The Size of the Army of Xerxes
In Warfare in the Classical World, p.28, John Warry credits the Persians with 130,000 foot and 20,000 horse supported by 1,200 triremes at Thermopylae and Artemisium. The Greeks have 7,000 infantry and 271 triremes.
In The Spartans by Paul Cartledge, p. 118, the author states, “Herodotus reports a total of 1,700,000 backed up by well over 1,000 ships. Sober modern estimates by the best military historians cut down the Persians’ land forces to numbers estimated as low as 80,000 to a quarter of a million and the navy to roughly 600 ships.â€
In Thermopylae, The Battle for the West, pp 33-34 and 87, Ernle Bradford covers the topic at some length.
“Although it is true that Herodotus, who was born some four years after the invasion of Xerxes, had access to all the records available, it is impossible to accept the figures that he gives for the size of the Persian army and of the fleet. If one first of all bears in mind that Herodotus was trying to make his figures square with a famous war memorial that had been set up in the pass at Thermopylae, it is not so difficult to see where his added noughts come from in his computation of the Persian numbers. The memorial in traditional Spartan or laconic style reads:
Against three million men fought in this place
Four thousand Peloponnesians, face to face.
“The fact is that the Greeks, when it came to numbers beyond their normal usage, tended to use the term ‘myriads’ (tens of thousands) as we, centuries later, loosely use millions or billions – meaning no more than an almost uncountable number.
“The figures as given by Herodotus show an army totalling 1,700,000 infantrymen, 80,000 horsemen, a camel corps and chariot contingent numbering 20,000, and a mixture of Greeks from Ionia, the islands and Thrace, to the total of 300,000. Burn adds the dry but apt comment
…finding himself still short of the war-memorial’s three million, he cheerfully doubles the whole total to allow for non-combatants (cooks, drivers, women – the Guards are reported to have brought their women along in wagons) and reaches a grand total of 5,283,220. The most remarkable thing, he adds, with a decent descent into realism, is how such a multitude was fed.
“Other scholars and military historians have debated the size of the army – and of the navy – but the most realistic viewpoint seems to be that Herodotus confused the Persian term myriarchs, which meant commander of 10,000 men, with the other named commanders who, in their lesser sphere, commanded no more than thousands or hundreds. (The Persians worked on the decimal systems.) If one removes a nought from all of Herodotus’ figures one comes up with an army of 170,000 infantrymen, 8000 cavalry, 2000 camel corps and charioteers, and 30,000 Greeks and Thracians. This seems a far more likely figure in view of the populations (as far as they can be conjected) at the time. It would still make sense in that it would nevertheless suggest to a Greek accustomed to battles involving at most a few thousand men an almost inexhaustible flood of troops.
“General Sir Frederick Maurice, who had the opportunity of covering the area of the march of the Great King not long after the First World War, came up with the conclusion that the total of the Persian army was about 210,000. Unlike most desk-bound scholars he had the opportunity to travel the whole area, and had excellent military and logistical knowledge of the terrain. He based his conclusions particularly on his observation of the water-supplies available. Maurice had also had experience of moving British military units together with animal transport, and he reckoned that such a force would probably have needed with them about 75,000 animals. Even at this, he reasons that what has sometimes been taken as an unbelievable comment by Herodotus, ‘except for the great rivers, their fighters drank the waters up’, was probably correct. A river, of course, unlike a pond or even a lake, cannot be drunk dry in one sense, for it is constantly being reinforced. One may also reasonably assume that the rivers in Asia Minor at that time were somewhat larger than they are today. Centuries of the ubiquitous goat, killing saplings, leading to deforestation, coupled with land changes in the earthquake-prone area of Turkey have certainly depleted the forests as well as interfering with natural water sources.
“Nevertheless, working on whatever system one prefers, it seems that there is no possibility of the army of Xerxes having exceeded 250,000 men. Even this number, together with all their animals, baggage train and (possibly) camp followers, would have been sufficient to exhaust the water resources at a number of places along their route.â€
And, concerning the crossing of the Hellespont:
“On the salient question of the water-supply of the invasion army no authorities can equal Maurice, with his practical military experience as well as his personal knowledge of all the terrain covered by it in the long march. To paraphrase his conclusions after the forces had left Asia Minor – limited quantities of water could have been obtained by boring ‘but this was beyond their resources’. At the point where the columns crossed, although there was water at Maidos near the bridges this was not on their route, so throughout this stage the army would have had to carry its own water with it. Except for occasional springs and wells, it seems that the troops would have had to take along four days’ supply with them in water–skins. It is possible that, along with the careful preparations of food depots, water-troughs had been erected for the animals in open ground around Gallipoli and at the northern end of the marsh midway between Melas and Aenos, which could have been kept filled by regular convoys from the Melas itself. The allowance for the troops would have worked out at about two quarts per day, ‘not an over-generous allowance for men marching in hot weather, whose food is dry grain’. Taking the estimate of the grand army’s numbers as 210,000 this would have amounted to 420,000 gallons of water. Presuming that much of this was carried by the camel corps, something like 15,000 camels would have been required (a good camel being capable of carrying 300 lb of water). It is the essential matter of water-supply which disproves the Greek tradition of the army being composed of three million men.â€
Peter Connolly clarifies and concurs with this estimate in Greece and Rome at War, p. 12.
“The ancient Greeks believed that Xerxes’ army numbered three million plus camp followers – Herodotus gives the total as five and a quarter million, but he is clearly skeptical about how such an army could be fed. At the end of the 1920’s, General Sir Frederick Maurice made a detailed study of Xerxes’ route from the Hellespont, examining in particular the problem of water supply, and concluded that the Persian army could not have numbered more than 210,000 men plus 75,000 animals. It seems probably that the rainfall at this time was considerably higher than today, therefore these figures could be increased slightly. Even so, the figure could hardly have been over 250,000. Of this number about three quarters would have been combatants.â€
Peter Green also goes along with these figures for much the same reasons in The Greco-Persian Wars, pp. 58-60.
“What was the actual size and composition of the force with which Xerxes invaded Greece? Even though Herodotus had access to the official Persian army roster, his overall figures (for the land forces at any rate) are flatly incredible: 1,7000,000 infantry, 80,000 cavalry, 20,000 chariots and camels, and 300,000 Thracians and Greeks picked up en route – which latter figure, as Burn says, ‘might be tolerable as a guess at the whole manpower of the Balkan Peninsula’. Reducing these astronomical totals to reasonable proportions has kept scholars and military historians busy for many years, without any final consensus of opinion being reached. However, it may be possible to arrive at a rough estimate. The Persian command structure operated – at least up to corps level – on the decimal system, with officers commanding units of 10, 100, 1,000, and 10,000 men. One attractive theory suggests that Herodotus may have confused the Persian terms of chiliarch and myriarch (the two highest ranks in this list), thus automatically multiplying all his figures by ten. Strike a nought from every total given above, and the picture at once becomes far more plausible: 170,000 infantry; 8,000 cavalry; 2,000 chariots and camels; 30,000 Thracians and Greeks = 210,000.
“It remains to check this figure against estimates reached by different methods, and to integrate it, if possible, with Herodotus’s allotment of the various army corps to thirty generals under six field-marshals.
“According to Munro and others, the thirty generals were in fact myriarchs, commanding a total of 300,000 men: this figure was then increased to 360,000 on the assumption that each general also had 2,000 cavalry at his disposal. Taking the Herodotean figure of 60,000 as the paper strength of a Persian army corps, Munro divided his total of 360,000 between the six field-marshals. From various slight indications (e.g. the division of the expeditionary force into three operational columns) he assumed that Xerxes took no more than half his available military reserves – i.e. three full army corps – into Greece, thus reaching a provisional invasion total of 180,000 men. This theory has been criticized in detail – most telling is Burn’s comment that, if we accept Herodotus’s list of regional contingents, the thirty divisions cannot possibly have been uniform in size – but it does agree substantially with the figures produced by decimating the Herodotean total. General Sir Frederick Maurice tackled the problem in a quite different manner, by traveling over the Gallipoli Peninsula, observing local conditions (the availability of water in particular), and applying his knowledge of military logistics to work out how large a task force Xerxes’ known route could possibly have supported. He concluded that the Persian army, at most, numbered 210,000 (of which perhaps 150,000 would be fighting men), together with perhaps 75,000 cavalry horses and pack animals.
“Thus, by several independent methods, we reach a very fair modicum of agreement over absolute totals. As regards general troop-availability, we know that several army corps stayed behind on garrison duty throughout the empire: Greeks were too prone to credit the Persians with their own practice of total emergency conscription. Xerxes’ six marshals each probably commanded a corps of about 30,000 men, with two corps forming an operational field army. Herodotus’s thirty generals (archontes, a significantly vague term), command forty-six regional contingents between them. Fourteen of these are large enough to rate a commander of their own, while the remaining thirty-two are brigaded into fifteen operational units. Bearing Burn’s criticisms in mind, it seems better to posit a varying strength for each command – as we would expect considering their nature – rather than to start from the assumption that these generals were myriarchs in any but a titular sense. We know that Hydarnes, the commander of the Immortals – Xerxes’ crack Guards Division – did, in fact, have 10,000 men under him; but this was a special case, and Herodotus goes out of his way to emphasize, as a fact worthy of note, that the Immortals were always kept up to strength, with reserves ready to replace battle-casualties or those who died of disease. Other units, it is clear, did not enjoy such favoured treatment. Any of Hydarnes’s twenty-nine colleagues who had six thousand men on his muster-roll probably considered himself lucky.â€
In The Wars of the Ancient Greeks, pp 95-96, Victor Davis Hanson says:
“While there is no credibility in ancient accounts that the Persian army numbered more than a million men, we should imagine that a force of even a quarter to half a million infantry and seamen was the largest invasion that Europe would witness until the Allied armada on D-Day. Neither do we need to agree with ancient accounts that the Persian cavalry numbered over 80,000 horse. But it may well have been half that size, still nearly 5 times larger than the mounted forces that Alexander would use to conquer Asia more than a century and a half later. To the Persians, the real trick was just in assembling such a horde and getting it into Greece intact. In comparison to thatlogistical nightmare – the Persian army was perhaps three times larger than Sherman’s entire federal force that cut a 60-mile swath through Georgia on its march to the sea – the destruction of Greek military forces was felt to be relatively simple.â€
