Mapping 186,000 Miles of Roman Roads Shows Rome Was Never the Empire's Main Hub

In 20 BC Augustus set up a gilded marker in the Forum, the Milliarium Aureum (the Golden Milestone), to show where every road in the empire began. Two thousand years later we still say all roads lead to Rome. A new study says the empire’s own road map disagrees.

The paper, published in Nature Communications on 15 September, is the first to look at the whole Roman road system at once. The team, from Aarhus University, Queen Mary University of London, the Autonomous University of Barcelona and several other European institutions, worked from Itiner-e, the open digital atlas of ancient roads released last year. It holds close to 300,000 kilometres of road, about 186,000 miles, running from Britain and Morocco to Syria and Arabia.

Rome Sat at the Edge of Its Own Map

The researchers turned that tangle into a graph: 10,516 junctions and endpoints, joined by 14,901 road links. Then they ran network centrality on it. Put simply, it measures how often a place sits on the best route between other places, and so how much traffic has to pass through it if people travel sensibly.

On land, Rome didn’t come out on top.

Look at a map and it makes sense. Italy is a long peninsula hanging off the southern edge of Europe. Someone going from Gaul to Syria had no reason to go down to Rome, and on foot or behind an ox they wouldn’t. The cities that scored well sat where the empire’s pieces met: Byzantium on the Bosporus, Ancyra in the middle of Anatolia, Antioch on the way into Syria, Corinth on its isthmus.

Byzantium is the telling one. Constantine made it his new capital in 330 and renamed it Constantinople. The roads suggest he chose well. For overland travel across the whole empire it was the better centre, and the old saying fits it far more than it ever fit Rome.

One caveat is worth keeping. This is land travel only, and Rome lived on the sea. Its grain came from Egypt and Africa by ship, through Ostia and Puteoli, and no road could compete with that on cost. Rome’s pull was political and maritime. Its roads were doing a different job.

Provincial capitals scored high right across the network, and so did the main trunk roads. The authors read that as a sign the system was never a starburst drawn out from the centre. It grew from decisions made at many levels, by governors and army engineers, city councils and local landowners, over a few hundred years.

Straight Where They Could Be

Then there’s the ruler-straight Roman road. Plenty of them were remarkably direct, especially across flat country. But when the team measured how much each road bends (its real length against the straight-line distance between its ends) Roman roads came out a little less straight than major modern ones. The engineers went round mountains. They followed valleys and rivers, and they tied in towns that were already there. A modern crew can blast through a hillside. A Roman one mostly couldn’t.

What Survives Under the Tarmac

The last claim is that Europe still drives on Roman roads. Partly. Across the former empire the old and new networks show a moderate match, but it swings a lot from one region to the next. Terrain matters, and so do rivers, mountain passes and which cities grew or shrank over the centuries. Where a pass is the only way through, the motorway follows the legions. Where a Roman town withered, its roads usually went with it.

And the whole picture rests on a thin base. The people behind Itiner-e have said only a small share of Roman roads can be placed with real certainty. The rest is reconstruction from ancient itineraries, milestones, old surveys and excavation reports. The map will keep changing.

Rome probably won’t end up in the middle of it.