How Large Was Xerxes' Army? 2

In the Histories, Herodotus asserts that Xerxes had over five million men (combatants and non-combatants) with him when he arrived at Thermopylae. In an earlier post, I referred to a theory by which Herodotus – who did not speak Persian, after all – arrived at such an impossible number in part through a misunderstanding, mistakenly believing that units of 1,000 men were in fact units of 10,000 men.

Robert Strassler’s remarkable The Landmark Herodotus includes an essay devoted to this topic. In “The Size of Xerxes’ Expeditionary Force,” Michael A. Flower has no doubt that Xerxes’ forces must have been orders of magnitude smaller:

The requirements of supply, maneuverability, logistics, finance and command and control would have made it impossible for Xerxes to have led an army even a quarter of the size of the one that Herodotus has given him.

Professor Flower, however, apparently rejects the idea that the miscalculation was based on the mistake described above:

How can we arrive at a more realistic figure? One very simple solution has been suggested. It is to suppose that the Greeks consistently misinterpreted a Persian chiliad (a unit of 1,000 men) as a myriad (a unit of 10,000 men), and thus one should divide Persian numbers by ten. That would given an original invasion force of 170,000 infantry. But even that is too large an army, given the inevitable problems of logistics, supply, and command. Other solutions must be sought.

In the end, a range of considerations leads Flower to the conclusion that Xerxes’ land army was probably no more than 80,000 infantry plus 10,000 cavalry, and perhaps as few as 40,000 to 50,000 infantry. On the naval side, Flower guesstimates that Xerxes’ fleet consisted of 300 to 600 triremes rather than the 1,207 claimed by Herodotus (and by Aeschylus) – a number that looks suspiciously like the number of ships that Homer says sailed for Troy.

I won’t go into greater detail because I want to motivate anyone who has an interest in Xerxes and Herodotus to run out and buy The Landmark Herodotus. It is true that The Landmark Herodotus represents a substantial investment: it currently sells at Amazon for $29.70. The Penguin paperback, in contrast, is only $8.00. Even so, if 300 has piqued your interest in Xerxes and the Persian invasions of Greece, I encourage you to consider the Landmark version.

Even if you’re generally familiar with some of the basics, ancient Greek history is strange and distant. We tend to portray the Greeks as our immediate cultural ancestors, and we can certainly see reflections of ourselves in them – that is why they have garnered so much attention and fascination over the centuries. But at the same time, the ancient Greeks are profoundly alien. In many ways, the more you learn about the Greeks, the more you realize how utterly different their views and assumptions concerning the world were from ours. (Take a look, for example, at this recent paper on The Laws of War in Ancient Greece.)

Herodotus compounds the difficulty, in several ways. First and foremost, he was composing for a contemporary audience of fellow Greeks. It was unnecessary for Herodotus to spell out his basic assumptions – about religion, morality, politics, geography, history and a host of other issues – for auditors who shared a common culture.

Second, Herodotus’ Histories are not simply an annalistic recital of events during a discrete period. Herodotus is historian, geographer, ethnographer, archaeologist, travel reporter and a few other things wrapped into one. Names of persons, places and regions race by. His delightful stories are by turns largely fictional or based in substantial fact. At times displaying logic and analysis, at others profound gullibility, he can leap from then-recent history to legend, myth and folk-tale without batting an eye.

In short, it is all too easy to read the Histories – and even enjoy them – without appreciating half of what Herodotus is saying. But to really understand the author, the work and the earthshaking events he is describing, the reader needs a lot of help. That is what the Landmark supplies. The maps alone are worth the price of admission. There are scores of them interspersed with the relevant text, so you don’t have to go around hunting for them, as well as dozens of helpful pictures and illustrations. Hundreds of annotations accompany the text providing background and explanation. At the end, there are twenty-one essays (including the essay on the size of Xerxes’ forces, discussed at the beginning) that analyze topics from the Spartan state and Greek religious festivals to the nature of hoplite and trireme warfare. A Glossary explains terms from Achaimenids to tyrant, and there is an index that is compulsive in its detail and completeness.

The picture at the top purports to depict Aeschylus (not Herodotus), who fought at Marathon, and whose play The Persians forms another primary source for the history of Xerxes' invasion.

How Large Was Xerxes' Army?

At Brett Schulte’s new Civil War group blog, TOCWOC (The Order of Civil War Obsessively Compulsed - Informed Amateurs Blog the American Civil War), Ray B has a fun post in which he applies the lessons of Thermopylae to the Civil War.

As I usually do, I’ll only pick at the details. Ray says:

A total force of some 7000 Greek and Spartan allies was able to hold off an invading force of perhaps millions. (Xerxes’ Persian army consisted of a little over 5 million, but not all were present at Thermopylae. Estimates range anywhere from 300,000-800,000 actually took part in this battle.)

Let’s look at the Persian numbers in greater detail.

In his History, Herodotus tells us that Xerxes’ infantry numbered 1.7MM men. There were also 100,000 cavalry, camel corps and chariot corps, plus 300,000 Thracians and Greeks (more than the entire population of the Balkans!), for a total of 2.1MM. There were 541,610 men manning the fleet, for a total armed force of 2,641,610 men. He then adds non-combatants (cooks, drivers, etc.) to double the total and reach the awesome number of 5,283,220.

Demographics, logistics and common sense demonstrate that these numbers are impossible. The most attractive theory that both explains how Herodotus could have come up with these numbers and suggests the true order of magnitude of Xerxes’ forces lies in the structure of the Persian army. The army was organized in a decimal system up to units of ten thousand. Units of ten men were formed into units of one hundred; units of one hundred into units of one thousand; and units of one thousand into units of ten thousand men. The 10,000 men units were known as Myriads, under the command of Myriarchs. The Myriads were then organized into corps of 60,000 men.

Herodotus understood there were 30 corps commanders, 29 of whom commanded 60,000 men each. He made his basic calculation of 1.7MM infantry simply by multiplying the number of corps commanders by 60,000.

But what if Herodotus (who did not speak Persian, after all) or his sources had confused the Myriarchs and the corps commanders. 30 Myriarchs times 10,000 men each would produce an infantry force of 300,000 men.

This number is still unreasonably high, but at least it is in the right ballpark. Other other considerations may get us closer to an actual figure.

The first is the difference between actual strength and nominal strength. As any student of the United States Civil War knows, there is often a dramatic difference between the two figures. Even without fighting a battle, tens of thousands of men crowded together create unsanitary conditions that in turn generate disease. Assuming that each Myriarch started with a full complement of 10,000 men (and that in itself is a questionable assumption – was there no recruiting fraud in Persia?) those figures would have been quickly reduced.

Second, Xerxes' army clearly included both regular troops and more lightly armed allies. By analogy to the later Roman army, the allied units may well have been smaller. In short, many of Herodotus' thirty commanders may have been commanding units of, say, 5,000 men rather than 10,000.

Finally, it is not credible to believe that the entire army simply marched to the front. Xerxes’ commanders would have left sizable contingents along the march, to guard fords and bridges, man forts, protect supply depots, and the like.

Historian A.R. Burn has pointed out that the limited water sources along the route, if nothing else, "lead[] to the conclusion that 200,000 is about the limit for the total manpower of the expedition." All in all, I would guess there were substantially fewer than 200,000 Persian infantry -- perhaps somewhat more than half that number -- in the neighborhood of Thermopylae when the Persian army encountered Leonidas and his colleagues. It's not 5,000,000, I know -- sorry to disappoint -- but still a huge number when you consider that 2,400 years later large Civil War armies were in the range of 100,000 men.