On Roman roads, people on foot typically managed around 30 kilometres a day, ox carts rather less, and riders or state couriers much more. These are modern estimates, such as those in the Stanford ORBIS model, and the real figure varied with terrain, season and urgency.
Key takeaways
- ORBIS, the Stanford model, assumes about 30 kilometres a day for people on foot and for mule carts, and about 12 kilometres a day for ox carts [1].
- Routine horseback travel is set at about 56 kilometres a day, and fast state carriages at about 67 [1].
- Ancient sources report typical rates for the state post of around 50 kilometres a day, with much faster dashes in emergencies [2].
- A road of 372 km, like the Fosse Way, would have meant many days on foot and fewer by horse, and the map does not calculate journeys.
How fast people travelled
There was no single Roman travel speed. How far you got in a day depended on how you travelled, what you carried, the terrain and the season. Modern researchers give figures as averages, and these should be read as approximate.
The most-used set comes from ORBIS, the Stanford model of the Roman world [4]. Its authors set average daily distances by mode of travel [1]:
- About 12 kilometres a day for ox carts.
- About 20 kilometres a day for porters or heavily loaded mules.
- About 30 kilometres a day for people on foot, including armies on the march, and for pack animals with moderate loads, mule carts and camel caravans.
- About 56 kilometres a day for routine travel on horseback.
- About 67 kilometres a day for fast carriages, either the state post or private couriers.
- About 250 kilometres a day for continuous relays of horses, which the authors say is meant mainly as an absolute speed ceiling for passing information over land [1].
The authors say they aimed for values that would be sustainable for days or weeks, and the model deliberately uses a conservative approach. They warn that we should not assume Roman travel was consistently much faster than in other pre-modern societies [1].
The state courier system gives a second line of evidence. Wikipedia's article on the cursus publicus works through examples of news travelling overland, such as the news of an emperor's accession reaching Egypt. These give an average of about 50 kilometres a day for the post, though a single rider in an emergency could manage 160 kilometres or more in a day [2]. Ox carts, in the same article's summary of the mansio system, could cover about 30 kilometres a day, with pedestrians a little farther [3]. That is why the overnight stops, the mansiones, were about 25 to 30 kilometres apart [3].
An example from our data
Our map does not calculate journeys, and it does not tell you how long a trip would take. What it can do is show how long the roads are, so you can picture the scale in words.
The Fosse Way is 372 km in our data. Watling Street is 352 km. For someone on foot at the ORBIS average, a road of that length would mean many days on the road, while a courier on fast carriages or relays would need far fewer. For the Via Egnatia, at 1,165 km across Greece, Turkey, Albania and North Macedonia, the gap between a foot traveller and the state post would be larger still.
All of these are lengths of mostly conjectured lines: 0% of Watling Street and 0% of the Fosse Way is classed as certain, so even the distance is approximate. The facts page explains how the lengths are measured.
The thing people get wrong
People often quote one number, such as "20 miles a day", as if it settled the question. It does not. Every figure in the lists above is an average chosen for a model, and the real journey depended on weather, terrain, the load and the traveller. ORBIS itself notes that mountain roads and winter conditions are handled only roughly [1]. The record-breaking journeys that appear in ancient sources were recorded because they were unusual, and they are not a guide to a normal day [2].
A second mistake is to think good roads made everything fast. The ORBIS authors point out that by the late eighteenth century, improved roads in France only sometimes beat their modelled speeds, and then only on a few privileged routes [1].
Frequently asked questions
How far could a Roman soldier march in a day?
ORBIS puts foot travellers, including armies on the march, at about 30 kilometres a day, with about 60 for a rapid short military march without baggage .
How fast was the Roman post?
Examples from ancient news journeys give an average of about 50 kilometres a day for the state post, and urgent single riders could do far more .
Did Roman roads make travel faster than later centuries?
Not necessarily. The ORBIS authors say we cannot assume Roman speeds were consistently far better than in other pre-modern systems .
Sources
- ORBIS: The Stanford Geospatial Network Model of the Roman World, version 1 paper (Scheidel, Meeks and Weiland, 2012), Stanford University.
- Cursus publicus, Wikipedia.
- Mansio, Wikipedia.
- ORBIS: The Stanford Geospatial Network Model of the Roman World, Stanford University.
- 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).