A POSSIBLE OUTER BAILEY DITCH TO MARLBOROUGH CASTLE 



105 



The third coarseware, represented by a single 

 sherd, is a Laverstock type from the Salisbury area, 

 again probably 13th century. This is also the source 

 of the dominant fineware type, a fine whiteware 

 found here in glazed jug forms, decorated with 

 incised or red-slipped motifs. Comparable jugs were 

 found during the excavation of the Laverstock kilns, 

 which have an archaeomagnetic date of 1240-75 

 (Musty et al. 1969), although production here (or 

 elsewhere in the immediate vicinity of Salisbury) is 

 likely to have been longer-lived. 



Two other glazed wares were identified. One is 

 calcareous (tempered with oolitic limestone), of a 

 type found in north Wiltshire, for example 

 associated with the Minety production centre; two 

 of the three sherds have combed decoration, which 

 can be characteristic of 1 2th century tripod pitchers, 

 although could equally well be later. The second 

 type is sandy and is not particularly distinctive. 

 Medium-grained sandy wares such as these are 

 widespread across central southern England and, 

 like the Kennet Valley wares, are likely to have 

 several different sources - one known kiln source is 

 at Ashampstead in west Berkshire (Mepham and 

 Heaton 1995). 



CONCLUSIONS 



Archaeological Interpretation 



Two broad groups of archaeological deposits have 

 been revealed at this site: a broad ditch of defensive 

 proportions, the upper backfills of which had been 

 cut into by two smaller ditches containing 13th 

 century pottery; and a group of pits and smaller 

 features to the south of it, redolent of domestic 

 functions, some of which contained 1 3th century 

 pottery in their lower fills. Of these, the latter are 

 more readily interpretable. 



In form and fill characteristics, pits 1010 and 

 1020 are similar to features identified as cess pits 

 at many urban archaeological sites. Pit 1012 is rather 

 more enigmatic, being of similar form and 

 dimensions to 1010, but having apparently been 

 backfilled immediately after its excavation. 

 Furthermore, their linear alignment follows the 

 patterns identified at many urban sites, in which 

 cess pits invariably indicate both the orientation and 

 extent of individual burgages (cf. Schofield and 

 Vince 1995), though the orientation of these 

 features is contrary to that presumed on the basis 



of the present property layout in the town (cf. 

 Haslam 1978, fig 11) in being at an angle of 

 approximately 60 degrees to the High Street. 



Feature 1014, being square in profile and plan 

 form, containing flint rubble in a chalky matrix, is 

 likely to have been a building foundation that has 

 been robbed of its larger masonry units and mortar. 

 Though undated, the absence of brick and the 

 purity of its matrix suggests a medieval rather than 

 Roman or post-Medieval date, whilst 

 stratigraphically die deposits pre-date the southern 

 pit group. The absence of domestic materials such 

 as charcoal or animal bone from the immediate 

 vicinity of the feature suggests a non-domestic 

 function. 



The other features, comprising isolated post 

 holes and smaller linear gullies, are of a form and 

 spatial density that might be anticipated on the 

 margins of a medieval settlement, and potentially 

 relate to a wide variety of domestic and other 

 functions. Though undated here, there is no reason 

 to preclude contemporaneity with the other 

 medieval deposits on the site. 



It would be reasonable to conclude, therefore, 

 that the above represent the backland of a burgage 

 fronting on to the southern end of the High Street, 

 established during the 12th - 14th centuries and 

 demarcated on its northern side by the pair of 

 smaller ditches cutting into the upper backfills of 

 ditch 1037. This much, at least, is unremarkable. 

 Ditch 1037, however, is slightly more enigmatic. 

 Material from the two upper 'recuts' indicates that 

 1037 had fallen out of use by the late 13th century 

 and must therefore have been cut and initially silted 

 at an earlier date. Its steep-sided 'V -shaped profile, 

 and considerable depth, are indicative of a defensive 

 function, rather than a simple boundary or quarry. 

 Though it is situated outside the accepted extent 

 of the Norman castle earthworks (cf. Haslam 1978), 

 the length observed within this site appears to be 

 broadly concentric with the western end of 

 Bridewell Street, an urban feature accepted as 

 respecting the alignment of the castle defences. If 

 we accept the assertion of Bradley etal. (1923) that 

 a second bailey was added to the north-eastern edge 

 of the Norman castle earthworks, and that the 

 medieval name for St Peter parish was 'The Bailey 

 ward' (Bradley er al. 1923), it is possible that 1037 

 represents the outer edge of the later bailey and 

 that it is likely to extend as an archaeological feature 

 across the southern end of the High Street, 

 containing SS Peter and Paul. Its steep profile and 

 lack of the broader base characteristic of Norman 



