Colchester Archaeological Trust

CAT Report 556: summary

An archaeological excavation at 21 St Peter's Street, Colchester, Essex in 2008: 2010
by Adam Wightman (with contributions from -)

Date report completed: July 2010
Location: 21 St Peter's Street, Colchester, Essex
Map reference(s): TL 99595 25534.
File size: 7,806 kb
Project type: Archaeological excavation
Significance of the results: ***
Keywords: Roman, street, drains, town wall, culvert, ?lilia

Summary. Although this was a relatively small site, the investigation produced a series of interesting and significant results covering a range of issues. Some of the conclusions are unexpected and most are problematic in some way. The bulk of the site is outide the Roman town wall and was expected in at least the Roman period to have been a kind of boggy 'no man's land' between the wall and the southern edge of the town ditch about 10 m to the north. In the case of no 21 St Peter's Street, it was supposed that the base of the town wall would pass unbroken across the south boundary of the site and that an internal Roman tower would lie immediately behind it (ie further north) with a Roman culvert on its west side. However, the excavation revealed a much more complicated story. The centre of the site was dominated by part of a Roman street heading northwards towards the river. The street was flanked by wooden drains, one on the west and two on the east. It is very hard to reconcile these remains with anything other than a street. The thick gravelled metalling was typical of the streets found inside the walled town as was the presence of flanking wooden drains, as were the flanking drains and the width of the streetat The problem then becomes if this was part of a street, what was its relationship to the wall? Various strands of evidence suggest that the street had been contemporary with the wall but was laid a short while before its construction. The presence of three drains seems excessive and usual and suggests that their main function was to channel away unwanted water from a spring directly uphill from the site (in other words, from the south). This must be the spring that was later to feed medieval Stockwell. In Period 1, Phase 1 (AD 50-61), the street consisted of a thin metalled surface. The presence of drains or drainage ditches cannot be established but either or both the drains on the east side could be this early. The street lines up with the Town Period 2 pre-Boudican street grid inside the town. (The street cannot be military in origin (ie AD 44-50) because no street could have existed in this position until after the legionary defences were filled in.) In Period 1, Phase 2 (AD 61-65/80), the street was re-established or relaid soon after the Boudican period. The west drain was laid sometime after AD 62. The town wall was built c AD 65-80. The drains were all cut through and effectively rendered useless by the trench for the foundation of the new wall. The southernmost plank covering the west drain was removed and the drain was partly filled in with some of mortar and stone used for the construction of the wall. The street continued in use throughout this period. It was re-metalled and repaired on various occasions so that the metalling became much thicker with time (as was normal). Presumably, although no direct evidence was found for its presence, the wall incorporated a gate and this explains the presence of the street. The various exposures of wall point to a single or double arch structure without provision for pedestrians. The water which had previously been conveyed to the ditch or river by the wooden drains was now channelled away in a new drain which passed through the culvert in the wall on the west side of the site. In Period 3 (c AD 275-300), the butt ends of the town ditch on either side of the street were joined up and the street effectively became a cul-de-sac. It cannot be determined if the gate was blocked at this time as happened to the Balkerne Gate. Small pits with upright logs in the centre were dug into the street between the redundant gate and the town ditch. Some of these were sealed by a thin layer of metalling showing that they are Roman in origin and late in the sequence of metalling. Their position in relation to the gate and street indicates that they post-date the use of both as a ingress and egress into the town. The most likely explanation for the pits is that they were lilia - defensive devices containing sharpen posts pointing upwards. Very few lilia have been found in Britain or abroad. The ?lilia from St Peter's Street are not very firmly dated and there is a small but distinct possibility that they are post-Roman in date and hence not lilia at all. Given their potential archaeological significance, it is proposed that one of the timbers in the ?lilia is radiocarbon dated. This should reveal with a high degree of certainty whether or not they belonged to the Roman period and thus if they are likely to have been lilia. In Period 4 (post-Roman), the area appears to have been left as open land until it was gradually built over possibly from c 1700 with the evolution of Dead Lane, now St Peter's Street.