasons). I believe that Xerxes before leaving for Asia, left to a considerable part of the Immortals in order for him to have at his disposal an infantry corps of some account to confront the formidable enemy hoplites. Mardonios’ 7,000 Medizing Greek hoplites could not do much against the 38,700 hoplites of the Alliance.

Iranian Achaemenid Cavalry

Achaemenid Iranian Cavalryman

The Medes, Persians, Kossaeans (or Kissians) and Hyrcanians were the military core of the Achaemenid imperial army. In the battles, the Kossaeans joined the Persian units and the Hyrcanians joined the Medes. The Bactrians, Sogdians, Saka (mercenaries), Sagartians, Arians and Arachotians had also a major role in the army, being Iranians as the aforementioned peoples.

Concerning the “medizing” Greeks who fought for the Persians, Herodotus quotes a figure of 50,000 men. It is not difficult to calculate the forces that they provided to Mardonius, based on the numbers that they usually mobilized in wartime. The lightly armed Macedonians, Thessalians and the perioikoi (vassals) of the latter (Malieis, Perrhaibians and others) numbered a few tens of thousands of men, but it is known that they were poor-trained troops and it is very doubtful that Mardonios would dare to line them up against the enemy hoplites. Additionally, the armies of those Greek tribes lacked hoplites. Instead, the Macedonian and Thessalian cavalry was of quality similar to that of the Iranian cavalry. It is well known that the Thessalians had around 6,000 cavalrymen, but the higher number of those who usually mobilized for a campaign was 2,000 as it is demonstrated by various ancient references (most prominently the 1,800 Thessalian cavalrymen of Alexander the Great). The figure of 2,000 Thessalian cavalry in Mardonios’ army is the most likely.
We also know that Macedonia before King Philip II had 1,000-2,000 cavalry (as has been demonstrated during the Peloponnesian War), thereby in proportion to the aforementioned Thessalian contribution to the Persians, Mardonios’ Macedonian cavalrymen would had not numbered more than 500. The Greek city-states of Chalkidike and Thrace gave no land forces because they were charged providing warships and crews for the Persian fleet. Considering the military forces that the Boeotians usually mobilized in the 5th-4th centuries BC, and the fact that the Plataeans and Thespians were fighting on behalf of the opposite side, the combatants that the former provided to Mardonius would be rather 5,000 hoplites and 500-1,000 cavalry (also of high military quality). Similar calculations lead us to estimate the Phocians and Locrians in Mardonios’ army to 2,000 hoplites and (a few) horsemen.

Illyriko kranos

Doric helmet of the Illyrian type. This type was invented by the Peloponnesian Dorians but later it became popular in Illyria.
Thereby, the medizing Greeks could provide to Mardonios 7,000 hoplites and 3-3,500 cavalry at most and a few tens of thousands of light infantry. Herodotus quotes that they formed the right horn of the Persian battle line, which might give the false impression that they consisted a much higher percentage of the Achaemenid army. But the ancient historian himself states that the lines of Mardonios’ “shock troops” – being the Iranian peoples (Persians, Medes, Sakas, Bactrians), the Indians and the Greek vassals – were framed by the more numerous warriors from other nations of the Achaemenid Empire (Thracians, Phrygians, Mysians, Ethiopians, Armenians, Arabs, Egyptians and many others). They were obviously lined behind the medizing Greeks, Iranians and Indians (according to the Persian tactics) or possibly on their sides, being in any case the majority of the army.

Concerning the Greek Alliance, the Spartan army at Plataea consisted of 50,000 men according to Herodotus. This is ultimately a number of men that Laconia and Messenia (comprising the territory of Sparta) could provide. 10,000 of them were hoplites (5,000 Spartans and 5,000 perioikoi [free Lacedaemonians but non-citizens]) and 40,000 were light-armed auxiliaries (35,000 helots and 5,000 perioikoi). Every Spartan was accompanied by seven helots and every perioikos hoplite was accompanied by a perioikos auxiliary. The Spartans were joined in the right horn by their trustful Tegean allies (1,500 hoplites). The Spartans and Tegeans would fight against the Persian warriors who had been relocated in new positions opposite to them. The 8,000 Athenian hoplites were supplemented by 3,000 Megarian and 600 Plataean hoplites, forming the left wing which was going to confront the Medizing Greeks.
The other Greek allies forming the center of the battle line, 15,600 hoplites in total, would face the Medes, Sakas, Bactrians and Indians whom Mardonios had lined up opposite to them. The Tegeans and the Greek forces of the left wing and the center were accompanied by light infantry warriors approximately of equal figure with the respective hoplites. The Athenians had also a corps of 800 archers. As mentioned above, the total number of the 110,000 Greek combatants has been considered as protuberant by some historians. However the Greek regions that resisted the Persian invasion were heavy-populated (archaeologically proved) and thus had no difficulty in gathering 110,000 men for a battle of “life or death”. After all 71,300 of them were light infantry (peltasts, “psiloi“, javeliners, slingers, stone-throwers, archers and others) and just 38,700 were heavy infantry (hoplites). There was no cavalry.
However, in reality not all of the 71,300 alleged “light infantry” were combatants. Modern researchers have mainly questioned the number of the 35,000 helots but they do not seem to be right on this matter, because those helots had been “dragged” to Attica by their Spartan masters, in order not to rebel against them if they remained in their homeland. It was a measure of the Spartans to avoid a helot revolution at home, on their rear, while they would be fighting the Persians. Considering the ratio of hoplite to light infantry for the forces of the other city-states at Plataea which was 1:1, this was rather the case for the Spartans and the armed Helots who accompanied them. That is to say, from the seven helots accompanying each Spartan hoplite, only one helot was lightly armed while the other six were non-combatant auxiliaries; to wit 30,000 helots (from the 35,000 in total) belonged to the latter category. Thereby from the 110,000 Greek allies at Plataea, probably only 80,000 were combatants. It is also probable that some of the other Greek alleged ‘light infantry’ who accompanied the hoplites, were actually non-combatant auxiliaries so perhaps the number of the real combatants was less than 80,000.

The new location chosen by Pausanias to deploy the army has been the subject of research for most scholars, who find it difficult to understand his strategy. The most probable case – as has been estimated – is that the Spartan commander intended to use the tactics of double envelopment of the enemy army, by which Miltiades (the Athenian commander) had crushed the Persians at Marathon eleven years ago. He probably hoped that the enemy cavalry would attack the Greek center located in the lowlands and thus both the two horns of his battle order – the Spartans and the Athenians – would gain the opportunity to encircle and destroy it. If Mardonios’ cavalry was destroyed (being the only corps of the Achaemenid army that the Greek allies considered as formidable), his infantry could not hold back the charge of the hoplites.


A detailed map of the battle of Plataea, describing among other data, the three successive positions of the two armies.
Pausanias and Mardonios kept their troops alert, waiting for the attack of each other. On the eighth day after the relocation of the two opposing armies in their second position, Mardonios upon recommendation of the Theban commander Timagenidas, sent his cavalry to occupy the passage of Cithaeron at Dryos Kephalae from where the Greeks were resupplied. The Persian cavalrymen met there a convoy carrying food and other supplies for their enemy, which they destroyed. After that, the former were conducting fierce attacks for two days against the Greeks, nailing them down to their positions. The result was that the latter did not have any more at their disposal the water of the Asopos River.

On the eleventh day after the new relocation/encampment, Mardonios convened a council of war, in which he strongly argued with Artabazos, his lieutenant commander (the second in command in the Achaemenid army). Mardonios insisted on an offensive movement against the Greeks while Artabazos and his supporters believed that the army had to retreat to Thebes in order to ensure its supply and better defense. Artabazos considered that they must avoid any battle against the enemy in order to gain as much time as possible for the Persian agents who were trying to buy over as many officials of the cities of the Greek Alliance as possible. Mardonios was urgent to give battle because of the problems of supplying his populous army because the western Aegean Sea was controlled by the Alliance fleet. I believe that Mardonios was afraid that sooner or later the Greek naval forces would go on the offensive to oust the Persian fleet from the Aegean. The latter was anchored in Samos Island. A Greek naval victory in the Aegean would cause a general uprising of the Greeks of Thrace and Asia Minor. The revolt would shortly spread to the Thracians (a non-Greek people), Macedonians, Thessalians and possibly to the Boeotians including the powerful Thebans. The latter were bitter enemies of the Athenians – which situation was possibly the main reason that they ‘medized’ – but they could not remain the sole pro-Achaemenid people in the geographical center of the Greek mainland, thereby they would probably turn against the Persians. In this case, the Achaemenid army would be isolated in the geo trap of Greece and would be wiped out. I believe that Mardonios’ opinion was the right one.

For the mentioned reasons, Mardonios overcame Artabazos’ objections and the next day ordered a general attack of cavalry against the enemy. The Persian cavalry managed to oust the Greeks from Gargaphia Spring whose water supplied many of them, and then destroy the fountain. The Greeks had already lost contact with Asopos while their supply was problematic after the destruction of the convoy at Dryos Kephalae. The army of the Alliance was facing thirst and hunger and Pausanias convened a council of war because of this danger. The Greek generals decided to move the army during the night in the area of Plataea, in positions where they could ensure water and food supply and the effective protection of the army from the enemy cavalry. But the situation was dangerous for the army of the Alliance.
Periklis Deligiannis