Like adjacent OA2, OA3 is conjectured to extend between OA5 and the lower terrace edge. No eastern limit is defined and the northern half of its interior is unexplored.
Within its investigated parts (i.e. within the northern end of Excavation Area N and Area P), the quantity of features and complexity of the inter-cut stratigraphic sequence suggests at least two phases of activity, or at least a relatively complex process of feature abandonment and replacement through Period 2A. The earlier activity, generally dating to the later 1st century BC, is described here. Later activity, dating to the early 1st century, is described in Lower terrace (south).
Feature 25094, an apparent elongated pit (segs 19115, 19144), is 11m long, up to 2.4m wide and aligned east-north-east/west-north-west, with slightly rounded terminals at either end. At in excess of 0.6m deep, it is notably deeper than other ditches and pits in this vicinity. Its dimensions therefore set this feature apart and it is difficult to decide whether it was an elongated pit, trench or ditch. It is even possible that it is instead a line of sequential inter-cut pits; the two excavated segments reveal notably different fill sequences. If a single cohesive feature, its positioning orthogonal to the perceived western boundary may suggest that it marks something of a subdivision of the enclosure. Whether one pit or a number of inter-cut pits, its artefactual content dates it to the late 1st century BC. It post-dates pit 19159 (Group 44) dated to the LIA by grog-tempered pottery. The pottery retrieved from 25094 includes a component of handmade Late Iron Age forms in sandy fabrics, which are possibly of Middle to Late Iron Age transitional date (KPG1). Briquetage, a loomweight and a spindlewhorl are also present.
Building 14 (Figure 20) comprises a semi-circular arrangement of eight post-holes later overlain by Period 2B Building 32. Assuming that this structure is built as a complete circle, but subsequently truncated by elongated pit(s) 25094, it has a diameter of 4.35m that defines a roundhouse with a floor surface of approximately 15m². Though small, it compares well with numerous similar circular buildings elsewhere across the southern part of the lower terrace. Like the buildings in adjacent OA2, Building 14 occupies a possibly central position in relation to its east and west boundaries and is located toward the south of the enclosure.
Alternatively, though perhaps less likely, these post-holes may represent the entirety of a structure enclosing the eastern end of the elongated pit and so be contemporary with it.
The stratigraphically earliest defined pitting is represented by pits 8525, 8547, 8748 and 19159 (Group 44) at the southern end of the enclosure, and 10896 (Group 45) further north. Of these, large pit 19159 is probably filled before the end of the 1st century BC, as it is cut by well-dated elongated pit 25094. Pit 8525 dates to the late 1st century BC or perhaps early 1st century AD. The remainder contain only grog-tempered pottery of a general Late Iron Age date or are undated.
Also at the south end of OA2, pit 8786 (Group 48) is notable for its pottery assemblage (KPG3). On traditional dating, it should be assigned to the late 1st century BC from a Dressel 1 amphora handle, or the early years AD, and probably the 'earlier' handmade pottery would be dismissed as residual. However, their association here, and the forms of the grog-tempered sherds, suggest that this assemblage constitutes a Middle to Late Iron Age transitional group. It is conceivable that the Dressel 1 amphora could reach the site even earlier, but there is little other evidence pointing to any activity pre-dating the middle of the 1st century BC. If this assemblage, and with it the middle Iron Age to late Iron Age transition, can really be dated to as late as the late 1st century BC, then perhaps many more features currently in Period 1 ought to be thought of as Period 2A. There are features whose pottery may suggest a similar date (in this area, 8525, 8586, 8755, ?8839, 19023, 19115, 19144). This clearly has important implications for the dating of the Middle to Late Iron Age transition.
The assemblage also suggests possible ritual aspects to its deposition, with two or three half-complete vessels present. The inclusion of loomweight fragments, an iron binding strip and a nail, a flint blade and an amber bead are also noted.
Relatively substantial (1.2m wide and 0.66m deep) ditch 25174 (segs 8508, 8516), at the south end of OA3, is traced for c. 35m northward to a bulbous irregular terminal. Where investigated within Excavation Area P, its fills contain small quantities of grog-tempered and sand-tempered pottery, some handmade, which may indicate a mid- to late 1st century BC date. Some residual prehistoric pottery is also present. This ditch is possibly the earliest evidence of the linear land division that becomes an important aspect of Period 2B. This ditch runs almost on the same alignment as the later divisions, although its course seems to display a slight curve at its south end. The curved portion provides a partial counterpart to ditch 25097 (Group 49), suggesting the two are perhaps associated. It is surely significant that the ditch extends north only as far as elongated pit 25094.
Ditch 25264 (segs 10850, 10857, 10913) is a c. 6m length of ditch that is on the same alignment as, and presumably related to, ditch 25166 (Group 1), which forms the boundary between OAs 2 and 3. Like 25166, 25264 is a shallow, narrow feature running NNW-SSE and extending northwards under the line of Road/Track 3 (Period 2B). The two ditches are c. 8m apart and both ditches are c. 0.8m wide and 0.44m deep with steep sides and flat bases. Ditch 25264 seems to subdivide Open Area 3 and may have defined the eastern side of an entrance corridor into Open Area 3 - or else between OAs 2 and 3. It may also be aligned on ditch 25174. There is no sign that this entrance/track is surfaced, nor does its line survive the alterations of Period 2B.
The apparent termination of ditch 25264 c. 1.5m to the north of Building 12 suggests that the two are contemporary. The small amount of amphora and grog-tempered pottery present in the fills of 25264 is consistent with a Late Iron Age infilling date.
Although recorded in the field as a pit containing a distinct re-cut, 11744 is re-interpreted as a construction cut for proposed well-shaft 11745, the full depth of which was not determined because it was not excavated beyond the water table. As wide as 5m diameter at its top, the construction cut fills are notable for the relatively large quantities of pottery they contain. Both cut and well-shaft 11775 are probably filled before the end of the 1st century BC.
Reuse of the slump hollow over this feature as the stokehole/rake-out of Kiln 1 (in cut 11477) is described elsewhere. Some of the material in the well fills may, in fact, derive from this secondary use and subsequent disuse and infilling.
Internet Archaeology is an open access journal. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.
Internet Archaeology content is preserved for the long term with the Archaeology Data Service. Help sustain and support open access publication by donating to our Open Access Archaeology Fund.