| Summary : Earthwork remains of a Roman temporary camp. Camp I, the largest element on the site, is almost square and enclosed an area of about 7.7 ha (19 acres). The NE and SW sides are parallel and the W and N angles are right angles. The camp faced NE towards Dere Street but its size, and the dimensions chosen, determined that the rear rampart had to take in lower ground, especially at the S angle, which is 10 m below the s corner of the later fort (IV). A camp of more elongated plan would have made better use of the level land available to the NW. The earthworks are most impressive on the SE side where the outer scarp of the rampart stands 2.0 m high above the bottom of the ditch which is 0.4 m deep. The outer scarp of the rampart is accentuated here by the slope of the hill, but nowhere does the inner scarp survive to more than 0.5 m in height. For a short distance on the SW the defences are covered by the counterscarp bank of the later fort. |
More information : This camp has been re-assessed in connection with RCHME's survey and publication of Roman Camps in England. The following descriptive account is taken from the published text. All elements of Chew Green complex now assigned unique identifiers under parent record NT 70 NE 3.
Camp I, the largest element on the site, is almost square and enclosed an area of about 7.7 ha (19 acres). The NE and SW sides are parallel and the W and N angles are right angles. The camp faced NE towards Dere Street but its size, and the dimensions chosen, determined that the rear rampart had to take in lower ground, especially at the S angle, which is 10 m below the s corner of the later fort (IV). A camp of more elongated plan would have made better use of the level land available to the NW.
The earthworks are most impressive on the SE side where the outer scarp of the rampart stands 2.0 m high above the bottom of the ditch which is 0.4 m deep. The outer scarp of the rampart is accentuated here by the slope of the hill, but nowhere does the inner scarp survive to more than 0.5 m in height. For a short distance on the SW the defences are covered by the counterscarp bank of the later fort.
Farther NW, immediately to the SE of the SW gate, a small counterscarp bank would normally be interpreted as part of the natural hillside, a consequence of the ditch being cut into the slope. However, Richmond's section through the defences at this point revealed what the excavators interpreted as a low glacis mound of upcast. The ditch here was said to be 8ft 3ins (2.5 m) wide and 3 ft 9 ins (1.1 m) deep, including a basal channel. The main rampart, 10 f (3.0 m) wide, was laid on a foundation of turf, which retained a core of rock brash (Richmond and Keeney 1937, 133-5 (1a); the measurements given in the text do not agree with the published section). Towards the W angle, where the ground is crossed by a series of drains, the preservation of the defences of camp I becomes markedly poorer as the ground levels out. Here the defences survive only as a broad bank spread to a width of about 3 m, its smooth outer scarp nowhere more than 0.5 m high; the line of the ditch is marked by a change in the vegetation.
On the NW side, the earthworks appear to be out of alignment around the junction with the SW side of camp III; this may be an illusion caused by the drains that cross the defences here at a very oblique angle, but the distortion is shown on Robert Tate's plan of 1810 (Alnwick Castle Archives: Sir David Smith's Collection, Northumberland Atlas 2:-187a/2). Nevertheless, the cutting of the drains makes it impossible to determine, from the surface remains alone, the sequence of construction of camps I and III, the perimeters of which intersect at this point; natural scouring down the hillside has largely contributed to the impression, apparently erroneous, that the ditch of camp I cuts through the bank of camp III. To the NE, a section of the rampart of camp I has been almost levelled (pace Richmond and Keeney 1937, 136 (see auth 1a) to the W of the gate on this side. The slighted rampart is only 0.1 m high; the remainder of the material was probably pushed into the ditch. The latter is only 0.2 m deep and seems to have been recut as a modern drain; upcast material has been spread outside the ditch for a distance of 3 m.
The single gate on this NW side is on the SW edge of the broad summit between the two burns; to the NE, the preservation of the defences is much better, and here the outer scarp of the rampart stands 0.9 m above the bottom of the ditch, which is 0.4 m deep. The inner scarp of the rampart is 0.5 m high. Richmond and Keeney (1937, 136 (see auth 1a) stated that their trenches at the NE junction with camp III showed that the rampart of camp I had been cut through by the ditch of camp III, which was therefore later in date. Immediately adjacent to this intersection, the NW half of the NE side of camp I is crossed and overlain by the low earthworks of medieval enclosures.
The ditch of the camp seems to have been recut in places as a drain and the rampart was probably reused as part of the later enclosures; despite this, the outer scarp of the rampart survives to a height of 0.7 m although its inner scarp and the ditch are each only 0.2 m in height and depth respectively. The SE half of this NE side of camp I evidently determined the position and alignment of the NE side of the later fortlet (V), and of the two 'annexes' which overlie the E angle.
The NE gate of camp I probably lay at the crest adjacent to Dere Street and should be overlain by the N angle of the fortlet (V). The other three gates are, however, visible. The SW gate, set almost exactly at the centre of that side, is on the crest of the ridge; no external traverse can be seen and thus the inner side of the gate was chosen for investigation in 1936, when the springing of a clavicula was apparently identified 'in the requisite position, bedded upon an extra thickness of turf, as in the main rampart' (Richmond and Keeney 1937, 134 (see auth 1a)). The NW gate is now little more than an interruption in the line of the ditch, with the slight stub of an internal clavicula, up to 0.2 m high, still apparent. A trial trench across the expected position of this clavicula in 1936 again revealed laid turf (see auth 1a 134). In the comparable position on the SE side, a broad gap immediately to the NW of the SW defences of the more southerly 'annexe', to the S of the fortlet, was probably an original gate into camp I, and may have remained in use throughout the active life of the site. Full information is included in the NMR Archive. (1)
Additional references. (2-3) |