SORVIODUNUM - A REVIEW OF THE ARCHAEOLOGICAL EVIDENCE 



formal excavations and small scale trenching as well 

 as watching brief activities conducted during the 

 course of major gas and water pipe-line operations. 

 The evidence is presented under the separate site 

 headings of Old Sarum, Bishopdown and Stratford- 

 sub-Castle. 



OLD SARUM 



A detailed description of the excavations carried 

 out by the Society of Antiquaries is given by St John 

 Hope and Hawley (St John Hope 1910; St John 

 Hope and Hawley 1911; Hawley 1912,1913). The 

 work took place in the area of the Norman Castle 

 and in the NE quadrant of the hillfort which 

 included the site of the Norman Cathedral. 

 Professor Haverfield, then a Vice President of the 

 Society, summarized the Roman finds in WANHM. 

 He reported that the only artifacts found were eight 

 or nine coins, all of the later empire and illegible 

 except for a Maximian, some potsherds, a bronze 

 armlet, and other metal trifles, a quern from 

 Andernach, some tiles, three pieces of painted wall 

 plaster, and lastly a piece of wall in situ (Haverfield 

 1915,26). 



Montgomerie writing some 30 years later gives 

 a summary of the excavation results with certain 

 adjustments 'made possible through personal 

 knowledge'. In particular he gives a detailed 

 description of five late Iron Age/early Roman finds, 

 a bronze belt-link, three bronze brooches and a 

 bead-rim potsherd. He also gives an account of how 

 the section of wall was found and its construction 

 (Montgomerie 1947, 132-47). It was discovered on 

 the old ground surface when two untimbered 

 exploration galleries were dug laterally east and 

 north from an unfinished post-Norman well which 

 had been sunk down to the natural chalk through 

 the material of the Castle Mound. 



A pebble floor appeared in the northern gallery 

 above which were occupation layers 0.9m thick 

 containing Romano-British refuse including pottery 

 and a coin thought to be of Maximian. Building 

 foundations were encountered in the eastern gallery. 

 They were bedded about 0.15m below the level of 

 the pebble floor. A further wall core was located at 

 the far end of the gallery some 2.5m after 

 encountering the first wall which ran almost N-S 

 and was traced over a distance of 3.5m to 'an 

 external corner' (Montgomerie 1947, fig. 4). The 

 building was clearly of some size. It was built of 

 ashlar and flints, on a foundation of chalk lumps 

 with a clay and chalk bedding laid on natural gravel. 

 Unfortunately further exploration of the building 



had to be abandoned due to the extremely 

 hazardous working conditions. 



As mentioned above, the principal aim of the 

 1957 excavations was to ascertain the integrity of 

 the curtain wall around the outer bailey following 

 the earlier excavations from 1909-1915 which had 

 shown that it existed in the NW (Cathedral) sector 

 (Rahtz and Musty 1960). An additional aim was to 

 try and locate a tunnel revealed in 1795 and last 

 seen in the 1820s. The excavations proved that the 

 wall was absent throughout the whole of the NE 

 sector where only a bank was a vestige of the pre- 

 Norman defences. The tunnel was found and 

 appeared to link the outer bailey with the north 

 east exterior of the earthwork. No dating evidence 

 was found but it was thought likely to be of Medieval 

 origin. 



The excavations showed that the Old Sarum 

 earthwork was of early Iron Age origin followed by 

 occupation well into the Roman period. A section 

 cut through the bank in the NE sector revealed in 

 the lowest, undisturbed levels, hundreds of sherds 

 covering the late Iron Age and Roman periods 

 (Rahtz and Musty 1960, fig. 9). The earliest were a 

 series of bead-rim vessels, some of pre-conquest 

 'Belgic' origin. Because there was no stratigraphical 

 division between 'Belgic' and other sherds, 

 identified as Roman, the excavators questioned 

 whether all the sherds were not in fact post Roman 

 conquest. Pottery finds continued with forms from 

 the 1st century AD while the 2nd century was 

 represented by samian ware and coarseware. 

 Occupation continued without a break until at least 

 the late 3rd or early 4th century. There was a total 

 absence of developed New Forest pottery in the 

 trench which suggested that occupation might not 

 have continued into the Constantinian period. 

 However the excavators did point out that a sherd 

 of New Forest pottery was found in the area of the 

 tunnel only 70m to the NW. 



In addition to the pottery, the lowest levels of 

 trench B gave up hundreds of pot boilers, several 

 dozen fragments of Roman tile, including four 

 combed pieces and one stamped LHS from Minety 

 (Darvill 1979, 328; 343), brick, part of a block of 

 dressed Chilmark stone, a few fragments of Purbeck 

 or Chilmark roof tiles, and a complete bronze 

 brooch, probably of mid-late 1st century AD. The 

 excavators highlighted the fact that earlier 

 researchers had believed searching for evidence of 

 pre-Norman occupation was likely to be 

 unsuccessful due to the thoroughness of Norman 

 re-construction in obliterating all traces. Their work 



