Camboglanna (Castlesteads) Fort

Camboglanna was a Roman auxiliary fort in north-west England and was part of the chain of fortresses on Hadrian’s Wall.

1. The Hastatus Posterior was the fourth-highest ranking centurion in a Roman Legion, usually serving in the First Cohort but in this instance seemingly placed in temporary command of the Fourth Cohort during construction work at the Castlesteads fort. The name of the legion in which Marcianus served is not recorded.

The small fort at Castlesteads is thought to have been added to the fortifications on Hadrian’s Wall at the same time as the Vallum was built. The fort is unique, as it is the only garrison fort which is detached from the Wall itself and lies a short distance to the south, between the Wall and the Vallum. The fort’s northern defences are roughly aligned with the line of the Hadrianic barrier, which here bows out north-westwards away from the direct line taken by the Vallum, which here runs from east-north-east to west-south-west.

The Wall fort is thought to have been preceeded by another fort of turf-and-timber built upon a different alignment, traces of which have been found by excavation beneath the south-eastern Hadrianic foundations during the 1930’s. It appears likely that the presence of this earlier encampment may have influenced the placing of the Hadrianic fort away from the line of the barrier wall.

The north-western side of the fort platform has succumbed over the passage of time to the eroding effects of the Cam Beck, giving the surviving platform a rectangular outline, although it is thought that Camboglanna was originally square in plan, measuring about 400 feet (c.122m) on each side and covering an area of about 3¾ acres (c.1.5ha).

The remains of an extra-mural military bath-house were excavated beside the Cam Beck about 220 yards (c.200m) north of the fort in 1741. It is thought that the fort’s parade ground may have been in the area about 330 yards (c.300m) north of the fort, just south of the site of Turret 56b, where a number of martial altarstones have been recovered (namely RIB 1979, 1981 and 1991).

Classical References to Castlesteads

The Roman names of the fort at Castlesteads and the neighbouring fort to the east along the Wall at Birdoswald, have been the subject of intense debate over the years. There are a number of ancient documentary and epigraphic sources which name these forts: the Ravenna Cosmography, the Notitia Dignitatum, a couple of Roman Souvenirs and an indirect reference on an Altarstone to Silvanus.

The Ravenna Cosmography places an entry named Banna between Esica (Great Chesters, Northumberland) and Uxelludamo (Stanwix, Cumbria). We know from archaeological evidence that there are two large forts on the Wall between these sites, namely Castlesteads and Birdoswald, but unfortunately there is nothing in the Cosmography to indicate which of these two forts is to be identified with the Banna entry.

The waters of investigation are further muddied by the appropriate section of the Notitia Dignitatum, which lists between Magnis (Carvoran, Northumberland), and Petrianis (Stanwix, Cumbria) the entry: Tribunus Cohortis Primae Aeliae Dacorum Amboglanna, Tribune of the First Aelian Cohort of Dacians at Camboglanna. Again, we could not be sure whether the Amboglanna entry refered to the fort at Birdoswald or the fort at Castlesteads.

The question seemed to have been settled with the discovery of a decorated bronze drinking-vessel at Rudge in Wiltshire and another similar vessel at Amiens in France. These artifacts, thought to have been Souvenirs of the Wall, list the names of places along the line of Hadrian’s Wall from west to east, they were: Uxelodunum (Stanwix, Cumbria), Camboglan[ni]s, Banna and Esica (Great Chesters); these pieces of souvenir cookware indicated that the name of Castlesteads fort was Camboglanna and that of the Birdoswald fort Banna.

The initial lack of epigraphic evidence from either of these two sites meant that this view became firmly ingrained into the history books produced over much of the twentieth century. It was not until the latter part of the last century that extensive excavations were conducted at the Birdoswald site and these unearthed evidence that the garrison was the First Aelian Cohort of Dacians, which is the unit associated with the name Camboglanna in the N.D.. This discovery seemingly overturned the evidence from the Rudge and Amiens artifacts and once more opened the discussion over the naming of these two forts.

Adding further fuel to the Banna / Camboglanna controversy, an altarstone to Silvanus the Roman god of the forest (RIB1905) was subsequently found at the Birdoswald fort, dedicated by a group calling themselves the Venatores Bannienses, or ‘the Hunters of Banna‘. It is very likely that these men formed an irregular auxiliary cavalry unit garrisoned at Birdoswald sometime during the fourth century, from which posting their unit name was derived.

The eminent Roman historian M.W.C. Hassall in 1976 suggested that the confusion is all caused by a lacuna in an early manuscript of the Notitia Dignitatum and all confusion could be avoided with a small restoration of said text, which should be read as follows:

This small and intuitive amendment was later championed in The Place-Names of Roman Britain by A.L.F. Rivet and Colin Smith (Batsford, London, 1979) and seems to have settled the matter – until the next piece of controversial evidence turns up.

Camboglanna – The Crooked Glen

The first mention of the Castlesteads fort is contained within the Notitia Dignitatum, the ‘Register of Dignitaries’ of the late-4th/early-5th centuries. In this document Castlesteads is listed as Amboglanna, between the entries for Magnis (Carvoran, Northumberland) and Petrianis (Stanwix, Cumbria).

The fort is also mentioned in the seventh century Ravenna Cosmography, where it is seemingly listed twice; the first and most-likely entry is named Gabaglanda (R&C#131), and occurs between Magnis (Carvoran, Northumberland) and Vindolande (Chesterholm, Northumberland), whereas the second entry Cambroianna (R&C#167), is listed between the unidentified stations Locatreve and Smetri.

The accepted name Camboglanna could be Welsh/Gaelic in origin and if so would translate as ‘Crooked Glen’, which refers, no doubt, to the fort’s spectacular setting overlooking the valley of the Cam Beck, a convoluted tributary stream of the River Irthing.

The Legionary Presence

The presence of the Sixth Legion at Castlesteads may be attested by a building stone (RIB 2000) found in 1732 near the eastern gateway of the fort.

1. The short-lived Gordian dynasty lasted from the accession of Gordian I in January 238AD until the murder of Gordian III in February 244.

The presence of the Twentieth Legion appears to be attested by an inscribed slab (RIB 1999) recovered from a hypocaust at Castlesteads. In the text the legion is styled Gordiana or Gordian’s Own, which was awarded to/adopted by units to show alligence to this emperor. The slab also mentions the third-century garrison unit Cohors Secundae Tungrorum (see below), and from the restored text it appears that the building work may have been carried out by this auxiliary unit under the supervision of a senior centurion on secondment from the Twentieth.

The Garrison Units

Cohors Quartae Gallorum – The Fourth Cohort of Gauls

Two inscribed altars recovered from Castlesteads show that the earliest recorded unit at the site was Cohors IIII Gallorum Equitata, a mixed unit of infantry and horse recruited from among the many Gaulish tribes of central France. The stones contain no clues in their textual content which may be used to positively date this unit’s period of tenure, but it is generally assumed that they represented the Hadrianic garrison, although they may have been moved here during the time of Antoninus Pius (120-160AD). ????

Cohors Primae Batavorum – The First Cohort of Batavians

The second century garrison of Castlesteads was possibly Cohors I Batavorum Equitata, a one-thousand stong mixed cavalry and infantry unit originally recruited from among the Batavi tribe inhabiting an island at the mouth of the River Rhine in Belgic Gaul. This unit is also recorded on building stones recovered from Carvoran nearby.

Cohors Secundae Tungrorum The Second Cohort of Tungrians

1. This unit also took the emperor Gordian’s name as a show of support for his rule. (see RIB 1999 by the Twentieth Legion, above).
2. The governorship of Publius Aelius Martinus is known only from this stone.
3. Imperator Caesar Marcus Antonius Gordianus Pius Augustus (II) and Pompeianus were consules ordinarii for the year 241AD (a.u.c. 993).

An inscription on an altarstone (RIB 1983) dated to January 1st 241AD records the Castlesteads garrison as Cohors II Tungrorum Milliaria Equitata, a one-thousand strong mixed unit of cavalry and infantry recruited from amongst the Tungri tribe inhabiting the Ardennes region of modern Belgium.

The Epigraphy of Roman Castlesteads

There are 40 latin texts recorded in the R.I.B. for the Castlesteads fort, many of which are given and translated on this page.

Dateable Latin Inscriptions from Castlesteads

The Minor Settlement

Geophysical surveys conducted between 1999 and 2001 recorded the course of the Vallum passing close by the southern defences of the fort and also revealed details of a civil settlement or vicus on the southern slopes of the hill, directly opposite the south-eastern gateway of the fort where there was a causeway across the Vallum. These same surveys revealed traces of Roman field-boundaries in the area to the immediate east of the settlement.

Tombstones from the Castlesteads Vicus

Only four tombstones have been recovered from the Castlesteads area, one of which (RIB 2005) is beyond any hope of translation, being too heavily damaged; the others are shown below.

1. Based on the expansion C[ustos] A[rmorum], literally ‘the Custodian of the Armoury’. The abbreviation C A may also be expanded C[uram] A[gente], or ‘responsible for conducting [things]’ (see note#2 below).
2. The second part of this inscription may also be translated as: ‘… His heirs saw to the making of this [memorial] through the agency of Flavius Hilario the centurion’.

The Gods of Camboglanna

There have been twenty-one inscribed altarstones and religious texts recovered from the environs of the Castlesteads fort, comprising: seven to Jupiter Best and Greatest, including one shared with the Numinibus Augustorum (RIB 1983) and another shared with the Genius Loci (RIB 1984); three to Sol invictus / Mithras; two to Belatucader, including one shared with Minerva (RIB 1976); two to Sanguine Mars, including one shared with Num Aug (RIB 1987); two to the Mother Goddesses; also single stones dedicated to the deities Neptune (vide supra), Victory, Vanauntis, Discipline and one unknown god (RIB 1996; not shown).

Altarstones Dedicated to Jupiter

Of the seven altarstones dedicated to Jupitter Optimus Maximus the king of the Roman pantheon, five are given above as they name military units, one is given below. The remaining stone is fragmentary, only the top part surviving, which simply reads: I O M VOTVM …, ‘To Jupiter Best and Greatest a vow …’, which is hardly worth the bandwidth.

Altarstones to Mithras the Invincible Sun

Altarstones Dedicated to the Mother Goddesses

The Mother Goddesses are mentioned on a couple of altarstones, both of which are given here. The temple of the Matres lay just outside the Castlesteads fort to the south-east.

Camboglanna Today

Camboglanna Related Links

References for Camboglanna

• Britannia xxxii (2001) p.333 & fig.12 p.335;
• Hadrian’s Wall Map and Guide by the Ordnance Survey (Southampton, 1989);
• Hadrian’s Wall in the Days of the Romans by Ronald Embleton and Frank Graham (Newcastle, 1984);
• The Roman Inscriptions of Britain by R.G. Collingwood and R.P. Wright (Oxford 1965).

Roman Roads near Camboglanna

Hadrian’s Wall: W (2.5) to Banks East Wall: E (7.5) to Birdoswald (Birdoswald, Cumbria) Wall: W (8) to Uxelodvnvm (Stanwix, Cumbria) Hadrian’s Wall: W (1.75) to Leahill