A new analysis of the road network of the Roman Empire has dispelled the enduring myth that all roads lead to Rome, showing instead that other important cities — such as Constantinople, or modern-day Istanbul — were more central in the network. “When we look at all terrestrial transport in the entire Roman Empire, there’s one path that is the most central path going through the empire itself, and that path doesn’t go through Rome,” said Tom Brughmans, an associate professor in the School of Culture and Society at Denmark’s Aarhus University and coauthor of the new analysis, published Tuesday in the scientific journal Nature Communications. At its greatest extent, in AD 117, the Roman Empire spanned from Britain to the edge of the Sahara, covering parts of Europe, Asia and Africa over about 2.1 million square miles (5.5 million square kilometres). Rome was at the heart of the empire politically and economically, but geography made it more central to Mediterranean shipping than the road system. “The Italian peninsula is just this land appendix sticking out into the Mediterranean,” Brughmans said, “so the most central path passed through the North African coast, then east through modern-day Turkiye and cities like Antioch and Ankara, and then through Constantinople, before heading north of Rome, through Milan.” In 324, the Roman emperor Constantine I moved the empire’s capital for strategic and economic reasons from Rome to Constantinople, which until then was named Byzantium, so it’s not surprising that the city was prominent in the road network. The analysis by Brughmans and his colleagues is based on Itiner-e, a digital atlas of the Roman road network that was published last year and is accessible online. Compiled using historical sources, excavation reports, satellite imagery and previous datasets, Itiner-e contains 186,000 miles (nearly 300,000 kilometres) of roads. However, within the atlas, the location of only 2.7 per cent of the roads is known with certainty, with about 90 per cent of the map less precisely known and about seven per cent simply hypothesized. In the new analysis, which considered parametres such as the roads’ elevation, orientation and slope, the researchers also looked at other widely held assumptions about Roman roads. They demonstrated that these roads weren’t as straight as commonly thought, and that while some key modern-day routes follow the same path as ancient Roman roadways, many others — even if labelled as Roman — don’t. The origin of the myth The proverb “all roads lead to Rome” dates to medieval times. It first appeared in writing in 1175 by the French poet Alain de Lille, who wrote the phrase “Mille viae ducunt homines per saecula Romam” in Latin, meaning “A thousand roads lead men forever to Rome.” The importance of the road network was very evident to the Romans themselves: In 20 BC, the emperor Augustus placed a “golden milestone” in the Roman Forum as a symbolic source of the main roads in the empire. The milestone is widely interpreted as having played a role in the rise of the myth, although the modern meaning of “all roads lead to Rome” is not literal, but rather that there are different ways to reach a goal or result. “It’s a medieval proverb, not an ancient one, and it indicates the symbolic centrality of this massive capital for this continental-scale empire,” Brughmans said. “Rome was on a nexus of major shipping lanes, and also where the government was, so it was definitely central. The unfortunate bit is the roads bit,” he added. “Perhaps what reinforced this myth is that there was a system of major roads that was used by the emperors to administer the regions and share information, like a communication network, and Rome was depicted as being at the centre of that major road network. The persistence of this proverb is incredible, even though it’s demonstrably false.” The researchers also verified whether Roman roads were uniformly straight, noting that while there is some truth to this widespread assumption — as well as surviving examples of straight Roman roads such as the Appian Way in Rome — the Romans had to contend with terrain and technical limitations. “If the topography allowed them to build straight roads, then they would build those straight roads,” Brughmans said. “But if the topography made it hard, then the roads would be wiggly.” Where the landscape allowed, the Romans indeed built straight roads, especially in lowland regions such as Western Europe and North Africa. However, even these roads were not as straight as modern ones, mostly due to engineering constraints. “We also proved a surprising degree of persistence between Roman roads and major modern roads,” Brughmans added. “It is absolutely exceptional that there is this ancient infrastructure that is still visible and that still determines the way in which people interact in the present day.” Examples include Via Augusta, which spans almost 1,000 miles (1,600 kilometres) from Gades (modern-day Cadiz) in the south of Spain to the Pyrenees in France. In the city of Tarragona, in Spain, the ancient road crosses a triumphal arch called Arc de Berà, while a modern-built road runs to the side of it. This arrangement, Brughmans said, illustrates one of the ways in which ancient Roman passageways have survived, with contemporary, larger paved roads running roughly along the old path. In other cases, people can still walk along Roman routes, such as the Appian Way, which features some of the original volcanic stones that the Romans expertly laid. However, most of the ancient road network has now disappeared or ended up being buried over the centuries. And not all “Roman roads” are actually Roman, according to Brughmans. “Everywhere in Europe, in North Africa, in the Middle East, you’ve got roads that are called Roman Way — Römerstraße in German or Calzada Romana in Spanish — but so many of these are not really Roman roads,” he said. “This kind of demonstrates how persistent the idea is that the Romans had such an impact on infrastructure, the assumption being that if we see a straight road that is quite impressive and a bit old, it’s probably Roman, even though that’s not quite true everywhere.” The results of the study are remarkable, but rather than revolutionizing earlier assumptions, they confirm and demonstrate mathematically something that had been previously described only in broad terms, according to Francesca Fulminante, a senior research fellow in the department of anthropology and archaeology at England’s University of Bristol. She was not part of the new analysis. The dataset of the road network is not fully accurate in all regions, she said in an email, but the issue does not detract from the results and is quite common for such a large-scale project. A mathematical demonstration Brughmans also noted that while the Romans built many roads, they also inherited some from areas they conquered, meaning their control over the structure of the network was somewhat limited. The researchers behind the analysis said that their work isn’t finished and they plan to add river and sea connections to the network in the future. The findings of the study are important because they provide a deeper understanding of the Roman road system as well as reusable data to improve future models, said Sarah E. Bond, an associate professor in the department of history at the University of Iowa. She also didn’t participate in the work. “An improved reconstruction of the Roman road system continues to fill in gaps in our knowledge of the ancient world,” Bond said via email. “Roads provide us better insight into how other things — beyond just donkeys, wagons, merchants and the occasional camel — travelled along roads. Things like religious beliefs, philosophical ideas, and even disease hitched a ride on the vehicles used to move goods and peoples from place to place.” The analysis also allows for some myth busting, she added. “Instead of blithely repeating the common notion that Roman roads were all uniformly straight, historians, archaeologists, and the general public can add a caveat,” she said. “Yes, Roman road systems were most often straight when the topography allowed it, but were also reactive surfaces. Roman surveyors responded to elevation changes, dips into river valleys, topographical features like mountains, and the natural lay of the land.” These findings were made possible through the work of previous digital projects that constructed maps of the ancient world before Itiner-e and “helped to blaze the trails within the field,” Bond noted. “All in all, this is a fantastic project that will only spawn new and more accurate models of the ancient world.”
