NGRef: ST 469 905
OSMap: LR171/172
Type: British Civita, Capital

The "Civitas Silurum Stone" (RIB 311)
Resides in the Porch of Caerwent Church
W (6) to Bvlmore (Coed-y-Caerau, Gwent)
NE (30) to Glevvm (Gloucester, Gloucestershire)
NW (12.5) to Lydney Park
River Wye: N (16) to Blestivm (Monmouth, Gwent)

Venta Silurum - The Market Town of the Silures

The Roman name for Caerwent first appears in the Antonine Itinerary of the late-2nd century. Iter XIV of this document, "an alternate route from Isca [Silurum] to Calleva [Atrebatum]", lists a station named Venta Silurum, 9 miles from Isca (Caerleon, Gwent) and 14 miles from Abona (Sea Mills, Avon), this latter part of the itinerary likely represents a journey by ferry across the Bristol Channel to the seaport serving the thriving Romano-British spa-town at Aquae Sulis (Bath, Avon).

The town is also mentioned in the Ravenna Cosmology of the seventh century, this time as Ventaslurum (R&C#48), between the entries for Pontes (Staines, Surrey) and the unknown road-station Iupania. The name of the Roman town, Venta Silurum, means 'the market town of the Silures', and is partially carried through in the present name of the modern town Caer-went, which is Welsh for 'fortified place with a market'.

Epigraphic Evidence from Venta Silurum

A grand total of seven Latin inscriptions on stone are recorded for Caerwent in the R.I.B.. Two of these texts are only fragmentary, and another two, reading ABCD and CDRF (RIB 313 & 314) may have been teaching aids; the remaining three texts are all shown and translated on this page, the most famous being the so-called Civitas Silurum stone (RIB 311; depicted above right; text below).

The Civitas Silurum Stone

This stone appears to have been the base of a statue (now lost?), and bears an important inscription which confirms the existence of the Civitas Silurum or the Romanized self-governing body of the Silures tribe. The so-called 'Civitas Silurum Stone' is now on display in Caerwent church.

LEG[atvs] LEG[ionis] II
LEG[ato] AVG[usti] PR[o]PR[raetore] PROVIN[ciae]
PVBL[cae] CIVIT[atis]

"For Tiberius Claudius Paulinus, legate of the Second Augustan Legion, proconsul of (Gallia) Narbonensis, imperial propraetorian legate of (Gallia) Lugdunensis. By decree of the ordines for public works on the tribal council of the Silures."

RIB 311; statue base; dated: c.220AD;
see RIB 1280 at Bremenium (High Rochester, Northumberland)

The "Model Settlement"

The Roman town of Venta Silurum, the civitas capital of the Silures tribe, underlies the present town of Caerwent in Gwent. The main Roman road from the colonia at Glevum (Gloucester) to the legionary fortress at Isca Silurum (Caerleon) passed through the geometric centre of the settlement. Thankfully, when the route was widened in recent times to become the modern A48 trunk road, Caerwent was bypassed to the north and this historic site was preserved.

The town was founded c.75AD after the tribe had been finally subdued by the campaigns of Sextus Julius Frontinus. The majority of the remaining population of the tribe was gathered from the hills and valleys of Gwent and Monmouth and housed together in a new settlement at Caerwent, which site was obviously chosen in order to keep separate the subdued parts of the tribe from those still causing trouble further to the west. The routing of the road through the centre of the settlement was also deliberate, in order to let the people who were relocated see the benefits of Roman civilization which were paraded before their eyes each day along the military supply route.

The towns defences were comprised originally of a single massive earthen rampart and ditch, which enclosed an area of forty acres (eighteen hectares). These early defences were replaced in stone during the late second or early third centuries, together with an external double ditch. These later stone walls may have originally reached a height in excess of twenty-seven feet (eight metres).

Within the defences described above, a regular grid of streets divided the town into twenty insulae (literally 'islands') or City Block's, of roughly equal size. Several public buildings have been identified, as one may expect, clustered near the centre of the settlement. A forum and basilica take up the entire central insula to the north of the main road, and a public bath-house was situated in the north east corner of the insula opposite, fronting onto the southern side of the main road. A small amphitheatre was built to the north-east of the forum at a later date; it violates the neat pattern of the street grid, being built over the top of the northward internal road leading from the north-east corner of the forum complex.

The houses in the settlement were generally small and modestly appointed, with only a few in possession of tessalated or mosaic floors, hardly any having hypocaust heating or adjoining bath-houses. The Silures were, it seems, never to attain the refined Roman qualities and its associated wealth, as did their close neighbours the Dobunni from Gloucestershire and Hereford & Worcester, who were to build many villas in the hills surrounding their own civitas capital at Corinium (Cirencester, Gloucestershire). There is no such outlying pattern of villas surrounding the Silurian capital.

No military finds have been recovered from within the enclosure, which points to the town being founded on a virgin site.

Excavations at Caerwent/Venta Silurum

ST467907 - Excavations in 1971 on the north-west polygonal angle-tower revealed that this structure had been added to the already-existing town wall some time before the mid-4th century. A hoard of coins having a terminal date c.350AD was found beneath the cobbled floor of this tower.

The Caerwent Temples

Romano-British Temple 1


Located in a temenos beside the forum. This square temple has an "eccentric" apse on the north wall of the cella and buttresses on the outer portico walls which may be an indication of some height to the building, but it is essentially a temple of the Romano-British type. The outer portico measured c.46 x 43 ft. and the cella c.24½ x 23 ft., with all walls a uniform thickness of c.2 ft. There are indications that an earlier shrine was associated with the temenos while a Romanised "temple" was later built on the same site, post 265AD. The temple faced south. (Type Ic or Id)

Romano-British Temple 2

This octagonal temple lies in a temenos outside the town walls, just to the north of the east gate. The exterior "portico" wall was about 1½ feet thick and c.66 feet across, the diameter of the cella about 53 feet. The temple possibly faced east.

Possible Classical Temple 3

The only evidence at Caerwent for a temple built along classical lines is a solid podium built in the middle of the west side of the forum facing east. This possible temple base measures 35 ft. wide by 50 ft. in length and stands about 4 ft. high. Signs of possible vaulting in the superstructure suggest an important civic building on the site.

Altarstones to Warrior Gods and Heroes

Altar to Mars Ocelus

Altarstone of Mars Ocelus


"For the god Martius Ocelus,¹ Aelius Augustinus the Optio,² willingly and deservedly fulfilled his vow."

(RIB 310; altarstone; pictured on left)

  1. Mars Ocelus would appear to be a conflation of the Roman war god Mars with a Germanic warrior god named Ocelus. The name may mean 'Mars the Ocelenian', in reference to a demigod/hero of the Ocelenses tribe from Lusitania, named the Lancienses by Pliny (iv.22). Alternately the name may be a conflation with the ancient philosopher Ocellus of Lucania and should perhaps be read as 'Mars the Philosopher'. There is another stone dedicated to Mars Ocelus at Carlisle (RIB 949), dated to the 3rd century. Another stone bearing these names has been found here at Caerwent (vide RIB 309 infra).
  2. An optio was a senior N.C.O. in the Roman military, second in command to the centurion. However, the letters OP in this inscription may, for example, be expanded op[itulavit] 'he brought aid to', or simply op[eram] 'this work'; both phrases implying that Augustinus was perhaps responsible for renovation of part of the temple or its precinct. There are a number of alternative readings; the ending chosen was the one that I disliked the least (see arguments below).

A Roman Legionary Retirement Home?

Given the proximity of the legionary fortress it is not inconceiveable that a veteran optio, upon receiving his discharge at Isca, may have chosen to live close to his old legionary buddies in the settlement at Venta, especially if he had taken a native Silurian wife, for example. Optiones would not be particularly rich, and the lower-status houses here, within easy sailing distance of the luxurious spa town at Aquae Sulis only twenty miles or so across the Bristol Channel, would have been an excellent retirement home. It should also be noted that although not rich by Roman standards, Augustinus would have been a veritable Croesus in comparison with many of his Silurian neighbours, after all, he had spare cash enough to dedicate an altarstone to Mars (RIB 310 supra), and it may not be by chance that this stone is one of only a few so far discovered (i.e. there are no others to be found). In addition, his Roman citizenship and military training would have made him a natural contender for a place on the town council, especially if, as I have surmised, he took a native wife.

Statue Base Dedicated to Various Warrior Gods from Caerwent


"For the god Mars Lenus or Ocelus Vellaunus¹ and the Spirit of the Emperor, Marcus Nonius Romanus, due to the privileges granted him by the guild [of magistrates], dedicated this out of his own funds, ten days before the Kalends of September² when Glabrio and Homulus were consuls.³"

(RIB 309; statue base; dated: 23 August 152AD)

  1. Mars is, of course, the Roman war god. Ocelus and Vellaunus may have been late Iron-Age heroes or demigods (vide RIB 310 supra). Lenus, however, was a iron-age god of healing depicted on this stone with the feet of a goose (see The Gods of the Celts by Miranda Green, p.114 & fig.56).
  2. The date, ante diem decimum kalendis Septembris, was the 23rd August [it was also the day of Mercury, a Tuesday].
  3. Mamercus Acilius Glabrio Gnaeus Cornelius Severus and Marcus Valerius Homullus were ordinary consuls for 152AD (a.u.c.905).

What's to see Now

The principle visible remains are the complete circuit of the town walls, in places still over sixteen feet (five metres) high, the foundations of several Roman houses and shops to the west of the forum in Pound Lane, and a small Romano-British temple, octagonal in outline, which lies outside the eastern gate to the north of the road. Some finds may be viewed at Caerwent church, but the majority are on display in Newport Museum and the National Museum of Wales in Cardiff.

Pictures of Venta Silurum
Taken in late July 2006

The Basilica viewed from the NW

Industrial? Buildings viewed from the main road to the S

Caerwent Church houses the famous Civitas Silurum stone (see above)

South West Corner Angle

The Eastern Gateway with the Coach & Horses Inn in the background

A Multangular Bastion viewed from atop the S wall

The South Gateway note the drainage channnel through the blocked-up gate

The South Wall viewed from the SW corner looking E

The South Wall looking W

The South Wall looking E

Temple Temenos in the centre of the town looking SW

The Temenos looking NW

A Town-House with Hypocaust in the NW part of Caerwent looking S

The Medieval Tump or Castle Mound in the SE corner looking SW

The West Wall looking S

A House with Courtyard in the W part of town looking NW

Other Useful RBO Pages

Click here for the Romano-British Walled Towns page

Click here for the RBO Temples and Shrines Index


See: The Towns of Roman Britain by John Wacher (2nd Ed., BCA, London, 1995) pp.378-391 & fig.170;
Britannia iii (1972) p.302 & p.353;
Temples in Roman Britain by M.J.T. Lewis (Cambridge 1966);
The Roman Inscriptions of Britain by R.G. Collingwood and R.P. Wright (Oxford 1965);
Air Reconnaissance of Southern Britain by J.K. St. Joseph in J.R.S. xliii (1953) pp.81-97;
All English translations, including any inherent mistakes, are my own.
If you're thinking of visiting Cerwent (or anywhere else in Roman Britain for that matter) you'll be doing yourself a big favour by obtaining a copy of Roger J.A. Wilson's fabulous book A Guide to the Roman Remains in Britain (4th Edition, Constable, London, 2002) - I never leave home without it!
Link to maps of the area from: StreetMap Old-Maps MultiMap

Roman Name

Ravenna Cosmography: Ventaslurum; Antonine itinerary: 14-2 Venta Silurum

Page Citation: Kevan White (2018) "Roman Britain: VENTA SILVRVM"