How Rome Held Together · Day 1 of 14

The Sea Did the Hauling

In 301 AD, officials working for the emperor Diocletian had a long price list carved into stone and set up in public places across the eastern provinces. The best surviving copy sits at Aphrodisias, in what is now south-western Turkey. It fixes maximum prices for hundreds of things: barley, pork, lions, a haircut, a lawyer’s fee for pleading a case. Buried in it are the freight rates. There is a charge for hauling a wagonload a mile. There are separate rates for a camel and a donkey. And there is a list of sea fares on named routes, priced by the measure of grain carried, with Alexandria to Rome among them.

Put those numbers side by side and the gap is enormous. Historians who have worked through the edict come out with road haulage costing somewhere between thirty and sixty times as much as the same cargo moved by sea. River transport sits in between, several times dearer than the sea but far cheaper than a road. Call the sea-to-road figure about fifty to one. The exact number depends on which route and which animal you pick, but the order of magnitude is not in doubt, and it is the single most useful fact about the Roman empire.

The reason is not mysterious. A ship is a box that the wind pushes for free. A grain freighter on the Alexandria run carried hundreds of tonnes, and the largest were startling: Lucian, writing in the second century, describes a ship called the Isis blown off course into Piraeus, about 55 metres long and 14 wide, deep-hulled, carrying enough wheat, he says, to feed Athens for a year. She needed a crew of a few dozen. An ox cart carried perhaps half a tonne and needed a driver and a team, and the team ate. Oxen pulling a load of grain across a province are steadily eating the cargo they are carrying. Go far enough and there is nothing left to deliver. In practice, hauling bulk grain overland stopped making economic sense after something like a hundred or a hundred and fifty kilometres. Past that the grain cost more than anyone would pay.

So the empire grew into the shape that the cost of carriage allowed. Look at a map of Roman provinces and the thing that should strike you is that it is not a blob with a centre. It is a ring around water. Baetica and Tarraconensis on the Spanish coasts, Narbonensis in southern Gaul, Italy, Greece, the cities of western Asia Minor, Syria, Egypt, the North African provinces from Cyrenaica to Mauretania. The Romans called the Mediterranean mare nostrum, our sea, and the phrase is closer to an accounting statement than a boast. The sea was the cheap part of the empire. Everything else was expensive.

Go inland and the empire thins out fast. Big cities stayed on the water or within reach of it. Alexandria sat on the Nile delta, Carthage on its bay, Ephesus at a river mouth, Corinth on its isthmus. Antioch, one of the three or four largest cities in the empire, was on the Orontes river with its own port at Seleucia a short way downstream. Where Rome built a great inland city it was almost always on a navigable river: Lugdunum, modern Lyon, at the meeting of the Rhône and the Saône, which together made a highway from the Mediterranean to central Gaul. Cologne and Mainz on the Rhine. London on the Thames. The genuinely inland regions — the high plateau of central Anatolia, the interior of Spain, the mountains of the Balkans — stayed poorer, less urban and less Roman in their habits, not because Rome disliked them but because the freight arithmetic never worked there.

You can see the traffic in the rubbish. In Rome there is a hill called Monte Testaccio, about thirty-five metres high, made entirely of broken pottery. It is the dump of the city’s river port, and it is built from the smashed amphorae that carried olive oil, an estimated fifty-three million of them, the great majority from the valley of the Guadalquivir in southern Spain. Spanish oil reached a Roman table more cheaply than oil from a hill farm two days’ cartage away. That is the fifty-to-one ratio showing up as a mountain of shards.

The sea charged for its cheapness in other ways. It was seasonal. Sailing more or less stopped in winter — the phrase is mare clausum, the closed sea — and the fourth-century writer Vegetius gives the safe window as roughly late May to mid-September, with risky shoulder periods either side and nothing much moving between November and March. It was also asymmetric. Running from Italy to Alexandria with the prevailing north-westerlies behind you could take nine or ten days. Coming back against them could take a month, sometimes two. A cargo that was cheap was not therefore fast or reliable, and an empire that depended on cargo depended on the weather.

Everything that follows in this course sits on top of that arithmetic. The grain that fed the city of Rome came by ship because it could not have come any other way. The roads, which are the thing everyone remembers about Rome, were not built to carry bulk freight and mostly did not, because on a road the cargo costs fifty times more — what they carried was soldiers and letters, and their real product was speed. The army sat on the edges of the empire partly because the edges were where the water ran out. And the two halves of the empire that eventually came apart were, among other things, two different sets of shipping lanes.

Here is the fact that reorders the map in your head. It was cheaper for the city of Rome to eat bread grown on the Nile, nineteen hundred kilometres away across open water, than bread grown in the Apennine valleys a hundred and fifty kilometres up the road. For Rome, Egypt was closer than Italy.