Parthian Empire

The Parni

Drafted from Overtoom, Oxford Studies in Early Empires (2020)formation

Around 247 BC a people called the Parni, from the steppe east of the Caspian, moved into the Seleucid satrapy of Parthia and took it. Their leader was called Arsaces, and every subsequent king of the dynasty took the name as a title, which is why the Greek and Roman sources call them all Arsaces and why the chronology is difficult.

Overtoom's book is the first full study of this period since 1938 and its argument is a reframing worth carrying. The Parthians are conventionally introduced as Rome's eastern enemy, which is where the surviving sources put them and which begins two centuries into their own history. What they actually were for their first hundred and fifty years is a Hellenistic power among Hellenistic powers — a state that emerged in the wreckage of Seleucid authority alongside Bactria, Pergamon, Pontus and the rest, competed in the same system, used the same diplomatic vocabulary, and won.

The circumstances made it possible. The Seleucid empire on the previous record was fighting the Ptolemies over Coele-Syria and had a satrap in Bactria who had already stopped acknowledging the king; the eastern provinces were four months from Antioch, and the answer to a lapsed province was an expedition when one could be spared. Antiochus III spared one, conducted the great anabasis between 212 and 205, obtained the Parthian king's submission, and left him where he was.

That is the pattern of the first century: the Parthians take a province, the Seleucids come, the Parthians submit, the Seleucids leave, the submission lapses. It is not a war of conquest and it is barely a war. It is a slow transfer of authority conducted through repeated acknowledgements of a suzerain who cannot stay.

Then in the 140s Mithridates I stopped submitting and took Media, and then Babylonia, and the arrangement ended.

Taking Iran, and meeting two empires

Drafted from Overtoom, Oxford Studies in Early Empires (2020)expansion

Mithridates I took Media in about 148 and Seleucia on the Tigris in 141, and the second of those is the one that mattered. Babylonia is the richest agricultural country between the Nile and the Indus, and taking it converted a dynasty holding a plateau into the government of an empire.

The Seleucids tried twice to get it back. Demetrius II invaded, was defeated and captured, and spent ten years as an honoured prisoner married to a Parthian princess. Antiochus VII invaded in 130 with a large army, won two battles, wintered his troops dispersed across Media, and was killed with most of them the following spring. That is the end of Seleucid Iran.

Mithridates II, who ruled from about 124 to 88, is where the empire reaches its height. He put down the nomadic incursions on the eastern frontier that had killed two of his predecessors, took Mesopotamia as far as the Euphrates, imposed Parthian suzerainty on Armenia, and took the title king of kings. That was the Achaemenid style and it was claimed deliberately: a Parni chief's descendant announcing himself the heir of Darius.

He also met both of the other great powers in the world within a few years of each other, and both encounters can be dated.

In about 121 an embassy from Han China reached him. Zhang Qian's missions west had opened the route, and the Chinese sources record the Parthians — whom they call Anxi, after Arsaces — receiving the envoys on the frontier with twenty thousand cavalry as a display, and sending an embassy back with ostrich eggs and conjurors.

In about 96 a Parthian envoy met Sulla on the Euphrates, and Rome and Parthia were formally in contact for the first time. Within forty-five years Crassus was dead at Carrhae.

Sitting between those two facts is how the empire made its money. Parthia was the land route between Rome and China for three centuries. It took a share of everything that crossed, and — as the Chinese sources complain — took care that the two ends never dealt with each other directly.

The empire Rome could not finish

Drafted from (ed.), The Cambridge History of Iran, Volume 3: The Seleucid, Parthian and Sasanian Periods (1983)peak

The Arsacids began as a nomadic group that took the Seleucid satrapy of Parthia around 247 BC and spent the next century working west, taking Media and then Mesopotamia. They held Iran for four hundred and seventy years, longer than the Sasanians who replaced them and longer than any other dynasty to rule the Iranian plateau.

They are also the worst documented. The Parthians left very little writing of their own; almost everything comes from Greek and Roman sources describing an enemy, and from coins. That imbalance produced a long scholarly habit of treating them as a loosely organised interlude between two real Iranian empires, which their four and a half centuries and their record against Rome do not support.

That record: Crassus destroyed at Carrhae in 53 BC with seven legions lost, Mark Antony's invasion beaten in 36 BC, and a frontier on the Euphrates that Rome crossed repeatedly and never held. Trajan reached the Persian Gulf in 116 and the gains were abandoned within two years.

The empire was governed loosely — great houses with their own armies, vassal kingdoms, Greek cities keeping their institutions — and that decentralisation is both why it lasted and why Ardashir, a vassal in Fars, could kill Artabanus IV in 224 and take it.

Carrhae, and the shape of a stalemate

Drafted from (ed.), The Cambridge History of Iran, Volume 3: The Seleucid, Parthian and Sasanian Periods (1983)peak

In 53 BC Marcus Licinius Crassus took seven legions across the Euphrates and was destroyed at Carrhae by a Parthian force perhaps a fifth the size. The Roman infantry could not close with horse archers who withdrew as they shot, and could not withstand the armoured lancers who charged when the formation tightened against the arrows. Crassus was killed; the legionary standards were kept in Parthian hands for thirty-three years, and their return by negotiation in 20 BC was advertised in Rome as a triumph because it was the best available outcome.

The battle set a pattern that held for two and a half centuries, and it is the reason the frontier between the two empires barely moved in all that time. Rome could invade Mesopotamia, and did, repeatedly — Trajan reached the Persian Gulf in 116. It could not hold what it took, because supply lines across northern Mesopotamia were vulnerable to exactly the cavalry that had beaten Crassus. Parthia could raid Syria and could not take fortified cities in numbers.

Two empires that could hurt each other and could not finish each other produced something neither intended: a stable frontier, a hostage exchange and embassy protocol, and a trade route across it that carried Chinese silk to Rome through Parthian middlemen who had every incentive to keep both ends ignorant of each other. When a Han envoy reached the Persian Gulf in 97 and considered going on to Rome by sea, the people who talked him out of it were the people whose living depended on the two ends never meeting.

An empire of magnates

Drafted from Overtoom, Oxford Studies in Early Empires (2020)aside?What aside means

There was no Parthian bureaucracy in the Achaemenid or Sasanian sense. There was a king of kings, over an assortment of things.

Great noble houses — the Suren, the Karen, the Mihran — holding hereditary territories and hereditary offices with their own armed followings. Vassal kingdoms in Persis, Elymais, Characene, Adiabene, Media Atropatene and Armenia, each with its own dynasty and its own coinage. Greek cities keeping their assemblies and their civic institutions. Nomadic groups on the eastern frontier under their own chiefs.

The army was the sum of what those parties supplied, and the most famous Parthian victory in history was won by a force raised and commanded by the head of one noble house: Surena at Carrhae, with cataphracts, horse archers and a baggage train of arrows. The king had him executed shortly afterwards, which tells you exactly what the relationship was.

What that costs, and what it buys

An army that can make a ruler will eventually make one it prefers. The Parthian version of that is chronic rather than acute. Arsacid civil wars are more or less continuous from the first century BC, with rival kings coining simultaneously and the great houses backing candidates on their own account. Rome repeatedly sponsored claimants, and the frontier wars are frequently interventions in a succession rather than campaigns of conquest.

It is also why the empire lasted four hundred and seventy years.

A state that is a coalition cannot be destroyed by defeating its king, because the coalition survives him. Rome demonstrated this three times, by taking Ctesiphon and finding that nothing followed. Trajan in 116, Lucius Verus's generals in 165, Septimius Severus in 198: three sacks of the capital, and no conquest.

The end came from inside the structure, which is the only place it could have come from. Ardashir was a sub-king in Persis, one of the vassal dynasties, and in 224 he killed Artabanus IV in battle and took the empire. The coalition did not fail; it changed which of its members was in charge of it.

Erased by their successors

Drafted from (ed.), The Cambridge History of Iran, Volume 3: The Seleucid, Parthian and Sasanian Periods (1983)aside?What aside means

The Achaemenids were described almost entirely by their enemies. Alexander was described by nobody who was there. The Parthians have both of those problems and a third that is worse than either.

They left very little writing. There is no Parthian history, no chronicle, no court literature that survives; what exists is coins in enormous quantity, parchments and ostraca from Nisa and Dura-Europos, inscriptions, and archaeology. Everything narrative comes from Greek and Roman writers describing an enemy across a frontier, or from Chinese envoys' reports, or from Armenian and Jewish sources with their own interests.

The third problem is that their successors removed them.

The Sasanians, who took the empire in 224, constructed a royal ideology that ran from the mythical Kayanian kings of the Avestan tradition directly to themselves, and the Parthians did not appear in it as an empire. The Iranian national tradition that comes down through the Shahnama treats the four hundred and seventy years between Alexander and Ardashir as the age of the muluk al-tawa'if — the party kings, the petty kings — a period of division with no central sovereign worth naming. Firdawsi disposes of it in a few hundred lines and says explicitly that there is little to tell.

Four and a half centuries — longer than the Sasanians who replaced them, longer than the Achaemenids they claimed descent from, longer than anybody else has ever governed Iran — compressed into a gap in the national story.

The habit persisted in modern scholarship, which for a long time treated the Parthians as a loose interlude between two real Iranian empires — feudal, unadministered, culturally derivative. Overtoom's book belongs to a body of work arguing against that, on the grounds that the judgement was not arrived at independently: it was inherited from the people who overthrew them, and repeated for seventeen hundred years by scholars reading sources the Sasanians had curated.

That is worth pausing on, because it is a rare thing to be able to watch. Almost every dynasty invents a claim to a predecessor; the Parthians did it themselves, to the Achaemenids. What happened here is the opposite operation and it is much harder to detect afterwards. A regime did not merely fail to claim its predecessor — it removed him from the account of the country, replaced him with a period of anarchy, and the substitution held so well that it was still the scholarly consensus in the twentieth century.

Twice taken

Drafted from Dio, Dio's Roman History (1906)contraction

Ctesiphon was captured by Roman emperors in 116 and again in 198, and on both occasions the Romans went home. Cassius Dio, who was a senator under the second of them, is unusually direct about what that meant, and the answer is not what either triumph claimed.

Trajan, 116

Trajan crossed the Tigris and entered Ctesiphon, was saluted imperator, established his right to the title Parthicus, and was voted by the senate, among other honours, the privilege of celebrating as many triumphs as he might desire.

Then he went down to the sea, because he wanted to. Mesene, the island kingdom in the Tigris, came over without difficulty. He was nearly wrecked by a storm and the backward flow from the ocean. The people of the palisade of Spasinus received him kindly.

And at the shore of the Persian Gulf he watched a ship setting out for India, and said — Dio gives it as direct speech —

I should certainly have crossed over to the Indi, if I were still young.

He was sixty-three. He gave much thought to the Indians and was curious about their affairs, and within two years the provinces he had just annexed had revolted and he was dead, and his successor gave them all back.

Severus, 198

Eighty years later Septimius Severus came down the Euphrates on newly built boats — the timber was abundant along the river — and found Seleucia and Babylon abandoned, which is its own comment on where the population of Mesopotamia had gone.

Then Ctesiphon. He took it, turned his soldiers loose on the whole town, and Dio records the result plainly: a great slaughter of men, and nearly a hundred thousand taken alive.

And then the sentence that is the point of the chapter:

However, he did not pursue Vologaesus nor yet occupy Ctesiphon, but as if the sole purpose of his campaign had been to plunder it, he thereupon departed.

Dio gives the reasons without softening them: ignorance of the country, and lack of provisions. The return had to be made by a different route because the wood and fodder on the first one were used up.

Hatra

Immediately afterwards Severus attacked Hatra, a caravan city in the Mesopotamian steppe, and accomplished nothing. His siege engines were burned, many soldiers died, and vast numbers were wounded. He withdrew.

The emperor who had just sacked the Parthian capital could not take a fortified Arab town a few days' march away.

What it did to Parthia

The obvious reading is that Parthia was being beaten. The more useful one is that both empires were discovering the same thing about this frontier.

Rome could reach Ctesiphon whenever it committed an emperor and a field army to doing so. Rome could not hold anything past the Euphrates, because the logistics were beyond it — Dio says so about Severus in as many words, and Trajan's annexations were abandoned within two years of his death.

Parthia, for its part, survived both sacks and neither of them killed the dynasty. But the capital had been taken twice in a century, a hundred thousand of its people had been carried off in one afternoon, Seleucia and Babylon were empty, and the vassal kingdoms had watched the king of kings fail to defend the centre.

What destroyed the Arsacids twenty-six years after Severus left was not a Roman army. It was a vassal from Persis who had drawn the obvious conclusion.

A sub-king in Persis

Drafted from Daryaee, Sasanian Persia: The Rise and Fall of an Empire (2009)end

The second century was, on the Roman side, a century of victories. Trajan took Ctesiphon in 116; Lucius Verus' generals sacked it in 165 and brought a plague back that killed a substantial part of the Roman empire's population; Septimius Severus sacked it again in 198 and reportedly took a hundred thousand captives. Rome annexed northern Mesopotamia and kept it.

None of it destroyed the Parthian state, and all of it damaged the Arsacids' standing. A king of kings who cannot keep his capital is a king of kings the great houses and the vassal dynasties have reason to reconsider, and the civil wars of the early third century were unusually severe — Vologases VI and Artabanus IV coining simultaneously and fighting each other while a Roman emperor was campaigning in the country.

Into that came Ardashir, sub-king of Persis: one of the vassal dynasties on which the whole structure rested, ruling the province the Achaemenids had come from, with its own coinage, its own fire temples and a family claim going back to a priest called Sasan. He took the neighbouring kingdoms, refused tribute, and in 224 met Artabanus IV in battle at Hormozdgan and killed him.

It was a replacement of one dynasty by another and nothing more than that. There was no foreign conquest and no collapse. One of the empire's own component kingdoms took the throne, kept Ctesiphon as the capital, kept the title king of kings, kept the frontier on the Euphrates, and kept fighting Rome for another four centuries.

What changed was everything about how it was held. The Sasanians centralised what the Arsacids had distributed: a state church, a land survey, a fixed tax, a paid army, a bureaucracy. It is the more impressive machine by a distance, and it lasted two hundred years less.

The head at the banquet

Drafted from Plutarch, Lives of the Noble Grecians and Romans (1859)afterlife

The Parthians were forgotten in Iran and remembered in Europe, and what Europe remembered was one afternoon in 53 BC and its aftermath. Plutarch's account of that aftermath has been retold for two thousand years as an atrocity story. Read slowly, it is evidence of something quite different.

The triumph at Seleucia

Surena, who had won the battle, staged a mock triumph in Seleucia. He dressed a prisoner named Caius Paccianus, chosen because he looked most like Crassus, in a woman's dress of the barbarian fashion, sat him on a horse, and had him answer to the name of Crassus and to the title Imperator. Trumpeters and lictors went in front on camels. Purses hung from the ends of the rods, and the heads of the Roman dead were fixed on the axes. Behind came the singing women of Seleucia with obscene songs about Crassus's cowardice.

Then Surena called the city's senate together and produced the pornography. Certain wanton books by Aristides of Miletus had been found in the baggage of a Roman officer, and Surena laid them before the assembly as a comment on the sort of people who could not leave such things behind even on campaign.

Plutarch, who disapproves of everybody, notes that the Seleucians were entitled to think of Aesop's fable of the wallet, since the general carrying a bag of Milesian stories in front of him had a whole train of wagons of concubines and lutes and castanets behind him — and adds that the Parthians had forgotten, while mocking Milesian literature, that a good many of the royal Arsacid line had been born of Milesian and Ionian mistresses.

That is a joke about the ancestry of the ruling house, and it only works if the house was as Greek as Plutarch thought it was.

The play

Meanwhile Orodes, the king of kings, had made peace with Artavasdes of Armenia and married his son Pacorus to the Armenian king's sister, and the two courts were feasting.

Plutarch stops to explain the company. Orodes knew Greek language and literature. Artavasdes was expert enough in it to write tragedies, orations and histories, some of which, Plutarch says, were still extant in his own day. The entertainment was Greek, and the performer was a tragic actor from Tralles named Jason, who was singing the scene from Euripides' Bacchae in which Agave comes on carrying the head of her son.

The head of Crassus arrived at the door as the tables were being cleared. Sillaces made his obeisance and threw it into the middle of the room. Jason handed the costume of Pentheus to one of the chorus, picked up the head of the Roman triumvir, and went on with the part —

We've hunted down a mighty chase to-day, And from the mountain bring the noble prey

— to great applause. When the dialogue reached the line asking whose hand had killed the beast, Pomaxathres, who had done it, stood up from the table and claimed the answer for himself. The king was delighted and gave presents. Jason got a talent.

What the story is actually evidence of

The scene has been recycled for two millennia as proof of oriental barbarity. Look at what it requires.

It requires a Parthian king who follows Euripides in the original. It requires an Armenian king who writes Greek tragedy himself. It requires a professional Greek actor on tour and employed at a court in Mesopotamia, a chorus with costumes, an audience that knows the play well enough to recognise a substitution in the middle of a scene and to enjoy it, and a Parthian nobleman who can catch a cue in Greek verse and answer it in character.

Nobody stages that improvisation for people who need it explained.

The Sasanians would later write the Parthians out of the history of Iran as an age of petty kings with no culture worth recording, and European scholarship took roughly the same view from the other end — a feudal, derivative interlude between two real empires. The single most famous Parthian scene in all of Western literature is a refutation of both, and it has been read for two thousand years as a confirmation, because the head in it is easier to remember than the play.

And afterwards

Plutarch finishes the account with the fate of the men in the room, and it is worth having, because it is the same story the Arsacid dynasty told about itself for the next two and a half centuries.

Surena was put to death by Orodes not long after, out of envy at his victory. Orodes lost his son Pacorus in a battle with the Romans, fell ill with dropsy, and was given aconite by his second son Phraates; the poison happened to cure the dropsy, so Phraates strangled him instead.

An empire that Rome could not beat spent four hundred and seventy years being run by a family that could not stop killing itself, and it still outlasted everyone who wrote it down.