The Roman empire was governed by fewer officials than a medium-sized modern city employs, and the way it managed that is the single most important fact about how it worked.
The previous lesson found an army of about one soldier per 240 inhabitants and a tax take of roughly five per cent of everything the empire produced. Both numbers point the same way: whatever held the provinces together, it was not the weight of the state pressing on them. This lesson counts the civilian administration and finds it smaller still, then asks what actually did the governing instead.
The arithmetic of an absent state
Under Augustus the empire's entire senior administration, senators and equestrians holding posts with real authority, numbered somewhere around 150 men. The equestrian service grew steadily, and by AD 200, with procurators, prefects and a much larger financial establishment, the figure is perhaps 350. Below them sat clerks, slaves and freedmen of the imperial household, and soldiers seconded to office work, but those are staff, not decision makers.
Example. Set 350 senior officials against sixty million people, then set the same population against the salaried officialdom of Han China, which ran to something over 130,000 men. What kind of state is Rome?
Rome gives one official per 171,000 inhabitants; Han China gives one per 460. Two empires of similar extent and similar population differ by a factor of nearly four hundred, which is not a difference of degree. The Han state reached into the county and the village, registered households, assessed individuals and moved grain by administrative order. The Roman state did almost none of that directly, and a Roman governor could not have registered a village if he wanted to, because he had nobody to send. So either the empire was not really governed, which the tax receipts and the law reports refute, or the governing was done by somebody who does not appear in the count of imperial officials. The rest of this lesson is about who. Keep the comparison rather than the figures: the Han total is disputed and the Roman one is a reconstruction, but no plausible adjustment to either closes a gap of that size.
Now you. A governor of Bithynia arrives with a staff of perhaps a few dozen, including his friends, his clerks and some seconded soldiers, to administer a province of a couple of million people. What can he actually do with his year?
Answer
He can travel, judge and write letters, and that is close to the whole list. The core of a governor's year was the assize circuit: he moved between a fixed set of towns, held court, heard the cases too large or too politically charged for local magistrates, received embassies, checked accounts if he was asked to, and moved on. He commanded whatever troops were in the province and he was the emperor's correspondent. What he could not do was administer, in the sense of maintaining a permanent presence that touched ordinary people, because he had no staff to leave behind. Everything continuous, meaning tax collection, policing, streets, water, markets, corn supply, festivals, buildings and the courts for small claims, had to be done by somebody local and permanent, or it did not get done. The empire's answer was to make that somebody a city.
The city as the unit of government
The empire was a mosaic of roughly two thousand self-governing communities, the civitates, each consisting of an urban centre and the countryside attached to it. Where cities already existed, in Greece, Asia Minor, Syria and Africa, Rome used them. Where they did not, in Gaul, Spain and Britain, Rome created them, converting tribes into civitates with a capital and a council so that the tribe would have an address, a set of magistrates and a body that could be held responsible for a tax total.
Each had a council, usually of about a hundred decurions, drawn from men who met a property qualification, at Pliny's home town of Comum set at 100,000 sesterces. Each had annually elected magistrates, usually two senior ones with judicial power and two responsible for streets, markets and public buildings. It is the constitution of the Roman Republic in miniature, reproduced in two thousand copies, and that similarity is not an accident: it was the only model of self-government Rome had.
Divide sixty million by two thousand and the average community holds 30,000 people. The average is misleading in the usual way, because Rome itself held around a million, Alexandria and Antioch several hundred thousand, and the median civitas was a small town of a few thousand with farmland around it. But the order of magnitude is right, and it is the useful figure: the empire delegated the government of a few tens of thousands of people to a hundred local landowners, and then dealt only with them.
What a charter actually says
The best evidence for how much autonomy this meant is a set of bronze tablets found near Seville in 1981, the charter of the small town of Irni, granted under the Flavians around AD 91. About two thirds of the text survives, and it is the fullest Roman municipal constitution we have.
It reads like a manual: how magistrates are elected and in what order the voting groups are called, how the council is summoned and what quorum it needs, how public contracts are let, how the accounts are audited, how a freed slave's status is recorded, how much a magistrate may spend on games. Two provisions are worth more than the rest.
The first is that the local courts may hear civil cases up to a value of 1,000 sesterces, and nothing above. That is a little over a year's pay for a legionary, so it covers the ordinary business of a small town and stops precisely where property of any real size is involved. Above the line, the case goes to the governor. Roman rule is not the absence of interference: it is a threshold.
The second is that a man who holds a magistracy at Irni becomes a Roman citizen when he leaves office, along with his parents, his wife and his children. The town is a Latin community, not a Roman one, and the charter turns its annual elections into a citizenship machine. That provision is where this lesson connects to the next one, and it explains a great deal about why local men competed for offices that cost them money.
Why the local rich paid for the privilege
They did pay, and heavily. A man entering a magistracy or priesthood owed a summa honoraria, an entry fee to the treasury of his own town, recorded at a couple of thousand sesterces in modest places and tens of thousands in the wealthy cities of Africa. On top of that he was expected to give: to pave a street, roof a portico, endow a festival, distribute cash at his son's coming of age, or build something with his name across the front of it. The system has a modern name, euergetism, and the inscriptions on which it rests number in the tens of thousands, because the whole point of the gift was that it was recorded.
Pliny the Younger, whose letters make him the best-documented benefactor in the empire, gave his home town a library with 100,000 sesterces for its upkeep, offered to pay a third of the cost of hiring teachers if the parents found the rest, and endowed a fund for the maintenance of local children. The endowment is worth doing the arithmetic on, because he explains his reasoning.
Example. Pliny conveyed an estate worth more than 500,000 sesterces to an agent of the town, then took the land back charged with a permanent rent of 30,000 sesterces a year. Why not simply hand over the cash, and what return did he set?
The return is 30,000 on 500,000, or six per cent, which is close to the ordinary yield on Italian land. He explains the structure himself: a gift of cash would be spent, and an estate given outright would be farmed badly by whoever the town appointed. So the town gets neither the capital nor the management, only an income charged on land that stays in the hands of somebody with every reason to keep it productive. Notice what that tells you about ancient public finance. There is no bank, no bond and no institution that can hold and invest money on a town's behalf, so an endowment meant to last has to be a charge on land, which is why so many benefactions take the strange-looking form of a rent or an obligation attached to a farm rather than a sum in a treasury.
Now you. Trajan's alimenta scheme worked the same way on a state scale. At Veleia in northern Italy an inscribed bronze table records loans of 1,044,000 sesterces secured on local estates at five per cent, funding 245 legitimate boys at 16 sesterces a month, 34 legitimate girls at 12, one illegitimate boy at 12 and one illegitimate girl at 10. Do the two sides balance?
Answer
Exactly. The interest is 1,044,000 at five per cent, or 52,200 sesterces a year. The payments are 245 times 16 times 12, which is 47,040; plus 34 times 12 times 12, which is 4,896; plus 144; plus 120. The total is 52,200. The scheme was designed backwards from the yield, and the number of children was chosen to consume it precisely.
Two things follow. First, the document is genuinely an account and not a monument, which is unusual and makes it good evidence. Second, look at what the state has done: rather than run an orphanage, it has lent public money to landowners against their estates and used the interest stream to fund child support administered by the town. Even the emperor's own welfare scheme is delivered through local proprietors, because there is no other machinery. The rates also record something the inscriptions rarely state so plainly: a girl was worth three quarters of a boy, and illegitimacy cost a further quarter.
The governor and the leaks in the system
Delegated government works while the delegates are willing and solvent. Pliny's posting to Bithynia and Pontus around AD 110 is the best surviving picture of what happened when they were not, because he wrote to Trajan about almost everything and the correspondence survives, 121 letters in the tenth book of his collection, most with the emperor's reply attached.
The province's cities had been spending. Nicomedia had put 3,318,000 sesterces into an aqueduct, abandoned it, demolished it, and started another. Nicaea had spent more than ten million sesterces on a theatre that was cracking, and had a gymnasium burnt down and rebuilt to no plan anybody could follow. Claudiopolis was excavating an enormous bath at the foot of a hill, financed by the entry fees of newly enrolled councillors. Pliny is not investigating corruption in any of these cases. He is reporting waste, on a scale that is easy to feel: the Nicomedian aqueduct is the annual pay of nearly 3,700 legionaries, poured into a hole and then knocked down.
Example. Trajan sent Pliny to Bithynia with a special commission rather than letting the province take its turn under an ordinary proconsul. What does the appointment itself tell you, before a single letter is read?
That the normal system had no way of noticing a problem until it was large. Bithynia's cities had been mismanaging their finances for years and two of its recent proconsuls had been prosecuted for extortion, and nothing in the machinery brought either fact to Rome's attention early. There was no audit, no inspectorate and no reporting line from a city council to anybody. Correction came as an exception, a trusted man with an enlarged brief, and it came after the money was gone. Rome paid that price knowingly, because the alternative was a salaried provincial civil service it could neither afford nor staff. Notice too what 121 letters implies about the volume of business reaching an emperor who governed forty-odd provinces: it was small enough that Trajan could read it, which is only possible because almost nothing came up.
Now you. Trajan repeatedly refuses Pliny's requests, most famously by rejecting a fire brigade of 150 men for Nicomedia after the city burned. Why refuse something so cheap and so obviously useful?
Answer
Because the emperor's worry was not fire but association. Trajan's reply says that whatever name such a body is given, and however few its members, men who come together for a common purpose will form a political club, and Bithynia's cities were already troublesome. He tells Pliny to equip property owners with fire-fighting gear and to call on the crowd if needed, which is to say the problem should be solved without creating an institution. Rome delegated government to cities but was consistently nervous about permanent organised bodies inside them, and the same nervousness runs through the law on collegia, the bans on unlicensed guilds, and, in a later lesson, the way the state reacted to a network of Christian congregations. An organisation it had not authorised and could not see was a category the empire had no comfortable place for.
What the cities got out of it
It is easy to read all this as extraction: Rome takes the tax, the local rich pay the costs of government and buy the honours, and the imperial power supplies almost nothing. That undersells the exchange.
A city under Roman rule was safe from its neighbours in a way no Greek city had ever been, and the wars that had periodically wrecked Asia Minor stopped. Its elite kept the local standing they already had and gained a route out of it, through equestrian rank, imperial service and eventually the Senate. Its ambitions could now be pursued in a currency other than violence, which is why the cities of the east put such extraordinary energy into competing over titles, temples and games. That competition looks trivial and was not: it was what political life became once the alternative had been closed off, and the buildings it produced still stand.
The price is worth naming now because it becomes lethal later. The men who ran the cities were personally liable for the tax total, and the honours were expensive. While being a decurion was worth more than it cost, the system financed itself and needed no enforcement. When the third century made the burden heavier and the honour cheaper, it stopped working, and the state's response was to make the office compulsory and hereditary, which is a good working definition of an institution that has died.
For now the exchange held, because the reward was real. The next lesson takes up what that reward was: membership itself, and how a status that began as the privilege of a few Italian towns spread until an emperor could grant it to every free inhabitant of the empire at once.