The emergence of civitas capitals

Lees deze text in

Political structure before the arrival of the Romans

Special places

kaart1ROBAmersfoort_aUnlike in France and Belgium, no major population centres had developed in the Netherlands before the arrival of the Romans. We do however know of central cult sites that drew people from far and wide to make offerings to the gods. One example is the temple at Empel, which is examined more closely in the article on ‘Cults’. Weapons and other special objects were already being offered there in the Late Iron Age. In the Roman period a Gallo-Roman temple was built on the site, indicating that it remained an important place in the eyes of the population.

Leaders and followers

In the southern Netherlands political, social and economic relations in the Late Iron Age were determined by the retinue system (Gefolgschaft). This basically involved an individual with higher status (the tribal lord) providing gifts and favours for his subordinates and followers. They in turn would give him their support, perhaps paying a proportion of the yield from their land to him, and following him into battle when necessary.

River offerings

Virtually nothing is known of many of the population groups that lived in the southern Netherlands. We only really know about the social and political structure of the Batavians, as they are mentioned by several Roman writers, including Julius Caesar and Tacitus. Furthermore, the eastern river area – their territory – is one of the most intensively studied regions of the Roman Netherlands. The region was probably special in several respects. There was probably a slightly larger population centre at the spot where the Waal and Maas almost meet, near Kessel and Lith, which was already in use in the Late Iron Age. Potsherds and other surface finds have been found spread over a very large area, providing evidence of a large settlement. There was a shrine, where many weapons, military artefacts and even a large number of human bones have been found. These objects, and many others, were thrown in the river as offerings to the gods.

013kaartjeKesselLithACVUAmsterdam_a 014botmatKesselVUACVUAmsterdam_a

A mint in the river area

This large site was probably occupied by the Eburones. They lived in the eastern river area before the Batavians. As well as this large shrine at Kessel and Lith, the Eburones also had their own mint. This is unusual, as coins were not generally used for trading at that time, only in social intercourse.

Batavian nobility

012summusmagistratusRMOleiden_aWith a population centre and their own mint, the Batavians (and the Eburones before them) seem to have been an unusual phenomenon in this region. However, we do not know for sure if they really were unique, as we simply have too little information about the other population groups. Since the Batavians have been studied so extensively, we know about their political and social organisation. We know from historical sources that there was a Batavian nobility, some of whom took Roman names in the first century. However, they were not the only group with power among the Batavians. Flavus, son of Vihirmas, the ‘summus magistratus’ of the civitas Batavorum, laid an altar stone in the first half of the first century. His title roughly translates as ‘highest magistrate’ of the Roman administrative district of the Batavians. His name indicates that he had no Roman civil rights, which is unusual, as someone with such a high status in Roman society would tend to be a Roman citizen.

Formation of a people

So who exactly were the Batavians? Roman writers wrote that the Batavians split off from the Chatti tribe, who lived in the area to the east of the Rhine (now the German Rhineland). They settled in the Dutch river area, which according to Tacitus was almost entirely unpopulated at that time. Excavations have however shown that the area was never in fact completely empty. When the Batavians arrived there they encountered people living on isolated farms. Those farmers were what remained of the Eburones, who had not been entirely wiped out by Caesar. The newcomers appear to have had little impact on aspects of daily life such as the style of houses. It is therefore unlikely to have been a large group. It seems more likely that only an elite group of Chatti moved to the new area along with their entourages. These newcomers mixed with the old population, and together they became known as the ‘Batavians’. The society did however change considerably in the last decades BC and the first decades AD. The houses and settlements became a fixed feature in the landscape, as they were no longer moved by each new generation. The houses also became more solid, which meant they lasted longer. Whether this change was prompted by the newcomers is unclear, as the process appears to have begun earlier. The changes in Batavian territory can also be seen in other parts of the Netherlands. For more about these developments, see ‘The rise of a villa landscape’.


Another population group we have learned of from historical sources is the Cananefates. Tacitus tells us that they were related to the Batavians in ‘their origin, their language and their courageous character’. This suggests that, like the Batavians, the Cananefates originated in the Chatti tribe. However, the archaeological finds differ from those from the Batavian river area, suggesting that a very different people lived along the Dutch coast. The handmade pottery most closely resembles that of the Frisians (who lived in Noord-Holland and Friesland), or the pottery that was made further to the south. So the coastal region was populated by people who took their influences from both their northern and their southern neighbours. Perhaps the occupants of the Zuid-Holland coast were immigrants from these areas. Or maybe only a small group of Chatti did indeed emigrate to the coastal region of Noord- and Zuid-Holland. At any rate, the Romans called the entire population in this area the ‘Cananefates’.

Roman geography

The early Roman centre of the Cananefates, Lugdunum, is suspected to have been near Leiden or Katwijk. It has never been found, however. The name and location of Lugdunum are known, among other things, from a geographical description by the ancient writer Ptolemy (c. AD 100-178). He mentions the names and geographical coordinates of eight thousand places, including Batavodurum (near Nijmegen) and Lugdunum, which he refers to as the most important civilian settlements in the territory that is now the Netherlands. There are strong indications that Ptolemy drew his information from Pliny, who had described the situation just after the middle of the first century AD. Ptolemy therefore probably unwittingly also described a situation that already existed in the second half of the first century, and had persisted unchanged for all that time. After the Batavian revolt of AD 69-70 Lugdunum had to relinquish its status in favour of another town. From that point on the town we now know as Voorburg grew in stature until, somewhere between AD 82 and 90, it became the official civitas capital of the Cananefates.

Indigenous predecessors

kaart4ROBAmersfoort_aIt is difficult to pinpoint major Iron Age population centres. The settlements were all very similar, consisting of no more than three or four contemporaneous farms. However, with the division of the province into civitates (administrative districts based on tribal areas), somewhere between AD 82 and 90, the Roman rulers designated civitas capitals. These capitals were to become the political, economic and religious centre of the civitas. We know of two such civitas capitals in the Netherlands. Voorburg was home to the capital of the Cananefates (Forum Hadriani) and Nijmegen housed the Batavian capital (Ulpia Noviomagus). They were both new towns, at locations that had not previously performed a central function for the indigenous people of the area.

Voorburg: an older Cananefate settlement

Below the civitas capital in Voorburg, remains predating the actual town have been found. Only a few traces of this earlier indigenous settlement have been found. It was probably no bigger or more special than the average rural settlement and cannot therefore be regarded as the forerunner of the town. We do not know what the settlement looked like and what kind of buildings stood there. However, remains of palisades and a large number of isolated pits have been found. Palisades were widely used to enclose indigenous settlements. The most important evidence of an indigenous settlement are the finds dated to the first half of the first century. The majority of these finds are potsherds from handmade indigenous pottery. However, potsherds from finer Roman pottery and a fairly large number of fragments of wine amphorae have also been found. Such finds are often seen as indications of the presence of a local elite, though we do not know for sure whether this was the case in Voorburg. 

Nijmegen: Batavians and Romans as neighbours

From AD 70 the Romans decided to use an existing indigenous settlement, Batavodurum, as the civitas capital of the Batavians. Batavodurum lay in the low-lying west of the city, on the banks of the river Waal. Only a few remains of the settlement have been found to date. There was probably a shrine dating from the Late Iron Age, around which a settlement had grown up. To the east of the indigenous settlement, there was already a real town (Oppidum Batavorum) back in the Early Roman period. So there were two contemporaneous settlements in Nijmegen: an indigenous settlement called Batavodurum in Nijmegen-West and a town built by the Romans called Oppidum Batavorum on the higher-lying Valkhof. After the Batavian revolt, during which Oppidum Batavorum was razed to the ground, the indigenous settlement in Nijmegen was transformed into a new town which, in around AD 100, became the new capital of the civitas Batavorum.

Nijmegen rondom 0   Nijmegen rondom 100 AD  

Establishment of the civitas capitals


Present-day Voorburg was the administrative centre of the Cananefate area in the Roman period, and was best known as Forum Hadriani. The town probably began in the mid-first century AD as a small population centre beside Corbulo’s Canal. After the formation of the civitates (administrative regions) in the province of Germania Inferior, between AD 83 and 90, the town was designated a civitas capital. Somewhere between AD 120 and 151 the town was granted a city charter, thus gaining the right to call itself a ‘municipium’. From that point on, it was known as Municipium Aelium Cananefat(i)um (abbreviated to MAC). The name Forum Hadriani was used for much longer than Municipium Aelium Cananefatium. It was noted on a Late Roman road map, from which it was copied to a Medieval version of the map, and so handed down to us.

Phases of Forum Hadriani – Voorburg

until c. AD 47

Indigenous settlement

c. AD 47

Corbulo’s Canal dug. Construction of a (small?) population centre beside the canal, at the site of the town

c. AD 90

Voorburg becomes the official civitas capital of the Cananefates

from c. AD 121/122

Final layout of town with grid street pattern and residential blocks, possibly an initiative of emperor Hadrian.

between AD 121 and 151

City charter granted: new name Municipium Aelium Cananefatium. Also known as Forum Hadriani.

AD 138-161

Major building work in town.

c. AD 275

Town ceases to exist.


018keteloppbatavValkhofMuseumhetValkhofNijmegenThe first foundations of a capital for the Batavians were laid by the Romans, at a site in the centre of present-day Nijmegen (on and around the high ground of the Valkhof). This probably occurred around the beginning of the present era (before AD 10). The new town lay amongst several military camps which at that time played an important role in the military campaigns to the north of the Netherlands and Germany. The town, called ‘Oppidum Batavorum’, was set alight during the Batavian revolt in AD 69/70. The town was rebuilt, however, this time in the lower-lying western part of Nijmegen, where there was already an indigenous settlement known as Batavodurum. In the administrative division of the province at the end of the first century, the Roman town at Nijmegen was designated the capital of the Batavians. The town was granted market rights around the year 100, possibly earlier. It was then rechristened Ulpia Noviomagus Batavorum (‘the Ulpic market in the land of the Batavians’). Inscriptions also reveal another name for the town: Municipium Batavorum. This name indicates that the town had been granted a city charter, though it is not known exactly when. The name (Ulpia) Noviomagus is however by far the most commonly used to refer to the town at the site that is now Nijmegen.

kaart Nijmegen vrom
kaart Nijmegen mrom

Phases and locations in Nijmegen


Nijmegen-west (Ulpia Noviomagus)
Nijmegen-Valkhof (Oppidum Batavorum)
c. 10 BC-AD 69/70

Indigenous settlement: Batavodurum. With cult site by river

Oppidum Batavorum

AD 69-70

Batavian revolt. Consequences for Batavodurum unknown

Batavian revolt: Oppidum Batavorum destroyed

c. AD 70-c. AD 100

New town in Nijmegen-West: (Ulpia) Noviomagus. Replaced Oppidum Batavorum

Small-scale reconstruction

from c. AD 100

Market rights (and city charter?) granted: layout of town finalised, with grid street pattern and public buildings. Also known as Municipium Batavorum


c. AD 170

Major fire, followed by partial reconstruction


Construction of towns

Town plan

Careful planning

019kaartforumhadrianiROBAmersfoortM_aRoman towns were generally very carefully planned, particularly the larger and more important ones. The Roman town in Voorburg
known by the name Forum Hadriani, was probably just a modest affair in the first century. Buildings stood on either side of a main road, and probably along a few side roads. But when the town was chosen as the capital of the Cananefates, all that had to change. The Romans decided to replan the town from scratch. From the beginning of the second century, a planned town began to emerge. The town boundary was determined beforehand and defined by a ditch.

Islands in the town

The layout within the town boundaries was very ordered. The streets were laid out on a grid pattern, with blocks of houses that archaeologists call insulae (literally ‘islands’). The most important streets were the two main streets that crossed more or less in the middle of the town. The route of one of these two main streets in Forum Hadriani (from east to west) is known. The road was originally some 13 metres wide, with a pavement along each side. The wooden road surface, remains of which have been found, was not as wide as the road itself (at only six metres or so), as there was probably a dirt track on either side for horses and livestock. Remains of a sewer have been found beneath the wooden road surface.

New towns in Nijmegen

Like Forum Hadriani, the Roman towns in Nijmegen – first Oppidum Batavorum and later Ulpia Noviomagus – were carefully planned. Oppidum Batavorum was a new town, built at the beginning of the first century. Finds made there to date suggest that the main road running east to west was densely built-up on both sides. When the capital was moved, however, an indigenous settlement was chosen as the new location. Although this settlement had grown into a town by the end of the first century, as in Forum Hadriani there are signs that it underwent major changes. An orderly grid pattern was introduced, which can best be seen at the site of the large temple complex on Maasplein. When the temple complex was built the existing houses – mainly potters’ workshops – were demolished. A new street plan was immediately laid out, some of it over the foundations of the old buildings.

Town wall

Neither Forum Hadriani and Ulpia Noviomagus was enclosed by a wall in its early existence. A ditch and, at most, a rampart were the towns’ only defences. The first half of the second century was very peaceful, and the towns were under no threat. Around the middle of the second century a palisade was built in Forum Hadriani, with a large moat in front of it. But it was not until the second half of the second century that a town wall was constructed around both Forum Hadriani and Ulpia Noviomagus. After the wall was built, some of the existing buildings in Ulpia Noviomagus lay outside the town boundaries.

In Forum Hadriani two gates in the town wall have also been found – a large gate on the main street, and a smaller one elsewhere. In front of the large main gate there was a bend in the road, which forced any attackers to walk parallel to the wall with their unprotected right side exposed (shields were carried on the left). Two sets of heavy foundations suggest that some other structure had previously stood on the site of the gate, possibly a triumphal arch.

Building phases: from wood to stone

Stone was not routinely used for building until the second century. Natural stone had to be imported from far away, as there was no stone suitable for building in the Netherlands. The first example of the use of stone and roof tiles (in the Netherlands) has been found in early Oppidum Batavorum. In autumn 2005 the stone foundations of a large building that must have been built around AD 40 were revealed. Around AD 100 a large number of large public buildings were constructed in Nijmegen, partially of stone, with tiled roofs. The buildings were probably put up with the help of the army. The increasing use of stone in the town can be seen in the other buildings in Ulpia Noviomagus. New houses with a different layout and heavier foundations were built where wattle and daub houses had once stood. The new structures themselves were generally made of wood, however. First the cellar would be reconstructed in stone, then a stone extension might be added, which had the great advantage of being more fire-resistant.

nijmegen oudste huis
nijmegen stenen kelder


Public buildings

A number of public buildings have been found in the two Dutch civitas capitals. In Forum Hadriani only the remains of a bath house have been excavated. More public buildings have been investigated in Nijmegen, though some of them were not in the town itself. This unusual situation arose out of the fact that a large legionary fortress (a ‘castra’) big enough to house 6000 soldiers was situated near the town. There were therefore a number of public buildings used by the soldiers in the nearby civilian settlement (the ‘canabae’). An amphitheatre, forum and inn have been found in Nijmegen. At the site of Ulpia Noviomagus itself, only the remains of a bath house and a large temple complex have so far been found.

Bath house

025BadhuisForumHadrianiBoth Ulpia Noviomagus and Forum Hadriani had a public bath house for the citizens of the town. Since the bath house in Forum Hadriani is the best known, it will be used as an example here. It was excavated back in the early 19th century by the world’s first professor of archaeology, C.J.C. Reuvens. He meticulously recorded his finds and even measured the height of the features, which was very unusual for that time. Since he had so little material for comparison, it was difficult for him to interpret the features. But his precise records and beautiful drawings have allowed them to be interpreted more recently. The bath house was probably built between AD 120 and 150. The remains consist mainly of stone foundations, fragments of wall paintings and other finds. Little remained of the standing walls. It appears to have been a bath house of the ‘row type’, like the bath house in Heerlen. On leaving the changing room one would first enter the frigidarium, the cold plunge pool. Beyond that lay the tepidarium, where the temperature was more pleasant. The final pool in the row was the caldarium, the hot water bath. The water in this room was heated by a furnace behind the caldarium. The bathing rooms were along one side of an inner courtyard. On the other side there was probably a water tower. The walls of this structure have heavier foundations, and buttresses were even added at a later stage to support the great weight of the water. There was also a large public toilet in the building, which was probably also used by people entering from outside the bath house.

23 nijm forumThus far, a forum (market hall) has been found only in Nijmegen. It was situated in the canabae, the settlement beside the legionary camp. The people of Noviomagus would probably also have bought and traded here. The forum in the canabae of Nijmegen consisted of a building covering more than two hectares, surrounding an open courtyard. Around this courtyard was a colonnade, beyond which lay shops and storerooms. There was a large hall (basilica) at the back of the building where wholesalers plied their trade. The eastern wing had a floor supported by wooden posts, with a space beneath, probably designed to combat damp. The courtyard, covering a hectare, also had a wooden floor, supported by more than 50,000 wooden posts. In the courtyard itself the remains of large rectangular foundations consisting of roof tile fragments have been found. This was probably where two large statues stood opposite each other in the centre. The large market hall was built at the same time as the legionary fortress was being rebuilt in stone, about AD 100. It is not clear how long the building stood. Shortly after construction work began on the forum the tenth legion left Nijmegen. There are indications that the building was still in use in the second century. Opposite the market hall was an industrial district, where a number of buildings, workshops of varying sizes and a granary have been excavated.

026nijmamfitheaterrecBobBrobbelHilversum_aThe only known Roman amphitheatre in the Netherlands was in the canabae of the Nijmegen legionary camp. It was probably very similar to the one in the town of Xanten (Colonia Ulpia Traiana). The amphitheatre consisted of an oval arena separated from the seating area by a wall made of tuff. The seating itself stood on the sand dug out to create the arena, which measured some 58 x 46 m. In the centre there was a square underground area (hypogeum) from which wild animals, for example, were driven into the arena. There was seating for some five to six thousand spectators – more than the estimated total population of Noviomagus. The theatre was therefore originally built for the legionary soldiers, which must have numbered in the thousands. After the legion withdrew, in around AD 104, the amphitheatre remained. Finds made at the amphitheatre suggest that it remained in use into the third century, indicating that it was also intended for the population of the town.

028nijmtempelsRemains of a very large temple complex have been found on the Maasplein in Nijmegen. It consisted of two adjacent temples for the worship of Fortuna and Mercury, both of whom signified wellbeing and prosperity. However, Fortuna was intended mainly for the female section of the population, and Mercury for the male. The temples were built according to the Gallo-Roman principle. They both consisted of a square walled area. In the middle a small square building stood on a platform, with a staircase on the east side. This was the actual shrine, known as the cella. There were a number of other shrines inside and outside the temple walls, possibly to other gods, and galleries containing several rooms on the north and south sides of the complex. One of these rooms on the south side was probably used for performing rituals. The faithful might have been able to buy offerings for the gods or trinkets in the other rooms. Temples of this construction have also been excavated elsewhere in the Netherlands. See the ‘Cult’ section on this website.

029mansioArcheonA fairly large number of traces of the Nijmegen inn (mansio or praetorium in Latin) in the canabae have been found, allowing a replica to be built at the Archeon archaeological theme park in Alphen aan de Rijn. The inn consisted of an inner courtyard with rectangular buildings, probably several stories high, divided into small rooms on the two short sides. There were probably covered galleries on the long sides of the courtyard.


The main material used for building in the Netherlands throughout the Roman period was wood. Walls were covered in daub, like the half-timbered houses that still exist here and there in the Netherlands and neighbouring countries. Buildings in town were often rectangular, with the short side facing the street. The same type of houses are also found in smaller rural settlements. Features found in Forum Hadriani indicate the position of a row of rectangular houses and their internal layout. In the yards behind the houses there were wells. A row of pillars stood before the row of houses, supporting a roof that covered the pavement along the entire street. One house found in Noviomagus was larger than the rectangular town houses. This house more closely resembled the main buildings of the large farmsteads (villae) found in the countryside. There would most probably have been larger houses in Forum Hadriani, too. Unfortunately, however, we do not know what they looked like, as the traces have not yielded enough information.

insula huizen
nijmegen huis maasplein

Legal status of towns

Different types of town

There were several types of town in the Roman provinces. The differences between them were determined by their legal status. One type was the coloniae founded by the Romans. These new towns were intended to provide the Romans with a familiar home in foreign territory. The residents of coloniae enjoyed Roman civil rights, which gave them many privileges. There were no Roman coloniae in the present-day Netherlands, however. The nearest was over the border near Nijmegen, in the vicinity of Xanten, Germany: Colonia Ulpia Traiana. Parts of this colonia have been reconstructed at Xanten Archaeological Park.

Municipia had a slightly lower status. A settlement granted a city charter by the emperor had the right to call itself a municipium. The inhabitants of municipia could also become Roman citizens, although they did not gain citizenship automatically. Eventually, both the capital of the Cananefates (in Voorburg) and the capital of the Batavians (in Nijmegen) were granted city charters, thus entitling them to call themselves municipia. Sometimes the emperor would grant only market rights, which would allow a town to consolidate its position as a commercial centre.

Finally, there were towns with no special status. They had not been officially granted a city charter or market rights, though this did not necessarily stop them from growing into important centres in their region. Dutch archaeologists generally refer to such towns with no special rights or status as vici. This was a general term in the Roman period, and tells us nothing about the size of the settlement, as it might refer equally to a village or town. We therefore prefer to use the term ‘rural centre’ to refer to towns with no legal status.


Municipal rights for Voorburg

The Cananefate capital received a city charter somewhere between AD 120 and 151, from emperor Hadrian or his successor Antoninus Pius. From that point on it was allowed to call itself a municipium, and it changed its name to Municipium Aelium Cananefat(i)um (abbreviated to MAC). It is not clear whether the name Forum Hadriani (which refers to the emperor Hadrian) predates MAC. It could be that Hadrian himself granted the town market rights (hence the name Forum Hadriani) when he was on a working visit to the border of Germania Inferior, probably in AD 121/122. However, it is just as likely that both names were used simultaneously. We do know that the name Forum Hadriani was used for longer. It can be found on a road map from the Late Roman period, and on a Medieval copy of this map. However, no actual Roman inscriptions bearing the name have ever been found.

Mijlpalen naar de stad van de Cananefaten


The abbreviation MAC is mentioned on milestones found in the vicinity, which give the exact distance to the remains of the town in Voorburg. The inscription on one of the milestones translates as follows:

‘To the emperor Caesar, the divine Hadrian's son, the divine Traian Parthicus' grandson, the divine Nerva's great-grandson, Titus Aelius Antoninus Augustus, high priest, with tribunicial powers, the pious, for the fourth time consul.’
The final row states the distance to the nearest town: ‘A MAC MP IIII’, which translates as ‘To the Aelian town of the Cananafates, four miles’.

The abbreviation MAC is therefore believed to stand for Municipium Aelium Cananefatium. The name Aelium might refer either to emperor Hadrian or to his successor Antoninus Pius.


A new market in Nijmegen

031aureusTraianusRMOLeiden_aThe name of the town in Nijmegen also changed over time. The capital of the Batavians eventually became best known as Ulpia Noviomagus, sometimes with ‘Batavorum’ added. This name has been found in several inscriptions. The addition of the word Ulpia has prompted many to draw the conclusion that the town was granted market rights by emperor Marcus Ulpius Traianus some time around AD 104. However, the name might already have changed soon after AD 70, when the town was moved to the west of the current city of Nijmegen following the destruction of the old one during the Batavian revolt. In that case, the name Noviomagus (‘new market’) would seem highly appropriate, the name ‘Ulpia’ possibly being added later, during the reign of Traian.

Nijmegen as a municipium

We know that the capital of the Batavians also received a city charter, as the name Municipium Batavorum appears in inscriptions. However, we do not know when it was granted this status, though it is generally believed to have been in the course of the second century. However, it might also have been earlier, under emperor Traian. Around AD 104 (in the reign of Traian) 6000 troops left Nijmegen. They were no longer needed in the peaceful province of Germania Inferior and were redeployed in recently conquered Dacia (Romania). Only a small military unit remained behind in Nijmegen. This must have hit the local economy hard. It is generally assumed that Traian therefore granted the civilian settlement market rights to boost its economy. It is also equally possible that he granted it not only market rights but also a city charter. If Nijmegen really did receive a city charter in AD 104, that would make it the first municipium in the Netherlands. Whether or not the town was allowed to call itself a municipium, the name (Ulpia) Noviomagus was clearly most commonly used.

Names of the Roman towns in Nijmegen

0-AD 69/70


Indigenous settlement in Nijmegen-West (later became Ulpia Noviomagus)

c. 10 BC-AD 69/70

Oppidum Batavorum

First Roman settlement in Nijmegen, situated on the high ground at the Valkhof in the centre of present-day Nijmegen (to the east of Batavodurum)

from c. AD 69/70

Ulpia Noviomagus

Civitas capital in Nijmegen-West (on the site of the old Batavodurum)

c. AD 100?

Municipium Batavorum

Official name of Ulpia Noviomagus after granting of city charter


The inhabitants


Towns were where artisans, politicians and civil servants lived and worked. We will never know the names of the majority of the people who lived in the towns. Only those who could read and write – mainly the higher social classes – have left behind their names for posterity. The others usually left no more than the remains of their home or workplace.

32hazewindhondRMOLeiden_aPotsherds, clothing accessories and art objects give us a glimpse of daily life in a Roman town. Butchering waste, potters’ wasters or pits containing glass or metal intended for reuse tell us something about the artisans who lived there. Many veterans probably also lived in the towns, moving there once their military service was over, though no inscriptions referring to veterans living in town have ever been found in the Netherlands.

Cobblers and eye doctors

33schoenmakersRingMuseumHetValkhofNijmegen_aThe capital of the Batavians is also referred to on gravestones, altars and other inscriptions, thus providing us with the names and professions of a number of residents. One fine example is a silver ring on which an inscription has been impressed using a pointed object. It translates as: ‘To the goddess Salus hath Rusticus bestowed and dedicated (this ring) for the cobblers of Noviomagus (‘Sutoribus Noviomagensibus’) of the cult association of Esseravus’.

34OogartsRMOLeiden_aA seal belonging to an eye doctor has been found in Nijmegen, where he undoubtedly practised. His name – Marcus Ulpius Hercules – is inscribed on the seal. On the sides are the names of the medicines the seal was used to stamp: ‘melinum’, ‘tipinum’, ‘diarices’ and ‘diamysus’. There were various professional associations and other societies in the towns. A stone tablet found in Nijmegen, for example, bears an inscription that refers to the trade association of the carpenters.


The names and deeds of politicians have also been immortalised in inscriptions. To enter politics, one had to have plenty of money, since one would be expected to pay for most things oneself, and also do the odd good deed. We know, for instance, of an official (decurio) from Xanten who paid for the bath house in Heerlen to be built. The names of officials from Voorburg and Nijmegen are also known. Valerius Silvester, a decurio from Municipium Batavorum (Nijmegen), for example, dedicated a small altar to the Germanic goddess Hurstrga.

Signs of cults

Believers in Nijmegen

35tempelsMaaspleinBerryvanHarenThere is fairly good evidence that several faiths were practised in the Dutch civitas capitals. The two temples on the present-day Maasplein in Nijmegen are clear examples. The goddess Fortuna and the god Mercury were worshipped in the two largest temples. These were gods from the Roman pantheon. Around these two main temples stood scores of smaller temples in which other gods (some of them indigenous) would have been worshipped, although we do not know exactly which. A large piece of a votive pillar has also been found. This carved pillar was erected in around AD 17 to celebrate a victory. The emperor and his general were depicted on the pillar, alongside Roman gods.

A domestic altar in Voorburg

During excavations at Forum Hadriani in 2005 a number of indications of the religious life of the town’s inhabitants were found. A small shrine was often set up in the home, perhaps housing a domestic altar and statues of the gods. During the excavations fragments of at least eight pipe-clay figurines were found. One of the figurines depicted a seated mother goddess with a dog in her lap. A small domestic altar was also found. According to its inscription, it was dedicated to Jupiter.

Personal expressions of faith

036altaarhurstrgaThere are further indications of the religious life of townsfolk. Votive altars often display the name of the person who erected it and for which deity. One official (decurio) from Municipium Batavorum, for example, still had ties with his Germanic origins, as evidenced by the altar which, despite his high social status – suggesting he was to a large extent Romanised – he dedicated to the Germanic goddess Hurstrga.


A capital for the people

With the division of the province into civitates (administrative units), the Roman rulers also designated civitas capitals. These capitals were to become the main political, economic and religious centre of the civitas. We know of the existence of two of these capitals in the Netherlands: the capital of the Cananefates (Forum Hadriani) in Voorburg and the capital of the Batavians (Ulpia Noviomagus) in Nijmegen. Both these towns were built and designated by the Romans. They were probably not at the site of an old tribal centre. Indeed, it is by no means clear whether there were any real population centres before they were designated as such by the Romans.

Roman urban planning in Voorburg

Both the civitas capitals in the Netherlands had all the features of a Roman town. The streets were laid out in a perpendicular grid pattern and several large public buildings were located in the town. The Roman town in Voorburg, Forum Hadriani, was neatly arranged into blocks of houses. Some of these blocks have been partially excavated. The excavations revealed that the people lived in a kind of terraced housing with long yards behind. The houses were long and narrow, and built of wattle and daub. Wells and waste pits were found in the back yards. Remains of a bath house have also been found in Forum Hadriani. Other substantial sections of wall bore testimony to the fact that other large buildings must have stood there, though it is not clear what type of buildings they were.

Different towns in Nijmegen

Nijmegen has a long history. Back at the very beginning of the Roman period the Romans established a capital there. Built on a high point in the landscape, it was known as Oppidum Batavorum. The houses there were typical town houses, and some even had foundations or a cellar built of stone, something found nowhere else in the Netherlands so early in the Roman period. This town was above all a Roman initiative. During the Batavian revolt in AD 69-70 the rebels burnt down the entire town. It would be impossible to rebuild it, and the Romans decided to build a new capital for the Batavians at the site of an existing indigenous settlement. The settlement – known as Batavodurum – probably already existed at the time of Oppidum Batavorum. Prior to the Batavian revolt, therefore, a complete Roman town and an indigenous village stood just 1.5 kilometres apart. The new town was granted market rights by emperor Marcus Ulpius Traianus, hence the town’s new name: Ulpia Noviomagus Batavorum (‘the Ulpic market in the land of the Batavians’). Having also been granted a city charter, the town was also allowed to call itself Municipium Batavorum. However, the name Ulpia Noviomagus is the best known. Remains of a bath house and a large temple complex have been found in Noviomagus. A number of large buildings were not in the town itself, however, but in the large settlement (canabae) outside the legionary fortress. Despite the fact that most of the soldiers left in the early second century, the amphitheatre in the settlement remained in use.

Defending the town

Town walls were built around both Noviomagus and Forum Hadriani in the second half of the second century. Both towns appeared to face almost insurmountable difficulties in the mid-third century, driving many of the townsfolk away.


Margje Vermeulen-Bekkering

References (selection)

Bechert, T., 1983, De Romeinen tussen Rijn en Maas. Dieren

Dockum, S. van, 1993, Romeins Nederland : archeologie & geschiedenis van een grensgebied. Utrecht

Es, W.A. van, 1981, (3e herziene druk), De Romeinen in Nederland. Haarlem

MUST & B. Colenbrander (red), 2005, Limes Atlas. Rotterdam.

Tacitus - Germania. In de vertaling van dr. J.W.Meijer, 1992. Baarn

Tacitus - Historiën; Vierde boek. In de vertaling van dr. J.W.Meijer, 1991/1995. Baarn

Bloemers, J.H.F., 1978, Rijswijk (Z.H.), "De Bult". Eine Siedlung der Cananefaten. Amersfoort.

Bridger, C., 2000, Zur römischen Besiedlung im Umland der Colonia Ulpia Traiana/Tricensimae. In: Germania inferior. (Ergänzungsbände zum Reallexikon der Germanischen Altertumskunde Band 28). Berlin/New-York. 185-211

Enckevort, H. van, 1991, Kelten, Germanen en Romeinen. In: J.H.F. Bloemers & T. van Dorp, Pre- & protohistorie van de lage landen. Houten. 265-276

Heinrichs, J., 2000, Römische Perfidie und germanischer Edelmut? Zur Umsiedlung protocugernischer Gruppen in den Raum Xanten 8 v.Chr. In: Germania inferior. (Ergänzungsbände zum Reallexikon der Germanischen Altertumskunde Band 28). 54-92 Berlin/New-York.

Roymans, N., 1990, Tribal societies in Northern Gaul. An anthropological perpective. (Cingula 12) Amsterdam

Roymans, H.G.A.M, 2004, Ethnic identity and imperial power. The Batavians in the early Roman empire. Amsterdam

Swinkels, L. (ed), 2004, De Bataven. Verhalen van een verdwenen volk. Amsterdam/Nijmegen

Willems, W.J.H., 1986, Romans and Batavians a regional study in the Dutch eastern river area (originally Published in: Berichten van de Rijksdienst voor het Oudheidkundig Bodemonderzoek 31(1981) and 34 (1984))

Wolters, R., Germanische Mobilität und römische Ansiedlungspolitik: Voraussetzungen und Strukturen germanischer Siedlungsbewegungen im römischen Grenzland. In: Germania inferior. (Ergänzungsbände zum Reallexikon der Germanischen Altertumskunde Band 28). Berlin/New-York. 146-168

Bazelmans, J.G.A. en D.H. de Jager, 2006, Forum Hadriani. Van Romeinse stad tot monument. Utrecht

Buijtendorp, T., 1982, Een Romeinse stad bij Voorburg-Arentsburg. Een interpretatie. Westerheem 31, 142-163

Buijtendorp, T., 1984, Nieuw onderzoek naar Forum Hadriani. Westerheem 33, 1984, p. 116-124

Buijtendorp, T., 1987, Periodisering van Romeins Voorburg. Westerheem 37, 1988, 107-117


Enckevort, H van, et al, 2000, Nijmegen. Legerplaats en stad in het achterland van de Romeinse limes. Abcoude

Enckevort, H. Van & Thijssen, J., 2001, Der Hauptort der Bataver in Nijmegen im 1. Jahrhundert n. Chr. Von Batavodurum und Oppidum Batavorum nach Ulpia Noviomagus. In: Gundolf Pecht (Hrsg), Genese, Stucktur und Entwicklung Römischer Städte im 1. Jahrhundert n. Chr. in Nieder- und Obergermanien (Xantener Berichte. Grabung - Forschung - Präsentation Band 9) Mainz

Enckevort, H. Van & Thijssen, J., 2003 - Nijmegen und seine Umgebung im Umbruch zwischen Römerzeit und Mittelalter. In: Kontinuität und Diskontinuität. Germania inferior am Beginn und am Ende der römischen Herrschaft. (Ergänzungsbände zum Reallexikon der Germanischen Altertumskunde 35) Berlin/New York. 83-118

Enckevort, H. Van & Thijssen, J., 2003, Nijmegen - A Roman town in the Frontier Zone of Germania Inferior. In: Pete Wilson (Ed.), The archaeology of Roman towns. Studies in honour of professor John Wacher. Oxford. 59-72

Haalebos, J.K., 1990, Neues aus Noviomagus. Archäologisches Korrespondenzblatt 20. 193-200

Koster, A., K. Peterse & L. Swinkels 2002, Romeins Nijmegen boven het maaiveld. Reconstructies van verdwenen architectuur. Nijmegen.

Willems, al (Ed.), 2005, Nijmegen, Geschiedenis van de oudste stad van Nederland. Dl. 1: Prehistorie en Oudheid. Wormer

Status of the towns:
Hiddink, H.A., 1991, Rural centres in the Roman settlement system of Northern Gallia Belgica and Germania Inferior. In: N. Roymans & F. Theuws (ed.), Images of the past: studies on ancient societies in northwestern Europe. (Studies in pre- en protohistorie 7). Amsterdam. 201-234

Waasdorp, J.A. 2003, IIII M. P. naar M. A. C. Romeinse mijlpalen en wegen. (Haagse oudheidkundige publicaties nr. 8) Den Haag.


Archeologienet (dutch)
Cultuurwijzer (dutch) (dutch)

Imperium Romanum (in German) (in English)