A lost Roman city with an 8,000-seat amphitheater turns up on the Danube in Austria

Geophysical surveys in a farming area of Lower Austria have outlined a Roman town that no written record describes in detail: a planned grid of streets covering 25 to 30 hectares, and an amphitheater built for roughly 8,000 people. Gerald Grabherr of the University of Innsbruck and Barbara Kainrath of Research Archaeology lead the work, and the town appears to have lasted only about a century before its people left.

The site lies near the hamlet of Stein, in the municipality of St. Pantaleon-Erla, close to the Danube between Enns in Upper Austria and the Lower Austrian border. In its release of September 25, the university says the team believes it is Claudivium, a place named by the geographer Claudius Ptolemy in the second century. The identification is a hypothesis, not a proven fact.

A street grid and a stadium beside a small fort

Nothing was dug up to produce the map. According to the University of Innsbruck release, the team combined aerial archaeology with geophysical measurements carried out by GeoSphere Austria, and Ancient Origins reports that magnetometry covered roughly 100 hectares while ground-penetrating radar covered about 50. Grabherr describes it in the university release as “a systematically planned urban area of 25 to 30 hectares with a regular street grid and an amphitheater for around 8,000 people,” in the German original “ein planmäßig angelegtes, 25 bis 30 Hektar großes Stadtgebiet.”

The amphitheater’s capacity is comparable to the one at Carnuntum, the better-known Roman city farther east on the Danube, according to Vienna.at’s coverage. What makes the Stein discovery unusual is the company it keeps. Next to the town stood an auxiliary fort holding about 500 soldiers. Civilian settlements normally grew up beside such forts to supply the garrison, but Grabherr says a settlement of 25 to 30 hectares with a street grid beside a fort that size has, so far, no parallel in the Roman Empire.

Other features recorded in the surveys include large cemeteries, with some burials suggesting wealthy inhabitants, and three structures identified as a Roman brewery. Kainrath calls the brewery “a special feature for this region,” according to the release. An unfinished legionary camp at nearby Albing belongs to the same network of Roman sites on both sides of the Enns River, which is the subject of the Innsbruck announcement.

Why the researchers think it is Claudivium

Ptolemy’s Geography lists a town called Claudivium with coordinates that, once converted, land close to Stein. His entry also mentions a river arriving from the north, which the team reads as the Aist entering the Danube here. HeritageDaily’s account notes the match is geographic rather than epigraphic, and other coverage stresses that no inscription naming the town has turned up.

The location also made strategic sense. According to Vienna.at, the site controlled a Danube crossing where the Aist joins the river and offered a short transport route toward the Vltava, the river running through Bohemia. A town there could have served both as a supply base for the fort and as a trading point on the frontier of the Roman province of Noricum.

A planned withdrawal around AD 180

The occupation ran from roughly AD 80 to AD 180. The endpoint is the most discussed result. Coin finds cluster in the period of Marcus Aurelius, the emperor who died in 180, and reach the reign of his son Commodus, who ended the Marcomannic Wars that had troubled the Danube frontier since 166. According to the Innsbruck release, the archaeological finds point to a planned withdrawal from the area around AD 180, with no evidence that the settlement was destroyed.

The team reads this as a deliberate move, not a destruction. La Brujula Verde’s summary of the release stresses that nothing points to burning or sacking. The likely destination for many residents was Lauriacum, the site of modern Enns a short distance upstream, where, according to the reference entry on Lauriacum, Legio II Italica built a legionary camp around AD 200 and excavations found walls dating from AD 180 beneath it. Similar bronze-working tools and brooches at both places are one of the links the team cites.

Why a settlement of this size was founded beside a garrison of 500 soldiers remains the sharper puzzle. The researchers hope further work will fill in the chronology between the Flavian emperors and the Marcomannic Wars.

Funding and a race against the plough

Grabherr and Kainrath are open about the obstacle. In the release they write that “despite the supra-regional importance and the potential of the region, it has so far not been possible to secure medium- and long-term funding.” They describe the potential of a modern mix of aerial archaeology, geophysics, targeted excavation and natural-science methods as “enormous.”

Time is also a factor. Vienna.at reports that farming could gradually destroy the remains within decades, since the town lies under working fields. The site sits within a UNESCO World Heritage area for the Danube Limes but is not yet formally listed itself, according to Ancient Origins, and the release names the University of Innsbruck, Forum Donaulimes, GeoSphere Austria, Research Archaeology and the archaeological services of Upper and Lower Austria as partners. Whether the money for excavation arrives before the next decades of ploughing is the open question the team leaves unanswered.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview