Persian Military and Warfare: From the Immortals to the Cataphracts

The image is in front of you the moment you step into the Persepolis museum, or open the second volume of Schmidt’s excavation report, or look at almost any introductory book on the Achaemenids. The Apadana staircase: the king on a high throne, the columned hall behind him, and twenty-three tribute-bearing delegations climbing toward him in long, carefully separated lines. The Persians and Medes lead, then the Elamites and the Parthians, the Bactrians and the Indians, the Arabians on camels and the Ethiopians with ivory, the Lydians with their gold vases, the Egyptians in linen. The picture is unmistakably a picture of a court in peace, not a battlefield. But the picture also tells you, without saying so, that the Achaemenid state is the sum of its peoples, and that the people who run it are doing the running. That’s the imperial idea. The military is the means by which the idea is held together, and the reliefs show you the means as well as the idea: every male on the staircase is armed, and the weapons on the staircase are the same weapons that you see in the reliefs of Darius’ tomb at Naqsh-e Rostam, and the same weapons that the Roman legionaries would later meet on the field of Carrhae.

Three points are worth making at the start, before we get to the chronology. First, the Persian military tradition is the longest continuous military tradition of which we have evidence in the ancient world — from the Median contribution to the fall of Nineveh in 612 BCE to the Sasanian defeat at al-Qadisiyyah in 636 CE, a run of more than a thousand years. Few traditions match it. Second, the tradition is poorly documented, in a way that the comparable Greek and Roman military traditions are not. The Persian sources are mostly silent on military matters: the Behistun inscription is about legitimacy, the Persepolis Fortification Tablets are about rations, the royal inscriptions at Naqsh-e Rostam are about succession. We depend on Greek, Latin, Armenian, Syriac, and Arabic sources, all of them written from outside. Third, the tradition changed. The Achaemenid, Parthian, and Sasanian armies are related to each other, but they are not the same army, and treating them as a single “Persian way of war” is a useful shorthand that the evidence does not strictly support.

The standard English synthesis of the Achaemenid side is Pierre Briant’s From Cyrus to Alexander (2002), with Josef Wiesehöfer’s Ancient Persia (1996) the closest competitor. For the Parthian side A. D. H. Bivar’s chapter in the Cambridge History of Iran, vol. III (1983), is still the most thorough treatment in English, with M. A. R. Colledge and various recent Parthian coin studies as supplements. For the Sasanian side, Parvaneh Pourshariati’s Decline and Fall of the Sasanian Empire (2008) overturned the older narrative, and Touraj Daryaee’s Sasanian Persia (2009) is the best short introduction. Kaveh Farrokh’s Shadows in the Desert (2007) is a popular synthesis that I find useful on details but not reliable on historiography. The Greek and Latin sources — Herodotus, Xenophon, Ctesias, Strabo, Plutarch, Arrian, Quintus Curtius, Ammianus — are all in translation in the Loeb and Budé series, and the relevant volumes of the Encyclopaedia Iranica are essential for specific points.

Achaemenid armies and the problem of the evidence

The Greek sources are abundant but interested. Herodotus wanted to explain the Persian Wars; Xenophon wanted to lionise Cyrus the Younger and at the same time excuse the Greek mercenaries who served the Achaemenids; Ctesias, the Greek physician at the Persian court in the late fifth and early fourth centuries, wanted to entertain. The Persian sources are few and mostly silent on military matters. The upshot: any account of the Achaemenid military is, in large part, an account of what the Greeks said about it, filtered through modern scepticism.

The test case is the army of Xerxes that invaded Greece in 480 BCE. Herodotus puts it at 2.6 million men, plus a support train. No serious modern historian accepts this; the logistical impossibility is obvious. The consensus estimate, after generations of argument, is somewhere between 60,000 and 200,000 effectives, with most scholars favouring a figure in the lower half of that range. W. Kendrick Pritchett’s The Greek State at War (vol. V, 1991) is the most thorough demolition of Herodotus’ numbers, and the consensus is not in serious doubt. The army drew troops from at least forty-six subject nations, which is itself significant: this was a cooperative force, held together by pay, prestige, and the political culture of the farr (the royal charisma), not by ethnic homogeneity.

The Ten Thousand Immortals are the most famous unit in the Achaemenid army, and the most misunderstood. Herodotus’ description — a force of exactly 10,000 heavy infantry, perpetually maintained at full strength, with wicker shields and short swords — is vivid but almost certainly idealised. The name probably refers to a unit that existed; the precise number, the perpetual replacement, and the wicker shields (a Greek cliché for “Easterners”) are likely a Greek literary construction. Maria Brosius’s The Persian Empire from Cyrus II to Artaxerxes I (2020) treats the Immortals as a real but smaller unit, perhaps a thousand or so, which served as the royal bodyguard and a strategic reserve.

I think the more useful question is what the Achaemenid army was for, not what it looked like. The answer is that it was an instrument of imperial policing and of selective offensive war. Cyrus used it to expand; Darius used it to consolidate; Xerxes used it to settle a score. The army was, in practice, a cavalry-and-chariot force with infantry support, not the reverse. The Persian tradition of mounted warfare went back to the second millennium BCE. The Median contribution to the destruction of the Assyrian Empire in 612 BCE was largely cavalry. Cyrus’ victory over Croesus of Lydia in 547 BCE depended on a force that included both horse and foot, with the horse element doing more of the decisive work than Herodotus admits. The Achaemenid reliefs at Persepolis and the glazed bricks of Susa show the mounted archer repeatedly: long cloak, quiver, sometimes a tasselled fillet on the brow. The infantry was a mosaic — Babylonian heavy foot, Indian archers, Arabian camel corps, Egyptian marines, Bactrian horse — and the Achaemenid doctrine was combined arms, with the cavalry working the wings and a heavy infantry reserve held back to deliver the decisive blow.

The Persians were not amateurs. They were fighting a different kind of war from the one the Greeks were good at.

What the Achaemenid army actually had

Iranian cavalry was the distinctive feature of the Persian military tradition, and it is the one that endured. The Achaemenid horse was drawn from the great breeding regions: Media, Armenia, Hyrcania (the Caspian), Parthia, Bactria. The Nisaean horse, from the Median plains, was the most prized. The Persian rider was typically a horse archer — bow, light spear or javelin, short sword — trained to shoot accurately from a galloping or retreating horse. Achaemenid cavalry tactics emphasised the Parthian shot — the ability to fire accurately to the rear while in full retreat — long before the Parthians. Xenophon, who spent time as a mercenary in Persian service, gives detailed instructions for mounted archery in the Hipparchicus. The Achaemenid kataphraktos, the mailed cavalryman, also existed — the funerary stele of a certain Artagerses, c. 350 BCE, shows a fully armoured rider with a kontos-like lance — but was a minority of the force.

The Achaemenid infantry wore little armour by Greek standards, and Greek writers were contemptuous of this. The contempt is a category error. The Achaemenid infantry doctrine was mobility and firepower, not close combat. The standard tactic was the volley of arrows at distance, with the cavalry working the wings and a heavy infantry reserve held back to deliver the decisive blow. This is, mutatis mutandis, the doctrine of every successful pre-modern army that did not have a Greek-style phalanx. The Romans copied it. The Byzantines refined it. The early Islamic armies fought the same way, and a good deal of their tactical vocabulary comes from Sasanian Persian.

The Achaemenid logistical system is what allowed the army to function at the scale it did. The Royal Road from Susa to Sardis, roughly 2,700 km with 111 way-stations at intervals of about 25 km, was supported by a courier service and protected inns. The actual movement of an Achaemenid army on campaign depended on what Xenophon calls the agēlaion — the obligation of the satrapy through which the army was passing to provide a fixed number of days’ supply. The system worked as long as the satraps cooperated; it failed spectacularly in the satraps’ revolt of 360 BCE, when the imperial centre was briefly unable to feed its own armies.

The Achaemenid siege tradition drew on Mesopotamian practice. Cyrus took Sardis in 547 BCE within a few days, allegedly by climbing an unguarded approach to the citadel. The Persian engineers of Babylon in 539 BCE diverted the Euphrates and entered the city by the river gates — a famous operation of antiquity, and almost certainly the basis for Herodotus’ account. The Jewish, Babylonian, and Egyptian specialists who staffed the imperial engineering corps brought traditions of fortification that went back to the third millennium BCE.

The Achaemenid empire was also a maritime power. The Phoenician cities — Tyre, Sidon, Byblos, Arvad — supplied the war fleet, along with Ionian Greek contingents, Cyprus, and Egypt. The decisive contribution of the Persian navy to the Greco-Persian Wars was the loss of Salamis in 480 BCE. The battle was, in effect, a defeat of the Phoenician and Ionian squadrons of the Persian fleet by the Athenian trireme. After Salamis, the Persians stopped projecting naval power into the Aegean at scale.

The Parthian army

Alexander ended the Achaemenid empire in 330 BCE, and the Hellenistic kingdoms that followed — Seleucid, Ptolemaic, Antigonid — inherited the eastern portions of the Persian military tradition in a form modified by Macedonian practice. The Parthians, an Iranian people from the northeast of the plateau, founded by Arsaces I around 247 BCE, built their army on a deliberate rejection of Macedonian-style heavy infantry and a revival of the pure Iranian tradition of massed horse archery. The Parthian spāh (army) was dominated by two units: the cataphract, the mailed lancer, and the asmān-khwār (literally “sky-eater”), the horse archer who delivered the Parthian shot.

The Parthian military system reached its high point in 53 BCE at the Battle of Carrhae. The Parthian general Surena, leading a small but tactically superior force, destroyed the Roman army of Crassus. Plutarch’s account, written more than a century later, gives the Roman force at seven legions — about 40,000 men — and the Parthian horse archers at perhaps 10,000. The numbers are inflated in both directions; modern estimates for the Parthians run from 8,000 to 12,000, and Crassus probably had closer to 35,000 effectives. The disproportion is what matters. Light horse archers, picking their distance, emptied the Roman formation. Then the cataphracts — heavy lancers in scale armour, on armoured horses — finished what the arrows had begun. Crassus died; the aquilae of seven legions were carried east and displayed at Ctesiphon. They were not returned for thirty-three years.

The Roman standards captured at Carrhae were returned to Rome in 20 BCE under Augustus, then re-captured by the Parthians in the reign of Trajan, and finally returned for good under the emperor Phraates V in 62 CE — a thirty-three-year humiliation that became a Roman literary topos. The Parthians never built a large infantry force of their own; they relied on subject peoples in Mesopotamia and Iran to garrison cities and provide logistical support. This structural weakness, combined with the chronic instability of the Arsacid dynasty, eventually allowed the Sasanian revolt in Persis under Ardashir I (c. 220–224 CE) to succeed.

The Sasanian army

The Sasanian Empire, founded by Ardashir I, restored Iranian military power to a level not seen since the Achaemenids. The Sasanian savārān (heavy cavalry) was the supreme mailed force of late antiquity. The Sasanian cataphract — the rider and horse in scale or lamellar armour, armed with a heavy lance, a mace, a sword, and a bow — could shatter Roman infantry in the charge, as it did at the Battle of Callinicum (531 CE) on the Euphrates. The Byzantine general Belisarius was, in fact, defeated at Callinicum by the Sasanian general Azarethes — a useful corrective to the older literature that treated Belisarius as invincible.

The Sasanian period saw four centuries of war with Rome and Byzantium. The high points from the Sasanian side: Shapur I’s capture of the Roman emperor Valerian at the Battle of Edessa in 260 CE — a Roman emperor taken alive by a foreign enemy, a first in Roman history. Khosrow I Anushirvan’s reign in the sixth century, a period of military and administrative reform. The early seventh-century campaigns of Khosrow II, which took the Sasanian army to the Bosphorus and to Jerusalem. The low points: the defeat at Callinicum, the Byzantine counter-offensive of Heraclius in the 620s, the Arab invasions of the 630s. The Sasanian frontier system of castles, watch-posts, and garrison cities from the Caucasus to the Helmand was an exceptionally sophisticated defensive network in the ancient world. The Sasanian logistical system — the aspakān (royal stud farms), the aspbed (master of the horse), the wuzurg-framadar (grand vizier), the dabīrbed (chief secretary), the marzbān (march lord, in charge of a frontier province) — is reconstructed from Pahlavi texts, Roman and Byzantine accounts, and archaeology. The English terms “marshal” and “marzipan” both derive, via Latin and Arabic, from marzbān.

The Sasanian army was multi-ethnic, drawing on Iranian, Mesopotamian, Armenian, Caucasian, and Daylamite troops. The Sasanians maintained a powerful navy in the Persian Gulf, drawing on the ancient maritime traditions of the Arabian and Iranian coasts. The Sasanian fleet controlled the gulf trade with India and east Africa, defended the Iranian coast against the Lakhmid client state and other Arab polities, and projected power as far as the mouth of the Indus. Sasanian naval terminology persisted in the early Islamic fleets of the Umayyad and Abbasid caliphs.

Weapons, armour, and animals

The signature weapon of the Persian mounted warrior, from Achaemenid to Sasanian, was the composite recurve bow — short, double-curved, made of wood, horn, and sinew, capable of penetrating scale armour at close range. The Achaemenid reliefs at Persepolis show the bow in repeated use; the Sasanian rock reliefs at Taq-e Bostan show a heavily caparisoned horse and rider as the central image of royal power. The spear, the akinakes short sword, the dagger, the battle-axe, the mace, and the sling were all standard. Sasanian swords, with their long straight blades and richly decorated hilts, were exported across Eurasia and are the ancestors of the Islamic sayf.

Persian armour mixed scale, mail, plate, and quilted textile. The Sasanian cataphract’s scale-and-lamellar harness set the standard for heavy cavalry equipment in late antiquity. Helmets were bronze and iron. Shields were wicker, wood, leather, and metal. Horse armour of scale and quilted fabric completed the panoply.

Chariots and elephants were used throughout the period. The Achaemenid army maintained a substantial corps of four-horse chariots, organised in squadrons of ten or twenty; the Persepolis reliefs show a complete chariot service, with grooms, farriers, and charioteers. Darius III deployed a force of war elephants at Gaugamela in 331 BCE; the number is unknown — Arrian mentions fifteen, but his number is unreliable. The Sasanians used elephants more sparingly. Camels were standard in the Arabian and Syrian frontier, both as supply animals and for the camel-mounted archers of the desert.

The royal banner — the Derafsh Kaviani, an emblem of Iranian kingship — is discussed in its own article. The Derafsh was not, in the older tradition, a “black banner” (that epithet belongs to the Abbasid revolution). It was an appliquéd leather standard, possibly with a sun or a star, and its loss in battle was a disaster of the first order. The traditional attribution of the Derafsh to the mythical blacksmith Kaveh, who raised it against the tyrant Zahhak, made it a focal point of Persian national memory in the centuries after the Arab conquest.

What was constant, and what changed

Across the three empires, certain features of the Persian way of war were constant, though the proportion of each varied by period.

Combined arms. Infantry, cavalry, chariots, elephants, and engineers acting in concert. The Achaemenid army at Gaugamela in 331 BCE had all of these. The Sasanian army at Yarmouk in 636 CE had all but the elephants.

The strategic reserve. Achaemenid, Parthian, and Sasanian doctrine all emphasised a centrally held reserve that could be committed at the decisive moment. The Greek and Roman sources sometimes mistook this for passivity, and were then surprised when the reserve was committed.

Terrain and manoeuvre. The Persian armies used terrain with care. The feigned retreat — the Parthian shot writ large — was a Persian invention. The enveloping manoeuvre was the Sasanian contribution to the art of war in late antiquity.

Diplomacy and intimidation. The Persians preferred to win without fighting when possible. The Achaemenid policy of installing friendly tyrants in the Ionian Greek cities, the Parthian diplomacy with the Armenians and the Huns, and the Sasanian management of the Lakhmid buffer state in southern Iraq were all examples of using military force as an instrument of a wider political strategy.

What changed most was the proportion of infantry to cavalry. The Achaemenid army was a multi-ethnic coalition with substantial infantry contingents from the subject peoples. The Parthian army was overwhelmingly cavalry. The Sasanian army was a compromise: heavy cavalry as the striking force, infantry as support, and a sophisticated logistical system that combined elements of both. The shift from infantry-heavy to cavalry-heavy and back is the single largest internal change in the Persian military tradition across the period, and it is not yet adequately explained.

What we don’t know

A word of caution, because the subject invites overconfidence. The Greek and Latin sources on the Achaemenid army are abundant but interested. The Parthian sources, in Parthian language, are very few: a handful of ostraca from Nisa, the Greek and Roman accounts, and a small number of coins and seals. The Sasanian sources are more numerous but biased: the Sasanian rock reliefs are propagandistic, the Pahlavi texts are mostly religious, the Roman accounts are from the other side, and the archaeological record is uneven. Reconstruction is the best we can do, and the reconstructions are improving as the Persepolis Fortification Tablets, the Nisa documents, and the recent French and Iranian excavations at Susa, Pasargadae, and Bishapur are published.

What can be said with some confidence: the Persian military tradition, from Cyrus to Khosrow, was a coherent and adaptive system that was studied, copied, and adapted by its enemies. The 1,300-year run from the fall of Nineveh in 612 BCE to the fall of Ctesiphon in 637 CE is, by any measure, one of the long runs of military history. Few traditions match it. The Persian military tradition did not end in 651 CE with the death of Yazdegerd III. The Sasanian general Hormozan continued resistance in Khuzistan for several years after the fall of Ctesiphon. The Iranian traditions of heavy cavalry, mounted archery, and combined-arms warfare passed into the early Islamic armies, particularly those of Khorasan, and the Persian military vocabulary survived in Arabic, Turkish, and Mongol armies for many centuries. The Buyid, Samanid, and Seljuk dynasties that revived Persian power in the medieval Islamic world did so in part by re-creating the Persian model of a multi-ethnic empire organised around heavy cavalry and the loyalty of the Iranian heartland. The Mongol conquest of the Khwarezmian Empire in the 1220s is the end of the line, in a strict sense.

What remains genuinely unresolved is the question of how much of the Achaemenid, Parthian, and Sasanian military systems is continuous and how much is reinvented. The Parthians clearly saw themselves as restoring an Achaemenid-style cavalry tradition after the Hellenistic interlude. The Sasanians clearly saw themselves as restoring the Achaemenid empire, and Ardashir’s rock reliefs at Naqsh-e Rostam and Firuzabad deliberately echo the Achaemenid iconography. Whether the tactical and logistical substance matched the iconography is, in my view, less certain. The answer will depend on the next round of publications from the Persepolis, Nisa, and Bishapur excavations, and on the next round of work on the Parthian and Sasanian ostraca. We are not there yet.

For further reading

Briant, Pierre. From Cyrus to Alexander: A History of the Persian Empire (2002)
Wiesehöfer, Josef. Ancient Persia (1996)
Brosius, Maria. The Persian Empire from Cyrus II to Artaxerxes I (2020)
Bivar, A. D. H. “The Political History of Iran under the Arsacids” in The Cambridge History of Iran, vol. III (1983)
Pourshariati, Parvaneh. Decline and Fall of the Sasanian Empire (2008)
Daryaee, Touraj. Sasanian Persia (2009)
Farrokh, Kaveh. Shadows in the Desert (2007)
Pritchett, W. Kendrick. The Greek State at War, vol. V (1991)
Tuplin, Christopher, ed. Xenophon and his World (2004) — several relevant chapters on the Achaemenid military
Dusinberre, Elspeth. Empire, Authority, and Autonomy in Achaemenid Anatolia (2013) — for the western satrapies