Structural Report

There were six main phases of activity on the site, and these can be outlined as follows:

Phase 1: The Primary Fort (Fig.2)

Where the natural material was identified, it comprised a yellow boulder clay (080, 160, 354) with lenses of sandier material (079, 186). The original Roman ground surface appears to have sloped upwards from east to west. No trace was found of the pre-fort plough-marks noted in Middle Chare (Evans et al. 1991, 15, Fig.5), the earliest occupation of the Church Chare site being the western defences of a turf-and-timber Roman fortification.

The north-south line of a ditch was evident where the foundations of the later stone building had been reinforced to cross it (Fig.3, Sections 2 and 3), and it was examined by machine-dug section at the extreme southern end of the site, next to the existing Parish Centre building. The ditch was U-shaped in profile and 3.00m wide by 0.85m deep, although it is by no means certain that this represents its full, original depth.

Some 3.40m to the east of the presumed eastern lip of the ditch, the westernmost extremities of a turf rampart (342) were found in the north-eastern quadrant of the site. Immediately beneath the later via vicinaria, it survived to a height of 0.13m, its upper part consisting of compressed sandy turves with characteristic dark streaks formed by the decayed vegetation, on top of clay blocks. A sherd of BB2 came from amongst the laminated turves (see Dore below). In the machine-cut trench, however, the rampart appeared to have been constructed of clay blocks with no trace of turves being recorded. Elsewhere on the eastern side of the site, deposits of sandy clay (351) and yellow clay with sandy inclusions (014) were identified as belonging to the rampart. Between the rampart and the ditch, sandstone pebbles (343) rested on the natural clay.

The western extremity of the rampart coincided with three north-south series of post pits on a more or less parallel course. The first series (363, 365, 367, 369, 373, and 375) were generally sub-rounded, c.0.20m in diameter and between 0.13 and 0.36m deep, and usually 1.40m apart (although 369 was inserted half way between 367 and 373). These pits were generally filled with rounded packing stones and dark-coloured gritty silt/clay (366, 368, 374). Post pit 365 retained the impression of a pointed oval post (0.10 by 0.06m) driven into its base.

There were two pits in the second series, the northern one (357) being 0.50 by 0.46m and 0.26m deep, and sub-circular in shape. Its partner (359), 0.70m to the south-west, was almost square (0.48 by 0.44m), and it was 0.22m deep. Both were filled with silty (but not identical) material and packing stones and, in both cases, these fills were partially overlain by rampart material (342). The alignment of these two pits differed from that of the first series.

The third series comprised three pits of varying shapes. The northernmost (371), 1.00 by 0.50m and 0.42m deep, was sub-rectangular in form, filled with light brown clay, sandstone, and sandstone fragments (370). 0.80m to the south lay 361, 0.50m long, 0.38m wide, and 0.38m deep, the fill being a dark brown gritty silt/clay similar to that found in the first series of pits. One of the packing stones in this bore a striking resemblance to the faced stone initially used in the Phase 2 stone structure (see below). A third pit (377) was situated 0.80m further south from 361; it was sub-rounded, 0.50 by 0.48m, 0.65m deep and filled with sandy material and sandstone packing (378).

Two possible further post-holes (350, 353) were noted to the south and east of the three main series, the fill of 353 (0.46m in diameter and 0.23m deep) included packing stones, although the fill of 350 contained pottery casting doubt on its primary nature (see Dore below, no.66).

Discussion

The first series of post pits associated with the rampart could represent a line of palisade posts (Jones 1975, 86), perhaps intended to support the breastwork, although the pits were positioned at the front of the surviving rampart material. They would not make for a practical structure, given the necessity for an angle of batter of around 65-75° for a turf-cheeked rampart (ibid. 70), nor do they seem likely to have been a frontal timber revetment like that at Verulamium (ibid. 83) or part of a box rampart, for the post diameters seem too slight. The most likely explanation may be that, since the land rises gently from east to west, the westernmost extremity of the rampart (i.e. that part beyond the line of post pits) may have been removed by levelling prior to the construction of the secondary fort.

Of the remaining two series of post pits, the second may represent a structure associated with the rampart, although they do not seem large enough to have belonged to an interval tower (cf. Jones 1975, 92). The third may not even have belonged with the first phase (a deduction hinted at by the presence of a dressed stone in the fill of one of them), even though its intrusion from higher levels was not noted during excavation.

The profile of the ditch located on the Church Chare site was U-shaped - unusual, but not completely without parallel (e.g. Deschler-Erb et al. 1991, Abb.55).

[PREVIOUS] [POPULAR ACCOUNT] [CAPTAIN'S LOG] [BEGINNING] [NEXT]