Administrative centres in the Province of Raetia

See this text in

The extent to which Raetia was divided into administrative districts remains largely unknown, which has obvious knock-on effects on the identification of central places. Only municipium Aelium Augustum-Augsburg, a Roman charter town, can be viewed as proven through epigraphic evidence. Its honorific ‘Aelia’ points to a Hadrianic (117-138AD) foundation, but whether its municipium status was preceded by a period under devolved local administration, is unknown. The existence of several civitates can be postulated from a fragmentary inscription from Isny. This dates to the mid 2nd century AD, but does not contain further details. The level of urban infrastructure at Cambodunum- Kempten and Brigantium-Bregenz is such that they are usually thought of as the central places of the Estiones and Brigantes, two pre-Roman tribes, and Strabo had labelled the two settlements with the Greek term polis, in the early 1st century AD, identifying them as centres of their respective tribal areas. A further settlement Damasia, which is probably to be identified with the settlement on the Auerberg in the Schongau, is classified by him as an akropolis. A further indicator of Kempten’s role as an administrative centre is its appearance on a milestone and the same status has traditionally been assumed for Bregenz, although this has recently been questioned. Certainly, there are no inscriptions, giving the community's title or mentioning local officers. That said, both sites have produced very few inscriptions, which might lie at the root of this apparent lack of evidence. On the other hand inscriptions honouring members of the imperial family are known from Kempten, Chur and Bregenz, all dateable to the early years of Roman rule, and these are often cited in support of the creation of civitates. The quality of this evidence is markedly reduced, however, by the lack of dedicant names for these stones. The legal status of the communities (chartered Roman towns such as colonia or municipia, or civitas capitals) thus remains uncertain.

For a province the size of Raetia just three administrative districts seems unusually low by comparison with its neighbours, however, and this is still more striking in view of the large number of surviving tribal names known from its initial occupation. Further possible administrative centres have thus been suggested for Curia-Chur, in the upper Rhine valley, Phoebiana-Faimingen, along the Danube, and vicus Scuttarensium- Nassenfels, to the north of the Danube. Of these, Faimingen's appearance on milestones and its possession of a city wall may be considered as supporting evidence for such a claim, but the existence of an important sanctuary and a visit by the emperor Caracalla might distort the picture. As a result, this account will, henceforth, only give detailed consideration to Augsburg, Bregenz and Kempten.
Because of its early provision with representational building complexes, in the 1st century AD, Cambodunum-Kempten has been suggested as Raetia's first provincial capital. If so, however, it was replaced in this role by municipium Aelium Augustum- Augsburg, in the course of the 2nd century AD.


Political structures before the Romans

Little is known about the political structures of the later province of Raetia at the time of its conquest in 15BC. The literary sources report the names of a comparatively large number of tribes, whose individual locations are sometimes far from secure. For example, even the Raetii, who gave the province its name, cannot be further located. Another important unit was the Vindelici, who settled in the Voralpenland (the relatively flat area north of the Alps) and consisted of several sub-tribes. The Celtic culture of the territory of the later province was already in decline in the first half of the 1st century BC, as is shown in the abandonment of the oppidum at Manching.


Indigenous predecessors

Both Bregenz and Augsburg have produced late LaTène finds but, as no associated buildings are known, little can be said about the size and character of the original settlements, or their possible continuity into the Roman period


Creation of the administrative centres

Cambodunum-Kempten was founded in the 10s AD and, although final epigraphic proof of its status as administrative centre is still absent, its early development and the provision of stone buildings by the mid 1st century AD, points to its unusual position within the province. This may be evidence for a possible role as the provinces first capital, but it might equally indicate an administrative centre for the Estiones.

Cambodunum-Kempten. Plan of the Roman settlement.
Cambodunum-Kempten. Reconstruction of the Roman town.

The military had occupied Augsburg by about 10AD and a fort has been found in the inner city which dates to the following decade. It remains unclear whether the town came to serve as an administrative centre immediately after the army withdrew in 69/70, but there was contemporary redevelopment of the fort area with civilian structures. What is reasonably well established, thanks to its adoption of the emperor's cognomen, is the creation of the municipium Aelium Augustum-Augsburg during Hadrian's inspection tour of the province in 121. We do not know, if its territory contained (or was restricted to) the land of the locally attested Licates or Licatii, and no tribal name is included in the official name. Moreover a Licatus is attested in the mid 2nd century AD as an auxiliary soldier without citizenship, which might suggest that not all of the tribal territory was included in the municipium.

Augusta Vindelicum-Augsburg. Reconstructed plan of the fort.
Municipium Aelium Augustum-Augsburg. Plan of the Roman town with reconstructed street grid.

Brigantium-Bregenz was also founded in the 2nd decade AD, but the few hints of an earlier, possibly military, presence do not allow us to trace a continuous development. In contrast to Kempten, we know of a fort of the first half of the 1st century AD AD, although its size and length of occupation remain unclear. Likewise, the question of when the settlement may have begun to serve as the administrative centre of the Brigantii remains equally unresolved. However, the introduction of stone structures and the construction of public buildings only began in the 2nd century AD.

Brigantium-Bregenz. Roman features in the centre of the settlement.
Brigantium-Bregenz. Model of the settlement from the north.

Building phases

At Cambodunum-Kempten, the transition from timber to stone construction for the public buildings dates to 50 AD. Residential buildings, on the other hand, began to employ stone dwarf walls for timber-framing from c.80 AD, with some timber buildings continuing into the 3rd century AD. At Augusta Vindelicum-Augsburg, stone building did not start until the early 2nd century AD, with timber construction again continuing in several areas, but the level of research is such that the change in construction technique is harder to pinpoint at Brigantium-Bregenz. The introduction of stone buildings is generally hypothesised to be in the 2nd century, possibly the second half, although a construction deposit in a temple dating to the second half of the 1st century AD might hint at older stone buildings.

Typical buildings

The very different levels of research in the three towns make it difficult to compare their public buildings. For, although the central parts of Cambodunum-Kempten and Brigantium-Bregenz are fully or almost fully known, excavations at Augusta Vindelicum-Augsburg have to be conducted in small sections because of the modern buildings, which results in features that are hard to interpret.


Forum and basilica

The fora of Augsburg, Kempten and Bregenz follow different design concepts and whilst those at Kempten (10,000 sqm) and Augsburg (9,400sqm) are similar in size, Bregenz's is markedly smaller (5,300sqm). The earliest forum is that at Cambodunum -Kempten which dates to the mid 1st century AD. Unusually, its basilica was initially a free-standing structure with no architectural links to the primary forum, but it was integrated into one of the forum's long sides during rebuilding in the early 2nd century AD. A temple occupied one of the narrow sides, whilst the other contained a further function room, perhaps the assembly chamber for the town council.

The forum of Augusta Vindelicum-Augsburg is only partially excavated, but it occupied the area of the former fort's main roads, possibly overlying the earlier principia. Parts of a two-aisled basilica have been found along the northern side of the square and the columns from the possible entrance area were covered in red plaster: a feature also encountered in Bregenz.

The forum of Brigantium-Bregenz was not just smaller than the other two complexes, it also lacks evidence for a basilica. It is noteworthy for large foundation slabs, which suggest the presence of (possibly equestrian) statues. This may suggest the presence of an imperial group, although this can not be confirmed by inscriptions. Within the courtyard a two roomed possible shrine was identified and heated rooms at the rear of the square probably served administrative functions.

Municipium Aelium Augustum-Augsburg. Features from the forum area and reconstructions of surrounding buildings.
Cambodunum-Kempten. The earlier period forum.
Cambodunum-Kempten. Later forum with walls of the earlier forum indicated.
Cambodunum-Kempten. Reconstruction of the later Forum.
Brigantium-Bregenz. Ground plan of the forum.


At least one large public bath is known from each of the Raetian central places. One of at least two periods was constructed in the 2nd century AD, in the north of the municpium Aelium Augustum-Augsburg, along with another to the west of the forum. The later is only known in part, and was originally interpreted as the residence of the provincial governor, but only the ground plan of a third set is known, and its dating remains unclear.

Municipium Aelium Augustum-Augsburg. Plan of the baths (Pettenkoferstraße).

Municipium Aelium Augustum-Augsburg. Plan of the baths (Georgenstraße).

Likewise, three baths are known from Cambodunum-Kempten. A smaller public bath, the so-called Thermenhaus, lay next to the temple district. It was constructed in stone around 40AD and contained a latrine as well as the standard bath-suite. These baths were rebuilt after a fire in the second half of the 1st century AD, and from then on were used for residential purposes. The Große Thermen were a 4,200 sqm complex, combining a bath suite with a large exercise area (palaestra). This was in use by the final quarter of the 1st century AD, and was redesigned in the mid 2nd century AD, when a new sequence of rooms was created and the size of the bath suite reduced by a fifth. The ‘Kleine Thermen’ were only accessible from the governor’s palace (praetorium) or the complex identified as a hostel (see below). Their use by the public is thus questionable, although an attached latrine was accessible from the road. The Kleine Thermen date to the mid 1st century AD and were later extended by the addition of a sweat bath/sauna.

Cambodunum-Kempten. Plan of the so-called Kleinen Thermen in its earlier (above) and later phases (below).
Cambodunum-Kempten. Plan of the so-called Großen Thermen in the second half 1st century AD (left) and from the 2nd century AD onwards (right).

The public bath of Brigantium-Bregenz was a (40 x 40m) building, surrounded by a wall, with nine rooms, some of which were heated. A two-aisled, free-standing portico with a large courtyard, stood adjoining these baths, but apparently without direct access, and it is unclear whether the two features were designed to operate in conjunction.

Brigantium-Bregenz. Plan of the public baths.



An emblematic 1st century AD building complex in Cambodunum-Kempten has traditionally been called the governor’s palace (praetorium) but after the presumed change of capital it supposedly continued as a hostel or inn. The complex adjoined the forum and its monumental entrance opened onto several large rooms and an internal courtyard, with a garden at the rear, flanked by further rooms. The Kleinen Thermen were part of this complex, and were only accessible from it, not from the street, but there was direct access between the complex and the basilica. Wattle and daub walls were later inserted in the front rooms, leading to a reduction in room sizes, and these are taken as indicators of a conversion to a hostel/inn.

Cambodunum-Kempten. The public buildings in the centre.
Cambodunum-Kempten. Reconstruction of the governor’s palace/hostel and the so-called Kleinen Thermen.

No hostels or inns can be identified in municipium Aelium Augustum-Augsburg, or Brigantium-Bregenz but, in Augsburg, a small structure with three rooms and a small bath building has been suggested as an inn or a station for beneficiarii. Present day thinking would doubt whether buildings previously suggested as inns in the west of Bregenz and the so-called Villa auf dem Steinhübel could have fulfilled this function.

Municipium Aelium Augustum-Augsburg. Plan of the putative inn. 2 = bath building, 3 = putative mithraeum.


Theatre and Amphitheatre

No theatres or amphitheatres are known, so far, in any of the Raetian town discussed here. The existence of a theatre in 2nd or 3rd century AD Augsburg might be assumed and the current lack of evidence might reflect the fragmentary nature of the excavations. At Bregenz and Kempten, however, where much larger areas of the towns have been studied, it is possible that no such entertainment venues existed.


A number of religious buildings are known in the administrative centres of Raetia. A rectangular structure on one of the narrow sides of the Cambodunum-Kempten forum is interpreted as a sanctuary, as is a two room building in the courtyard of the forum at Brigantium-Bregenz. In the latter, the existence of a capitolium temple in honour of the Roman gods Iupiter, Iuno and Minerva has also been considered proven through the presence of a tripartite temple of Roman design within a walled enclosure. This enclosure is probably also the find spot of a hoard of 80 brooches, which probably represent a mid 1st century AD construction deposit. Two possible podium temples are known from Augsburg, which date to the 2nd century AD and lay close to the former southern fort gate, and two structures to the south of the Kempten forum might have been podium temples with vestibules. The existence of a Mithraeum at Augsburg, which was apparently found close to the putative inn (see above), has not been confirmed.

Cambodunum-Kempten. Plan of the probable temple south of the forums.
Brigantium-Bregenz. Temple east of the forum.

At Cambodunum-Kempten, a 238 x 179m walled, sacred enclosure, with a central altar, lay directly south of the forum. Built around the mid 1st century AD, it was entered via a 50 m x 13 m monumental entrance. It has been connected with the imperial cult and may have been used by the provincial council, using the altar of Rome and Augustus at Lyon as a model, but this interpretation has not been confirmed by epigraphic evidence.
An open area in municipium Aelium Augustum-Augsburg has also been interpreted as a sacred space. Here, a large, gravelled area, has been identified, west of the former fort gate, which remained free of structures, except for several statue bases. It is debatable, however, whether the remains of a large enclosure wall, found north of the centre of Brigantium-Bregenz, also represented a sacred area.

Cambodunum-Kempten. Reconstruction of the sacred area with central altar.

Areas with Gallo-Roman temples have been identified at Cambodunum-Kempten and Brigantium-Bregenz. An enclosure wall with double portico on the western fringes of Cambodunum-Kempten contained two Gallo-Roman temples and an apsed, square temple, along with several smaller temples and shrines. The veneration of Hercules, Mercury and the Celtic goddess Epona, has been shown by inscriptions and statuettes. The stone buildings developed from the final third of the 1st century AD onwards but, from the second quarter of the century, the area had seen earlier occupation by rectangular wooden buildings, which were probably also temples. A small area with two Gallo-Roman temples and a shrine has been found to the east of Brigantium- Bregenz and an inscription found there was dedicated to all of the gods and goddesses.

Cambodunum-Kempten. Timber buildings in the temple area.
Cambodunum-Kempten. The stone buildings of the temple area.

City Walls

To date, only municipium Aelium Augustum-Augsburg has been proved to have a city wall but, according to the reports currently available, this was constructed in the 160s AD, albeit an older section supposedly survives in the north of the town. The projecting towers were only added in the 4th century AD and the area enclosed was smaller than the settlement as a whole, at 65 ha. The wall consisted of worked tufa, cladding a rubble and mortar core, and was fronted by two V-shaped ditches. City walls can probably be ruled out for Cambodunum-Kempten and Brigantium-Bregenz although, as demonstrated in neighbouring Upper Germany, this need not necessarily reflect on the towns' status.

Municipium Aelium Augustum-Augsburg. Section through the city wall and rearward buttress.


Timber-lined channels in the sacred area at Cambodunum-Kempten prove the existence of an aqueduct, but where this water came from and how it was distributed within the town, remains unknown. Likewise, we have insufficient information on the water supply of municipium Aelium Augustum-Augsburg for an aqueduct, postulated from aerial photographs, has not yet been studied in any detail.


The towns’ legal status

Roman legal status has only been confirmed for municipium Aelium Augustum- Augsburg but, as a municipium, this was a Roman chartered town, albeit current evidence does not allow us to decide whether it had Roman or Latin rights. There is no evidence for the legal status of Cambodunum-Kempten and Brigantium-Bregenz. Tacitus mentions the ‘exceedingly sumptuous colonia of the Raetians’ (splendissima Raetiae provinciae colonia), in his Germania (finished at the end of the 1st century AD), but fails to give its name. To judge from the level of urban infrastructure, this seems more likely to refer to Kempten than Augsburg, but this is far from certain. Moreover, there is nothing to indicate, that Tacitus used the term ‘colonia’ in its strict legal sense, rather than as a figure of speech, reflecting a certain distinctive level of urban design. An inscription from Aquincum-Budapest, which mentions a legionary veteran from Kempten whose father gained citizenship under Claudius, is sometimes quoted to support of this elevated status, but it does not prove that the town was a ‘colonia’ .

Aquincum-Budapest. Tombstone of the legionary veteran Tiberius Claudius Satto, from Cambodunum-Kempten.

A mid 2nd century AD inscription from Isny (see above) can be cited in support of the existence of several civitates in Raetia but, until further epigraphic evidence comes to light, the administrative situation must remain unresolved.

Information on the inhabitants

The little we know of the residents of the Raetian administrative centres is based largely on inscriptions from Augsburg and its hinterland, which include mentions of magistrates and members of the city council of municipium Aelium Augustum. An inscription from Aquincum-Budapest mentions a legionary veteran from Cambodunum-Kempten (see above).

Religious cults

Apart from the temples mentioned above, municipium Aelium Augustum-Augsburg has produced a number of inscriptions which shed light on the Raetian capital's religious life. These include references to the imperial cult, as well as other Roman and oriental deities. One inscription mentions the refurbishment of the temple of Mars and Victoria, and Sol, Elagabal and Isis testify to the presence of Eastern cults. Little can be said about the cults of Cambodunum-Kempten and Brigantium-Bregenz, thanks to their lack of inscriptions, except that the latter has produced an identification between Mercury and an indigenous deity.


Within the province of Raetia, only municipium Aelium Augustum-Augsburg can yet be safely identified as an administrative centre. This was a former garrison centre, which received its charter from the emperor Hadrian around 121AD, but it is not known whether the town had earlier served as the civitas capital of the Licates, of the upper R. Lech.

The surviving urban infrastructure of Cambodunum-Kempten and Brigantium- Bregenz suggests that both are likely to have served as administrative centres. The settlements were located in what had been the tribal territories of the Estiones and Brigantii in pre-Roman times, but their legal status remains uncertain. The mention by Tacitus of a colonia in the province of Raetia might refer to Kempten, but that can also not be proven. The existence of further civitates and thus administrative centres has to remain unresolved, although potential candidates include Curia-Chur, in the Alpine Rhine Valley, Phoebiana- Faimingen, and vicus Scuttarensium-Nassenfels in the north of the province.

Augsburg and Bregenz (two of the three settlements covered here) developed from military installations. Both also produced late Celtic material, although this does not allow predictions as to the nature of possible indigenous settlements. Kempten, by contrast, was a completely civilian foundation of the 10s AD, and had stone public buildings by the mid 1st century AD. This conversion from timber to stone and the provision of large public buildings happened much later (in the 2nd century AD) in Augsburg and Bregenz, despite equally early foundation dates. For this reason, it has been assumed that Cambodunum-Kempten served as Raetia's capital in the 1st century AD, and it even provides a distinctive building from that time which could have served as the governor’s residence. From the 2nd century AD, at the latest, however, the governor resided in municipium Aelium Augustum-Augsburg whose size of 80 ha (or 65 ha after the construction of the city defences) made it noticeably the largest administrative centre in the province. In this context, it is striking that Kempten’s monumental centre is unusually large by comparison with its residential quarters (its forum was larger than Augsburg's), which argues for a planned settlement that never reached its intended size. Thanks to the size of modern Augsburg, many of the Roman town's buildings are only known in fragments and are thus difficult to interpret but, of the three administrative centres, only it had defences, these being built in the 160s. The lack of amphitheatres and theatres may be due, at least in Augsburg, to the current state of research. The similarities between the sacred area of Cambodunum- Kempten and the altar of Rome and Augustus in Lyon, are exceptional, although a lack of epigraphic evidence prevents a final decision as to whether it fulfilled the role of a provincial assembly. Even if we take into account the markedly lower quality of the evidence, in comparison with neighbouring Upper Germany and Noricum, Raetia's much lower level of administrative centre development is striking.

Text: Thomas Schmidts


Select bibliography

Das römische Brigantium. Ausstellungskatalog des Vorarlberger Landesmuseums 124 (Bregenz 1985).

G. Alföldy, Latinische Bürger in Brigantium und im Imperium Romanum. Bayerische Vorgeschichtsblätter 51, 1986, 188-220.

L. Bakker, Augusta Vindelicum. Augsburgs Weg vom römischen Garnisonsort zur Hauptstadt Raetiens. In: Die Römer zwischen Alpen und Nordmeer (Mainz 2000) 88-94.

W. Czysz/K. Dietz/H.-J. Kellner/Th. Fischer, Die Römer in Bayern (Stuttgart1995).

Ch. Ertel, Landschaftsbezogenes Bauen in Brigantium. Jahrbuch des Vorarlberger Landesmusvereins 134, 1990, 63-86.

P. Gleischer, Topographisches zum antiken Brigantium. Montfort 37, 1985, 283-290.

G. Gottlieb (Hrsg.), Geschichte der Stadt Augsburg. 2000 Jahre von der Römerzeit bis zur Gegenwart (Stuttgart2 1985).

G. Gottlieb, Die regionale Gliederung in der Provinz Rätien. In: Raumordnung im Römischen Reich. Zur regionalen Gliederung in den gallischen Provinzen, in Rätien, Noricum und Pannonien. Schriften der Philosophischen Fakultäten der Universität Augsburg 38 (München 1989) 75-87.

G. Grabher, Bregenz/Brigantium. Archäologie Österreichs 5/1, 1994, 59-66.

G. Grabherr, Zur sogenannten Via Decia, Bayerische Vorgeschichtsblätter 61, 1996, 229-244.

H.-J. Kellner, Augsburg, Provinzhauptstadt Rätiens. In: Aufstieg und Niedergang der Römischen Welt II,5.2, 1976, 690-717.

W. Kleiss, Die öffentlichen Bauten von Cambodunum. Baubeschreibung und Rekonstruktion. Materialhefte Zur Bayerischen Vorgeschichte 18 (Kallmünz/Opf. 1962).

B. Overbeck, Raetien zur Prinzipatszeit. In: Aufstieg und Niedergang der Römischen Welt II,5.2, 1976, 658-689.

A. Schaub/L. Bakker, Zur Stadtentwicklung des römischen Augsburg. In: Genese, Struktur und Entwicklung römischer Städte. Xantener Berichte 9 (Mainz 2001) 177-189.

A. Schaub, Topographie und Stratigraphie des römischen Augsburg aufgrund neuerer Ausgrabungen. In: Römische Besiedlung zwischen Oberrhein und Enns. Schriftenreihe der Archäologischen Staatssammlung 3 (Grunbach 2002) 109-120.

W. Schleiermacher, Cambodunum-Kempten. Eine Römerstadt im Allgäu (Bonn 1972).

G. Weber (Hrsg.), Cambodunum-Kempten. Erste Hauptstadt der römischen Provinz Raetien? (Mainz 2000).

H. Wolff, Einige Probleme der Raumordnung im Imperium Romanum, dargestellt an den Provinzen Obergermanien, Raetien und Noricum. Ostbairische Grenzmarken 28, 1986, 152-177.

H. Wolff, Die verspätete Erschließung von Ostraetien und der Nordgebiete von Noricum – ein Forschungsproblem. Ostbairische Grenzmarken 30, 1988, 9-16.

H. Wolff, Die römische Erschließung der Rhein- und Donauprovinzen im Blickwinkel ihrer Zielsetzung. In: Römische Inschriften – Neufunde, Neulesungen und Neuinterpretationen. Festschrift H. Lieb (Basel 1995) 309-340.