Itiner-e, the road map of the Roman Empire that everyone is talking about
Rediscovering the Roman world through its network of roads
It is unusual for a cartographic topic to appear in the press, let alone for it to be everywhere. But that is what has happened with the Itiner-e project.
In recent years, we have become accustomed to seeing maps of all kinds, such as real-time air traffic maps, submarine cables, maritime container routes, and a thousand other imaginable topics. There were also “road” maps (or rather, carriageway maps) of the Roman Empire, but Itiner-e represents a huge leap forward from what we had until now.
Itiner-e is a project that has just hit the mainstream press and promises to become an essential reference for anyone curious about Rome, historical cartography, or how infrastructure shapes the world over time. Itiner-e is both a digital atlas and a set of high-resolution data on Roman roads around 150 AD, when the empire reached its maximum extent. It is coordinated by an international team led by Pau de Soto and Adam Pažout (Autonomous University of Barcelona) and Tom Brughmans (University of Aarhus).
The appeal of the idea lies not only in creating “another Roman map” (there are quite a few out there), but in the underlying data and the study behind it. The new map shows 299,171 km of mapped roads, compared to the approximately 188,554 km in the most comprehensive digital resource available to date (the DARMC, based on the Barrington Atlas). The routes are divided into 14,769 sections, classified hierarchically as main roads, with 103,478 km (34.6% of the total), and secondary roads, with 195,693 km (65.4%).
In other words, almost twice the length of roads that appeared on previous standard maps, distributed in much greater detail. And even so, the team itself points out that this is only a small part of what may exist. It is estimated that only 3% of the routes are located with complete certainty, the rest being more or less reliable reconstructions based on sources.
Itiner-e did not come out of nowhere. It is the continuation of previous lines of work such as Mercator-e (focused on the Iberian Peninsula) and Viator-e, which had already been compiling regional roads for years, and is integrated into larger projects (MINERVA, Desert Networks, etc.). Each section of road appears as a separate entity, with a stable identifier that allows it to be cited, linked from other projects, and reused. The geometry and attributes are stored in standard formats (GeoJSON and similar), designed so that anyone can download the dataset and work with it in QGIS, Python, R, or whatever they want. In addition, it is explicitly linked to other classical world repositories such as Pleiades (ancient places), Vici.org (archaeological atlas), Tabula Imperii Romani / TIR-FOR, etc., so that not only lines are visible, but also the associated system of cities, buildings, bridges, and milestones.
On that basis, the project website offers what is effectively a “Roman Google Maps” (one of the most widely used expressions in the press these days): a comprehensive network viewer with tools for exploring routes and travel times, designed for both curious minds and researchers. The process of creating the map has been divided into several phases. It began with the identification of the roads through an exhaustive review of archaeological literature and major repertoires (Barrington Atlas, DARMC, Tabula Imperii Romani, etc.). Intensive use was made of ancient sources such as the Itinerarium Antonini, the Tabula Peutingeriana, and other itineraries, as well as a study of the epigraphic database, LIRE (Latin Inscriptions of the Roman Empire), with 8,388 geocoded milestones that serve as “markers” in the territory. Each milestone indicates distances and, in many cases, public works financed by a particular emperor.
The spatial location of each element is based on the cross-referencing of this data with historical orthophotos from USAF photogrammetric flights in the 1950s, where the traces of roads are more visible than today in many urbanized areas (in many places they have now been completely erased). Current satellite images (ESRI, Google) have also been used and, for the Middle East, the legendary photos from the Corona spy program (1967–1972), which were essential for locating sections now submerged under reservoirs. Topographic maps from the 19th and early 20th centuries have also been consulted, where many roads were labeled as Voie Romaine or equivalent. Each section was then traced by hand on a GIS, with a target resolution of between 5 and 200 meters. When the road is visible (straight line on flat ground, bridle path on hills, remains of paving, centuriation alignments, etc.), the image and maps are followed. In other cases, a “topological” reconstruction is used, linking what we know (cities, milestones, excavated remains, bridges) by following logical relief corridors. This often means matching the Roman route with current secondary roads, rural roads, or tracks, which inherit the logic of the ancient route. In addition to geometry, attributes such as geodetic length and average slope are calculated for each segment, which then allows travel times or transport costs to be estimated systematically.
All of this is certainly impressive, but of course, there is the issue of certainties, doubts, and “confidence maps.” One of the most honest (and useful) aspects of the project is that it does not hide uncertainty, but rather maps it. Each section of road is associated with a certainty rating (accurate, conjectural, hypothetical), based on the quality of the sources that support it. It also adds information about the sources that document it: excavations, specific milestones, itineraries, regional studies, etc., as well as “confidence maps” covering the entire Empire, showing where the network is well documented and where we are practically in the dark. From all this, it can be seen that only 2.7% of the sections have a spatial location considered “certified.” 89.8% are considered conjectural and 7.4% purely hypothetical. From there, Itiner-e becomes a map of pending tasks for the archaeology of the future.
There are some curiosities that become clear with this project and others that are similar but less complete. A map of roads like this is, at heart, a map of power. When all the information is brought together, several interesting ideas emerge. The big “hubs” are not always where the clichés say they are. An analysis of density and connections shows that, in strictly terrestrial terms, the major nodes of the network are in the Po Valley and the Alpine corridor, not in Rome itself, which ends up being something of a cul-de-sac at the end of a peninsula. Rome only regains its centrality when maritime and river routes are added, something that other projects such as ORBIS model very well. The growth of the Roman network was based on older roads. Itiner-e includes pre-Roman routes as long as they were still active in the imperial era, reminding us that the Romans were great engineers, but also magnificent recyclers of previous infrastructure. The roads were vehicles for everything: legions, goods, imperial orders… and diseases. The project team explicitly mentions cases such as the Antonine Plague of the 2nd century AD, which would have spread at “continental” speed thanks to this network of roads and sea routes.
The press insists on an easy-to-remember figure: those almost 300,000 km are equivalent to more than seven times the circumference of the Earth. It may sound simplistic, but it gives an intuitive idea of the scale. In addition, many of the mapped roads are still part of the current road network. In some cases, highways have erased traces, but in others, the old corridors remain virtually intact, now paved. Itiner-e is not just a pretty map to illustrate headlines about “three hundred thousand kilometers of roads.” Above all, it is a serious attempt to bring order to two centuries of scattered bibliography, package it into an open dataset, and, on that basis, invite everyone to continue completing the puzzle. Like any good digital project, it is alive, will grow with new regional contributions, will correct errors and, with a little luck, will end up being for Roman roads what Pleiades is for ancient place names or ORBIS for mobility models.
