The earlier excavations on the defences, undertaken by various people since 1948, have been the subject of a final report by C.A.R Radford (1972). Radford's excavation report also contains details of otherwise unpublished excavations conducted by G.M. Knocker and A. Maddison, together with assessments of the ceramic evidence by E.M. Jope. It has been found necessary to reinterpret and reassess both the reporting and the conclusions in this report in the light of the evidence from the 1975 excavations. The original records have therefore been examined to test the conclusions put forward in Part 1. The following is a summary of the results, under the five periods set out in Part 1.
Note: where Radford is quoted in this section the references are to his 1972 publication.
There is no evidence in any previous excavations for the cultivation of the ground earlier than the construction of the bank of period 1, as was assumed by Radford.
a) There is no evidence in previous excavations, or in the 1975 excavations, for timber strapping within the bank, as suggested by Radford.
b) In many places the front turf revetment, into which the stone wall (period 2A) was inserted, has been misinterpreted as infilling behind the wall (e.g. in Radford's trench R.V, re-excavated in 1975).
c) The 'rear revetment wall' noted by Radford in many places and shown as a flat layer of stones at the back of the bank in most of his sections, is in fact the intra-mural walkway, which was identified in the 1975 excavations. This feature can be recognised on all four sides of the defences.
d) The 'palisade trench' along the line of the front of the wall can best be interpreted as a marker ditch or stabilising trench for the front of the wall of period 2A. There is no clear evidence that the original bank was fronted with a wooden palisade.
e) There are a number of places where evidence of ditches was noted, but either not mentioned in Radford's report or not correctly interpreted (by both Wainwright and Radford). This evidence shows that the inner ditch encircled the defences, and both the centre and outer ditches were present on all sides except the north.
2A. It is clear that in every place where the relationship between the bank and wall was observed, the front turf revetment was substantially intact at the time of the insertion of the wall, and had been cut back with a near-vertical face to receive it.
2B. Deposits of this period, in which the defences must have fallen into disuse, have been located in many trenches around all four sides of the defences. These include deposits on the berm, underlying the destruction deposits of the wall (period 3A); and erosion of, and infill in, the ditches, before they were cleaned out in period 2C.
In this period both the wall fronting the bank, and the wall at the rear of the bank, were destroyed completely.
a) The destruction debris on the inner berm has appeared in every section across the defences, in many of Wainwright's records showing clear tip-lines of stones. It is interpreted by both Wainwright and Radford as the product of robbing of the wall. In a number of places Radford has misinterpreted the sometimes large stones in this deposit as a structural feature deliberately placed to prevent erosion of the berm or the edges of ditches.
b) An ubiquitous spread of stones appearing in every section behind the bank, which can be interpreted as the product of the destruction of a wall at the rear of the bank in this period, has been mistakenly interpreted by Radford as the stones of an intra-mural street.
c) There is evidence of the presence of stone-filled ditches which were not recorded by Radford (and in some cases recorded by Wainwright but not marked on Radford's sections) on all sides of the defences except the north.
a) Palisade trench: this is shown clearly in nearly all of Wainwright's sections and site photographs, but not noted in Radford's report. In one place its fill contained pottery of the late 11th-early 12th century (identified by E.M. Jope). It was present on all sides of the defences.
b) Inner and outer ditches: these were recorded on all sides of the defences, but not noted in Radford's report.
a) There is no evidence for the existence of the Roman cemetery recorded by Radford (1972, 95). The burials are medieval (they are cut into the fill of the Saxon ditch), and the Roman tombstone fragment was built in to the Saxon wall.
b) The 'internal street' noted by Radford (1972, 79, fig. 7) is merely a ridge of a medieval ridge-and-furrow system.
c) There is not only no evidence for the existence of a gateway in the centre of the west side of the defences, but also much evidence from excavations for its absence. It seems most likely that the Saxon west gate was on the alignment of the present Bath Road.
© Internet Archaeology
Last updated: Mon Jul 7 2003