The earliest sign of habitation at Banna is a Neolithic burial chamber in the cliff which has been tentatively dated to 3,500-2,500 BC. The Romans later reused it. Prior to their arrival the area of the fort was woodland populated by hazel, birch and alder, and oak and hazel on the slopes down to the river Irthing. Unlike the general locality, Banna seems to have been uninhabited during the Iron Age.

The first building on the site was a signal tower similar to those at Walltown and Pike Hill, and was associated with the Stanegate frontier and the fort at Nether Denton in the valley below. The next phase of building was the 15ft high turf wall that was part of the original plan for Hadrian’s Wall and ran between Willowford Bridge (650m east of Banna) and Bowness on Solway 30 miles away. As the turf at Banna itself was unsuitable for wall building, turf had to be carried from Midgeholm Moss some 200 metres away. On the site of the future fort there was a stone turret, also part of the original plan. A temporary camp was built on the site of the future fort for the construction teams working on the turf wall who were accommodated in leather tents. The camp had defensive ditches built around it.

The Turf wall was completed in 124, the same year as the bridge at Willowford. Shortly after this there were 2 changes to the design of the wall which happened either at the same time, or one very closely after the other. The first change was the decision to base units in forts built upon the wall itself, rather than at their previous bases on the Stanegate, the second was to build the vallum, a defensive ditch that ran for almost the full length of the wall. The vallum around Banna is unusual in that due to the lack of space there was no room for a northern mound.

The first fort to be built at Banna is presumed to have been one of timber, though the evidence for this is fairly tentative. Soon afterwards a permanent stone fort replaced it. The stone forts on the wall were not completed en masse overnight, but rather over the course of the next 10 to 14 years. The fort at Banna was built straddling the wall like most of the primary forts and was probably finished by around 130, though the work was completed in 2 stages with a number of years separating them. Evidence for this is shown by the differing qualities of workmanship shown between the well crafted foundations and the less well carved structures above, and evidence that vegetation grew on the site and had to be burned off before work could continue.

The second phase of building was short. The site was drained, streets were laid out and the internal stone buildings were built. During this time a third major alteration to the Wall took place. The decision was made to replace the turf wall with one of stone. At Banna this lead to the bringing forward of the line of the wall so that it now met the fort at the north east and north west corners, rather than further back on the fort as it had previously. This meant that there were now 5 gates to the rear of the fort, the south gate, and a single portal and double portal gate on both east and west sides. As the single portal gates were no longer needed the towers were torn down and the rubble was used to fill in the gateways.
This period of occupation for Banna lasted until 138-142. There is no evidence that the garrison of the fort left during the occupation of the Antonine Wall, though its garrison may have been reduced. By the end of the 160’s the Antonine wall had been abandoned and Hadrian’s Wall and the outpost forts once again became the active frontier. At this time the defences around Banna were strengthened.

Further work at Banna took place during the reign of Septimus Severus when the 1st Cohort of Aelian Dacians under Aurelius Julianus, together with the 1st Cohort of Thracian Roman Citizens rebuilt the granaries, though the Thracians seem to have been at the fort to help with the construction work, they were not stationed at the fort and no other evidence of their stay there exists.

There was a lot more work undertaken by the Dacians, an example of which is the rebuilding of the east gate under Claudius Menander in 219. In the west gate the south tower was rebuilt using masonry from another pre-existing building or monument as the stone used was of a particularly high quality. Further rebuilding work took place at Willowford Bridge. The bridge had been damaged in the 160’s and now its rebuilding and the building of a road called the military way was to have a further impact upon the fort.

Banna was now connected by a road to the eastern part of the wall. Where previously there had just been the walkway of the wall itself, the road offered new economic opportunities to local inhabitants and to the soldiery. For the rest of the 3rd century the Dacians continued in occupation of the fort and at this time a small town, or vicus, grew up around it. Soldiers, for the first time, were legally allowed to marry and this coupled with the trading opportunities afforded by the new military road led to a large expansion of buildings to the east and west of the fort. The South gate was blocked around this time as it was largely defunct and bread ovens were placed there.

At the end of the 3rd century the fort was abandoned for a short time. The political situation was desperate. Britain, for a short time, became part of a breakaway “Gallic Empire” that split with Rome between 260 to 273 AD. Upon Diocletian’s accession and the forming of the Tetrarchy it looked as though Britain would be brought back in, but Carausius, and his successor Allectus continued their revolt.

The Dacians are recorded as being at the fort at some time during the reign of the emperor Probus (276 to 282), but then left, and the fort appears to have been derelict for at least 15 years. Heavy restoration took place at Banna under Constantius Chlorus between 297 and 305, but after the reforms of Constantine their numbers were severely reduced from the strength the unit had been when they first arrived. As a unit of frontier troops, or limitanei their strength was now just over 100 men. Also, the sons of the men serving with the unit were compelled to join the army. Military service was now hereditary. This must have had little impact at Banna as joining the army must have been the aim for the sons anyway, giving a solid income and career, when the alternative was trading or farming. The nature of the pay awarded to the soldiers changed at this time also. Although there were occasionally cash payments most of the soldiers were paid with goods, possibly those they managed to extort from the local population as tribute or tax. For the confused happenings of the later period of roman occupation you should refer to the history of the wall itself. At Banna the late 4th century saw a period of neglect with haphazard building work. The roof of the north granary collapsed and the building was quarried for its stone. Around the same time the south granary was used as living quarters showing that the large food stores of a year’s supply for 1,000 men was no longer needed. By 420 the south granary building had also collapsed and for the next 100 years the only construction at the site would be internal buildings of timber, and earthen banks for defences. From now until the 12th century evidence of occupation at Birdoswald is slight indeed with a solitary Saxon brooch being the only evidence of occupation between the 6th and 10th centuries.

Units Based at Banna

Sources: