Roman roads were built by surveying a line, clearing and digging a bed, filling it with stone and gravel, and finishing with a cambered surface and side ditches. Methods varied with local materials, terrain and the date of the road.
Key takeaways
- Builders surveyed a line first, then dug a bed and filled it with rubble, gravel or stone, so the road was a built structure, not a worn track.
- Many roads sat on a raised bank called an agger and were cambered so water ran off into ditches.
- Surfaces and depths differed a lot from place to place, so no single recipe fits every road.
- Our map shows where roads ran, not how they were built: Itiner-e has no excavation layers in it, and only 2.7% of the length is classed as certain.
How Roman road building worked
The homepage gives the short version. This page goes through the steps and the caveats.
Survey. An engineer inspected the ground and roughly fixed the route. Surveyors (agrimensores) then marked the line with rods, sighting along them, and used a groma to set right angles [1]. See why Roman roads are straight for how lines were chosen.
Clearing and digging. Topsoil was stripped. In many places a trench was dug down to firm ground, and its depth depended on the terrain [1]. In Britain the carriageway was often built on a raised bank, the agger, with shallow ditches on both sides [2].
Filling. The bed was filled with rubble, gravel and stone, sometimes with sand, and the upper part was rammed down [1]. Textbooks often describe layers with Latin names such as statumen, rudus and nucleus, and some Italian roads match that scheme. British excavations tend to show something plainer: a foundation of medium to large stones under a running surface [2].
Surface and drainage. The top might be packed gravel, or in some places stone paving. The road was crowned, or cambered, so rain ran off into the ditches [1].
Variation. Excavated British roads show a wide range. Surface build-up averaged about 51 cm across recorded roads but ran from about 10 cm to 4 m, probably because surfaces were repaired and added to over centuries [2]. Some Weald roads were surfaced with iron slag [2]. Treat any neat diagram of layers as an ideal, not a rule.
Who did the work. Army engineers surveyed and built many roads from scratch in Britain, and later repair usually fell to local authorities [2]. See who built Roman roads.
An example from our data
The Fosse Way runs for 372 km in 34 segments in our data, and Watling Street for 352 km. For the Fosse Way, 99.1% of the length is conjectured and none is certain. That tells you how well the line is known, not how it was built. Nothing in the dataset records a surface, a depth or a bridge. To learn how a particular stretch was built you need a local excavation report.
The thing people get wrong
People often picture every Roman road as a layered, paved highway. That picture comes from a few well-preserved stretches, mostly in Italy. Many roads, including many in Britain, were gravel-surfaced, and some were little more than levelled tracks. The Roman roads that survive best are not a fair sample of all of them.
Frequently asked questions
How long did it take to build a Roman road?
There is no single answer. It depended on the length, the ground and the labour available, and our data does not record building dates.
Were all Roman roads paved with stone slabs?
No. Many had gravel surfaces, and paving was more common in and near towns and on some major Italian roads.
Does the map show how a road was built?
No. It shows the route and how certain it is, not the surface or the layers underneath.
Sources
- Roman roads, Wikipedia.
- Roman roads in Britannia, Wikipedia.
- Every figure from our own map is worked out from the datasets on the data sources page (checked against Itiner-e version 2024 1.3).