History of the Persian Empire: From the Achaemenids to the Sasanians
What the sources actually say, in the autumn of 539 BCE, is this: the garrison of Babylon opened the gates. The Babylonian Chronicle tersely notes that “the soldiers of Cyrus entered Babylon without fighting.” The Cyrus Cylinder, found in the ruins of the Esagila by Hormuzd Rassam in 1879 and now in the British Museum under the catalogue number BM 90920, presents the entry as liberation — Marduk, the cylinder says, “scanned and looked” through all the countries for a just king, and chose Cyrus, an Iranian from the small kingdom of Anshan, who within a generation had assembled an empire stretching from the Aegean to the Hindu Kush.
The older reading of the cylinder as “the first declaration of human rights” is a construction of the 20th century, popularised in Iran under the Pahlavis and picked up by the United Nations in 1971 when a replica was displayed. Irving Finkel’s 2013 re-reading of the cuneiform is the modern philological position: it is a Babylonian royal inscription in the tradition of Nabonidus’s own self-presentation, and Finkel has argued that the text in our hands is itself a later copy, perhaps from the reign of Darius. The dating matters less than the propaganda claim, which is recognisably Mesopotamian: foreign conquest could be legitimate if the conquered god had willed it. The novelty was the conqueror.
Three empires followed, each of which deserves to be read on its own terms. The Achaemenids (c. 550–330 BCE) built the largest state the ancient world had then seen and lost every major war they fought with Greece. The Parthian Arsacids (247 BCE–224 CE) preserved the Iranian imperial idea for nearly five centuries, mostly in the shadow of Roman power. The Sasanians (224–651 CE) restored Iranian centralised monarchy, fought four centuries of war with Rome and Byzantium, codified Zoroastrian law, and bequeathed a state template that the caliphs of Islam would adopt within a generation of the Arab conquest. Each is treated in its own pillar; what follows is the long arc.
A geography that dictated the politics
The Iranian Plateau is roughly 2,500 km from west to east and 1,500 km from north to south — a block of mountains and high desert ringed by the Zagros, the Alborz, the Hindu Kush, and the coastal ranges of the Persian Gulf. The western edge drops into the Tigris-Euphrates lowland; the eastern edge fades into the Helmand and the Bactrian plain. Rain-fed agriculture is possible only in a narrow western strip and in the Caspian lowlands; the rest of the plateau depends on snow-melt, qanats (underground water channels), and oases.
The political consequence is that whoever controlled the western lowland and the Caspian gates controlled the population and the tax base. The Median kingdom of the eighth and seventh centuries BCE did this. The Achaemenids did it more thoroughly. The Seleucids tried and failed to hold it. The Parthians held it loosely. The Sasanians reasserted it. The whole political history of the empires is, in one reading, a story of varying degrees of control over this geography. The Iranian languages — Old Persian in the south-west, Median in the north-west, Avestan in the north-east — had been spoken on the plateau since the late second millennium BCE. Gernot Windfuhr’s The Iranian Languages (2009) is the modern reference. The transition from Old Persian (cuneiform, Achaemenid) through Middle Persian (Pahlavi script, Sasanian) to New Persian (Arabic script, post-Islamic) is the linguistic spine of the empires.
The Achaemenid achievement
Pierre Briant’s From Cyrus to Alexander (2002) is the indispensable English synthesis. Josef Wiesehöfer’s Ancient Persia (1996) is the closest competitor, broader in scope and slightly older. Maria Brosius’s The Persian Empire from Cyrus II to Artaxerxes I (2020) is the most reliable single-volume account with attention to the Babylonian evidence. Amélie Kuhrt’s The Persian Empire: A Corpus of Sources from the Achaemenid Period (2007) is the indispensable source collection. Matthew Stolper’s work on the Persepolis Fortification Tablets, the 30,000-odd administrative documents published in fits and starts since the 1930s, has done more than any other body of scholarship to put numbers on the Achaemenid economy. Margaret Cool Root’s The King and Kingship in Achaemenid Art (1979) remains the indispensable treatment of the visual evidence.
A few specific claims deserve attention, because the older popular literature is full of them and the modern scholarship is more cautious.
The Cyrus “tomb” at Pasargadae is a small stone chamber on a six-stepped platform, roughly 11.1 by 7.2 metres in plan, with a steeply pitched gable. It is austere in a way that the Achaemenid reliefs at Persepolis are not. Its function has been debated since Alexander the Great reportedly visited it and had it restored. The inscription on the platform, “Passer-by, I am Cyrus, who gave the Persians an empire,” is generally thought to be Hellenistic or later.
The “nineteen battles in a single year” of Darius I, after the death of the usurper Gaumata in 522 BCE, comes from Darius’s own Behistun Inscription, a trilingual text in Old Persian, Elamite, and Babylonian carved about 100 m up the cliff at Behistun in western Iran. Henry Rawlinson’s copies of the 1830s–1840s were the key to the decipherment of cuneiform. The “nineteen battles” is Darius’s own count. The Behistun relief shows him with his foot on the chest of Gaumata, with nine bound rebel kings in front of him and the winged figure of Ahura Mazda above. The location, the iconography, and the trilingual text all served the same purpose: to anchor the legitimacy of the new branch of the Achaemenid house in cosmic order.
The Achaemenid empire was, in territorial extent, the largest state the world had then seen: from the Balkans and the Nile to the Indus, from the Caspian to the Persian Gulf. It governed a population of perhaps 50 million. It lasted 220 years. It did so without the cultural or linguistic assimilation that Roman citizenship eventually required, and it did so while losing every major war it fought with Greece. The puzzle is worth a moment. My own view is that the standard answer — religious tolerance, satrapal administration, light centre, heavy local elites — is roughly right, but undersells the role of the farr and the political theology of kingship. The Achaemenid state worked because the bandaka, the subject of the King of Kings, was a legal status that applied to all free inhabitants regardless of ethnicity, language, or religion, and because rebellion was a religious as well as a political crime.
The fall of the Achaemenid state to Alexander between 334 and 330 BCE is treated in the Achaemenid pillar. What is worth saying here is that the Achaemenid state did not collapse under the weight of internal corruption, as older scholarship had it, but was defeated by a single new military power in the northern Aegean. Darius III was not incompetent. He was out-generalled.
The Hellenistic interlude and the rise of Parthia
Alexander’s conquest was followed by a century and a half of Seleucid rule over the eastern portion of the empire. The Seleucids were Macedonian in language and court culture; the Iranian population they ruled was, by their own account, restive. The Parthian state emerged from a revolt led by Arsaces I, a chieftain of the Parni (an Iranian people of the north-eastern steppe), in or around 247 BCE. A. D. H. Bivar’s chapter in the Cambridge History of Iran, vol. III (1983) is still the most thorough treatment in English; V. Sarkhosh Curtis’s The Age of the Parthians (2007) is a useful counterweight. Josef Wiesehöfer’s Iraniens, Grecs et Romains (2001) is excellent on the east–west cultural negotiation of the period.
Mithridates I (c. 171–138 BCE) is the first great Arsacid ruler. He conquered Media, Mesopotamia, and Elymais, entered Seleucia-on-the-Tigris in 141 BCE, and reduced the Seleucid kingdom to a rump in Syria. Mithridates II (c. 124–88 BCE) consolidated the empire, signed the first formal treaty with Rome in 92 BCE, and adopted the title shāhān shāh (King of Kings), the Old Persian xšāyaθiya xšāyaθiyānām that the Achaemenids had used. The shāhān shāh was a Persian title of legitimacy that would outlast every dynasty that used it, from the Sasanians to the Qajar shahs of the 20th century.
The Parthian state is hard to characterise, because the sources are thin. The Parthian-language documents are mostly ostraca from the site of Nisa (in modern Turkmenistan), the Greek and Latin sources are interested, and the archaeological record is uneven. Justin Wickersham’s recent work on the Parthian documents and the Parthian Sources Project, a digital edition of the Greek and Latin texts, has begun to make the field more navigable. The standard judgment is that the Parthian state was a federation of kingdoms, with a weak centre and powerful noble houses — the Suren and the Karen are the best attested — whose autonomy grew as the dynasty aged. The Parthian pillar discusses the administrative structure in more detail.
The Parthians are best remembered, in Western memory, for the Battle of Carrhae in 53 BCE, in which the Roman triumvir Crassus was killed and the aquilae of seven legions were captured and carried east. Plutarch’s account, written more than a century later, is the canonical source. The standards were returned to Augustus in 20 BCE, recaptured under Trajan after his Mesopotamian campaigns, and returned for good under Phraates V in 62 CE. The Romans never quite got over the humiliation.
The Sasanian century
The Sasanian revolution was, in significant measure, a Parthian restoration with new ideological content. Ardashir I, a minor prince of Persis, defeated and killed the last Arsacid king Artabanus IV at the Battle of Hormizdagan in 224 CE, and his son Shapur I, victor of three wars with Rome and capturer of the emperor Valerian at the Battle of Edessa in 260 CE, established the dynasty on a footing the Parthians had never managed.
The principal modern syntheses are Touraj Daryaee’s Sasanian Persia: The Rise and Fall of an Empire (2009), Matthew P. Canepa’s The Two Eyes of the Earth: Art and Ritual of Kingship between Rome and Sasanian Iran (2009), and the relevant volumes of the Cambridge History of Iran (Yarshater, 1983–). Parvaneh Pourshariati’s Decline and Fall of the Sasanian Empire (2008) overturned the older narrative by demonstrating the continuing political importance of the great noble families and the inadequacy of the older “Sasanian–Parthian conflict” model. James Howard-Johnston’s The Last Great War of Antiquity (2021) is the definitive account of the final Sasanian-Byzantine war of 602–628 CE.
The Sasanian state was more centralised than the Parthian, more aggressively Zoroastrian, and more consistently at war with Rome. Shapur I’s triple war of 241–244, 256–260, and 283–293 CE; Khosrow I Anushirvan’s reign of 531–579 CE, with its administrative reforms and its capture of Antioch in 540 CE; and Khosrow II’s campaigns of the early seventh century, which took the Sasanian army to the Bosphorus and to Jerusalem in 614 CE, are the high points. The Sasanian “Spring of Khosrow” carpet is a literary topos, attested in Arabic and Persian sources, not a surviving object.
A note on the Sasanian calendar, because the popular literature on the subject is full of errors. The Sasanian year was 360 days long, divided into twelve 30-day months, with five extra gāhānbār days appended at year’s end to bring the count to 365. The system was inherited from Achaemenid Mesopotamia and refined under Khosrow I. It was not a six-day week. The 30-day month and the 360-day year are the basic structure, and the five extra days are the recalibration. This calendar, transmitted through the Shahnameh and the Pahlavi astronomical tradition, was the ancestor of the modern Iranian solar calendar (adopted in 1925) and was refined in later centuries by astronomers who were not “Sasanian” — al-Khwarizmi (d. c. 850) and Omar Khayyam (d. 1131) are the two most famous, but they worked under Islamic, not Sasanian, patronage.
The fall of the Sasanian state to the Arab armies of the Rashidun Caliphate in the 630s and 640s is a separate subject. The decisive battle was al-Qadisiyyah (636 CE) on the Iraq-Iran border, followed by the sack of Ctesiphon in 637 CE and the death of the last Sasanian king, Yazdegerd III, near Merv in 651 CE. Howard-Johnston argues, against the older view, that the Sasanian state was not in a state of internal collapse when the Arabs arrived but was exhausted by the final war with Byzantium (602–628 CE) and unable to mobilise.
The Persianate aftermath
The Arab conquest did not erase Persian identity; it transformed it. The new Islamic ruling class adopted Sasanian court ceremony, bureaucratic practice, and royal titulature within a generation. Persian, written in the Arabic script, became one of the great literary languages of the Islamic world. The Tahirid, Samanid, Buyid, Ghaznavid, Seljuk, Ilkhanid, Timurid, and Safavid dynasties of the Islamic period traced their legitimacy to the Sasanian past, and the Sasanian model of kingship — religiously grounded, militarily aggressive, administratively centralised — became the template for Islamic imperial rule.
The Sasanian title shāhān shāh is one of the more durable items in the political vocabulary. It passed, via the Sasanian model, into the caliphal vocabulary; the Buyids used it in the 10th century, the Seljuks in the 11th, and the Safavids in the 16th–18th centuries. The last official user of the title in Iran was Mohammad Reza Shah Pahlavi, deposed in 1979.
The Persian language itself is the most legible continuity. Ferdowsi’s Shahnameh (c. 1010 CE) is essentially a long meditation on the meaning of Persian kingship across the Achaemenid, Parthian, and Sasanian periods, drawing on the lost Sasanian Khwaday-namag and on oral tradition. The classical Persian poets — Rumi (d. 1273), Hafez (d. 1390), Saadi (d. 1291), Nizami (d. 1209), Jami (d. 1492) — inherited a poetic vocabulary and a set of images that went back to the Achaemenid and Sasanian periods. Their works, in turn, became a foundational element of the Ottoman, Mughal, and Central Asian literary traditions.
The concept of Iranshahr — the Iranian realm, defined by language, religion, and political tradition — is the political form this continuity took. After the Arab conquest, the concept did not disappear but transformed. The Persianate world at its height, in the 14th and 15th centuries, stretched from the Balkans to Bengal, from the steppes of Central Asia to the ports of the Indian Ocean. The Mughal Empire (1526–1857) is the largest and longest-lived of the post-Sasanian Persianate states, and the Tajik, Hazara, and Uyghur communities of Central Asia are the demographic heirs of the eastern Iranian world. The Parsis of India, descended from Zoroastrian Persians who fled to Gujarat in the 8th and 9th centuries, are a distinctive diaspora community.
Specific contributions, briefly named
The Persian empires were, in the first instance, military and administrative organisations. They were also, in the longer run, civilisational projects whose ideas — about kingship, about religious tolerance, about the relationship between a centralised state and a multi-ethnic population, about the management of a continental economy — became part of the inheritance of Eurasia.
The Achaemenid model of a multi-ethnic, religiously tolerant, administratively sophisticated state influenced the Hellenistic kingdoms, the Roman Empire, the caliphs of Islam, the Byzantine emperors, the Mughal shahs of India, and the constitutional arrangements of modern states. Briant’s argument, in the second volume of his history, that the Achaemenid state was a more credible model for the modern world than the Hellenistic city-state is now widely accepted. The Sasanian model of a religiously grounded monarchy with a strong bureaucracy and a learned class serving as a bridge between court and people is the ancestor of the Islamic imperial tradition from the Abbasids to the Safavids.
The Sasanian Academy of Gondeshapur, traditionally founded under Shapur I in the third century CE, brought together Greek, Indian, Syriac, and Persian scholars in a programme of translation and research. Greek medical texts by Hippocrates and Galen were rendered into Syriac and Pahlavi; Indian astronomical, mathematical, and medical works were translated into Middle Persian. The Barmakid family — originally hereditary custodians of a Buddhist monastery in Balkh, not “Buddhist administrators” in the loose sense — patronised the great translation movement of the early Abbasid caliphate in Baghdad. The Canon of Medicine of Avicenna (Ibn Sina, d. 1037), a systematised and extended version of the Galenic tradition as transmitted through the Sasanian academies, was a standard medical textbook in European universities until the 17th century.
The Sasanian rock reliefs at Naqsh-e Rostam, Bishapur, and Taq-e Bostan, the Sasanian silver plates, and the Sasanian silk designs became a fashion across Eurasia. The Sasanian silk designs were copied in Byzantium and Tang China; the Sasanian imperial crown, with its distinctive globular form, influenced the crowns of medieval Europe and the Russian tsars. The Sasanian tradition of monumental rock-cut art, in which the king is shown in investiture by Ahura Mazda, is a distinctive contribution to late antique visual culture whose iconographic influence can be traced into Byzantine, Carolingian, and Islamic royal art.
The Persian calendar, the Persian garden, the Persian festival of Nowruz, and the Persian tradition of hospitality are the elements of material culture that have proved most durable. The modern Iranian solar calendar, adopted in 1925, is a direct descendant of the Sasanian system. The four-fold garden, the chahar bagh, with its water channels and its four quadrants, is a Persian invention of the Achaemenid period that was transmitted through the Sasanian gardens at Ctesiphon, the Islamic gardens of Isfahan and Shiraz, and the Mughal gardens of India. Nowruz, the spring equinox festival, is celebrated by hundreds of millions of people from Iran to Tajikistan, from Afghanistan to the Kurdish regions of Iraq, from Azerbaijan to the diaspora communities of Los Angeles and Toronto.
A note on the evidence
The Achaemenid sources are good by the standards of the ancient Near East: the royal inscriptions, the Persepolis Fortification Tablets, the Babylonian chronicles, and the Greek historians Herodotus, Thucydides, and Xenophon, who knew the Achaemenid court personally or by reputation. The Parthian sources are thin. The Sasanian sources are more numerous — the rock reliefs, the Pahlavi texts (Bundahishn, Denkard, Khwaday-namag in later recensions, Arda Viraf Namak, Zand-i Wahman Yasn, the Kar-Namag i Ardashir), the Roman and Byzantine accounts from the other side — but biased, and the Pahlavi texts are mostly religious, not historical. The Sasanian “archives” of administrative documents that one sometimes reads about do not exist in the way that the Roman papyri or the Persepolis tablets do.
For the Sasanian period, the Arabic and Syriac sources are often more informative than the Pahlavi ones, because the Arab and Syriac writers were not bound by the Zoroastrian theological framework that shapes the Pahlavi texts. Tabari (d. 923), the great Persian historian writing in Arabic, is the indispensable source for the late Sasanian and early Islamic periods; the Chronicle of Seert, a Nestorian Christian compilation in Arabic, is excellent on the Sasanian relationship with the Church of the East. Procopius, Agathias, and Menander Protector are the best Greek sources for the Sasanian-Byzantine wars of the sixth and seventh centuries. Ammianus Marcellinus, writing in the fourth century, is the best Latin source for the Sasanian state in its late antique form.
The reconstruction is improving. The Persepolis Fortification Tablets continue to be published, in fits and starts, by Stolper and his successors. The French-Iranian excavations at Susa, Bishapur, and Pasargadae have transformed the archaeological picture. The Nisa documents are being re-edited. The Pahlavi text tradition is being reassessed. But the reader who wants certainty will not find it, and the reader who pretends to certainty is not to be trusted.
The duration
The 1,100-year run from Cyrus to Yazdegerd III is, by any measure, a long run. The 1,400-year run from the Behistun Inscription of Darius to the fall of Ctesiphon in 637 CE is longer. Few ancient states survived as long. The Roman Empire is the only obvious comparison, and the Roman case is more often invoked than the Persian case because the Roman sources are richer and the Roman legacy in European political vocabulary is more direct.
The reason the Persian story is worth telling in detail is that the Persian empires did something that the Roman Empire did not. They built a continental state, on a Eurasian scale, without the cultural and linguistic assimilation that Roman citizenship required. These are the early versions of a political idea that the modern world is still trying to make work.
The Persian empires are also worth telling because they are part of the modern conversation. The Cyrus Cylinder is displayed at the United Nations. The tombs of Cyrus and Darius are protected as national heritage in modern Iran. The Persepolis reliefs appear on Iranian banknotes, passports, and official documents. The Persian New Year, Nowruz, is a moment of national and transnational solidarity that connects Iran to the wider Persianate world. The recovery of the Persian past is, in other words, not only a scholarly enterprise but also a living political and cultural process.