In 200 BC Rome held Italy, Sicily, Sardinia and Spain; by 146 BC it had beaten Macedon, Syria and Carthage, destroyed two of the great cities of the Mediterranean, and acquired provinces it had no machinery to govern.
The previous lesson left Rome with veteran armies, experienced commanders and a taste for long commands. This lesson follows those armies east and then asks the question that matters more than any battle: once Rome had taken these places, what did it actually do with them? The answer is that it improvised, and the improvisations became the empire.
Fifty-three years
Polybius set out to explain how Rome subjected almost the whole inhabited world in less than fifty-three years, and the compression is real.
The Second Macedonian War ended at Cynoscephalae in 197 BC, where legions beat a Macedonian phalanx of the kind that had conquered Persia. In 196 the victorious commander Flamininus stood at the Isthmian Games and had a herald proclaim the Greek cities free, garrison-free and taxless, which Polybius says caused such an uproar that birds fell out of the sky. Roman armies then left Greece.
Antiochus III of Syria, the largest of the successor kingdoms, was beaten at Thermopylae in 191 and destroyed at Magnesia in 190. The peace of Apamea in 188 took Asia Minor west of the Taurus away from him and imposed 15,000 talents, about 360 million sesterces, the largest indemnity in ancient history. Rome annexed none of it, and gave the territory to Pergamum and Rhodes.
Then the pattern hardens. Macedon revolted and was crushed at Pydna in 168; this time it was broken into four republics and later, after another revolt, made a province in 146. In the same year Roman armies destroyed Carthage, which had complied with every clause of the treaty of 201, and Corinth, after an Achaean revolt. In 133 the last king of Pergamum left his kingdom to Rome in his will, and it became the province of Asia.
That sequence contains two different Romes: one that wins and withdraws, and one that annexes and destroys. Explaining the change from the first to the second is the central problem of Roman imperialism.
Why did Rome keep fighting?
Two answers are on offer and both have serious evidence behind them.
The defensive reading, which goes back to the ancient sources themselves, takes each war as a response to a threat or an appeal from an ally, and points to the withdrawals: Rome left Greece in 194 BC, left Asia in 188, and did not annex Macedonia for twenty years after Pydna. On this account the empire was acquired absent-mindedly by a state that kept being drawn back in.
The structural reading, argued most forcefully by William Harris, starts from the Roman aristocracy rather than from foreign policy. Rome fought a war in almost every year of the third and second centuries BC. Its highest honour, the triumph, required a victory with a minimum body count. Its elections rewarded military glory, as the earlier lesson on canvassing showed, and its temples, roads and games were built from booty by men whose careers depended on winning commands. A political class organised that way does not need a threat in order to find a war.
The evidence tends to support the structural reading of motive without discrediting every particular defensive claim, and the most useful way to hold the two together is to notice they answer different questions. Individual wars had occasions, often genuine ones. The system had a bias, and the bias explains why the occasions were so reliably found.
Example. In 196 BC Flamininus proclaimed the Greeks free, and Rome withdrew its armies. In 146 BC Rome destroyed Corinth and made Macedonia a province. What changed?
Not the intentions of a single policy-making mind, because there was no such mind: the Senate contained factions, and its members served short commands and wanted results within them. What changed was the accumulated experience of what freedom-granting produced. Every settlement left in place a Greek state capable of quarrelling with its neighbours and appealing to Rome, so Roman armies kept returning to the same places, and each return was more expensive than the last. Annexation is what a power does when it has decided that repeated intervention is dearer than permanent presence. The change is a learned habit rather than a decision, which is why it is so hard to date.
Now you. From 167 BC Roman citizens in Italy paid no direct tax on their property, the tributum having been suspended after the Macedonian booty came in. What does that fact do to the debate above?
Answer
It strengthens the structural reading considerably. From 167 the ordinary costs of the Roman state were carried by conquered provinces, and the citizen body that voted on wars was the group that stopped paying for them, an arrangement that lasted until 43 BC, some 124 years. Once a war is funded by other people and its rewards are visible at home, the political constituency for stopping is thin. It is not proof of aggression, since the suspension followed the conquests rather than preceding them, and it is a strong statement about the incentives operating on everyone who voted afterwards.
Government by whoever was free
A provincia originally meant a sphere of duty assigned to a magistrate, not a piece of land, and the Roman empire was governed by that word long after it had come to mean territory.
The mechanism was simple and it broke on arithmetic. Rome created two extra praetors in 227 BC for Sicily and Sardinia, and two more in 197 for the Spains, giving six praetors and two consuls: eight holders of imperium a year. It then stopped creating them, while the provinces went on multiplying, so that by the 60s BC there were about fourteen provinces and still ten senior magistrates, of whom the consuls were usually needed elsewhere.
The gap was closed by prorogation: extending a magistrate's imperium past his year so that he governed pro consule, in place of a consul. Prorogation was a device of the Hannibalic emergency that became the ordinary way to staff an empire. Under Sulla's arrangements it was formalised: the year's magistrates served in Rome, then went out as promagistrates afterwards.
Notice what this does to the constraints the Republic had placed on imperium. Annuality is gone, because the commands last as long as the Senate finds it convenient. Collegiality is gone, because a governor has no colleague. Provocatio does not apply, because it protects citizens inside Italy and not the provincials. And there is no civil service: a governor arrived with a quaestor, a few legates, personal friends and his own slaves, and he was general, judge, tax supervisor and diplomat, answerable to nobody nearer than Rome and answerable there only after his term.
The tax farmers
Rome had no revenue department either, so it sold the job.
Provincial taxes were let out to companies of publicani, contractors who bid for the right to collect and kept the difference. In Sicily the arrangement was mild, because Rome inherited a working local system and the contracts were let in Sicily itself. In Asia, the richest province, the law of Gaius Gracchus in 123 BC moved the letting of the contract to Rome, where the censors sold it to companies of Roman equestrians.
The consequences are worth spelling out. A publican company bid a fixed sum for a province's taxes and profited from whatever it extracted above that, in a place where the only authority that could restrain it was a governor who needed its goodwill and its money. Cicero, defending a governor years later, described the provincials' hatred of the tax companies as one of the standing facts of Roman rule, and he was not an unfriendly witness to the equestrian order.
The Senate did respond. In 149 BC it created the first standing criminal court, the court de rebus repetundis, for the recovery of money extorted by magistrates. Its existence is the strongest evidence that Romans recognised the problem; its record is the strongest evidence that recognition was not enough. From 122 BC the juries of that court were staffed by equestrians, the same order as the tax companies, and control of those juries became one of the great political prizes of the late Republic, fought over for the next fifty years.
Example. Cicero, prosecuting the governor Verres in 70 BC, claimed he had extracted 40 million sesterces from Sicily in three years. Put that number in a scale a Roman would feel.
A legionary at this date was paid about 120 denarii a year, which is 480 sesterces. Forty million sesterces is therefore about 83,000 years of a soldier's pay, or the annual wage bill of roughly seventeen legions. One governor, one province, three years. Treat the figure as a prosecutor's maximum, since Cicero had every reason to inflate it and was addressing a court whose members might themselves govern provinces. Even halved, it says that provincial government was a route to wealth of an order that no career at Rome could match, and that is enough to explain everything that follows about the price of elections.
Now you. The extortion court was created in 149 BC and Verres was tried in 70 BC. What does the existence of the court prove, and what does it not?
Answer
It proves that extortion was recognised as systematic rather than exceptional, early, and by the governing class itself, which would not have built a standing court for a rare crime. It also proves the provincials had no vote and therefore had to be given a procedure instead. It does not prove that the problem was contained: the court's juries were themselves drawn from the men who profited from the provinces, convictions of well-connected governors were rare, and Verres is remembered because he lost. A remedy that requires the injured party to find a Roman senator willing to prosecute, and a Roman jury willing to convict, is a real check and a weak one.
What the money did at home
The wealth arriving in Italy after 200 BC was of an order the city had never handled. Aemilius Paullus's Macedonian triumph in 167 BC filled the treasury so completely that direct taxation of citizens was suspended. Pompey's eastern settlement a century later raised the state's annual revenues, on Plutarch's figures, from 50 million drachmas to 85 million, a rise of 70 per cent, which in Roman terms is roughly 200 million sesterces a year going to 340 million.
Alongside the public money came private money on a scale that unbalanced politics. A successful commander took a share of booty, distributed some to his soldiers, spent some on a temple or a road that carried his name, and kept the rest. The men who let and worked the contracts grew into a second wealthy order beside the Senate. And the whole system was self-reinforcing, because wealth won elections, elections won commands, and commands won wealth.
Roman writers noticed and moralised. Sallust argued that fear of Carthage had kept Rome honest and that its destruction in 146 BC removed the discipline, dating the decline of the Republic to that year. The theory is too neat, and something in it survives inspection: the generation after 146 is the one in which the political killings start.
Example. The Isthmian declaration of 196 BC promised the Greek cities freedom without garrison or tribute. Fifty years later Rome was destroying Corinth. Was the declaration a lie?
Probably not, and that is what makes it instructive. Flamininus needed a settlement that would let him take his army home within his command, and freedom for the Greek cities was a formula that cost Rome nothing and bought immediate goodwill. What the Romans meant by freedom was what a patron means: independence exercised within the expectations of a superior. What the Greeks heard was independence. Both sides used the word honestly and read it differently, and every subsequent quarrel, up to the war that ended with Corinth in ashes, grew in the gap between the two readings. Diplomatic language that both parties can accept for different reasons is not a lie, and it is a fault line.
Now you. Rome annexed no territory after Magnesia in 190 BC but imposed the largest indemnity in ancient history. What kind of empire does that describe?
Answer
A hegemonic one rather than a territorial one: Rome took the money, disarmed the loser, and left the administration to somebody else, which is cheap because it requires no garrisons, no governors and no tax collection. The weakness of that model is that it depends on the losers staying weak and the client states staying obedient, and it produces the repeated interventions that eventually make annexation look economical. So the fiscal choice of 188 BC and the territorial choice of 146 BC are the same policy at two stages of its life, and the difference between them is the accumulated cost of going back.
Where this leaves the Republic
By 133 BC Rome governed the Mediterranean with eight or ten senior magistrates a year, no bureaucracy, no permanent revenue service, no professional army, and a constitution designed for a city that could see its own farmland from the walls.
The strains that broke it, though, did not appear first in the provinces. They appeared in Italy, in the countryside that supplied the soldiers, which fifty years of overseas war and imported wealth had changed beyond recognition. What actually happened to Italian land and Italian people is the next lesson, and it is the one place in this course where the ancient explanation and the modern evidence come apart.