SECTION 6.2: WHELFORD BOWMOOR ARCHAEOLOGICAL DESCRIPTIONPHASE 1 (EARLY 2nd TO MID 2nd CENTURY
AD) PHASE 2 (c MID/LATE 2nd TO EARLY 3rd CENTURY
AD) UN-PHASED FEATURES FROM SALVAGE WORK Phase 1 (early 2nd to mid 2nd century AD; Figs 6.1.3: Phase 1, 6.2.1a: Plans of excavations: southern area and 6.2.1b: Plan of excavations: northern area )SummaryThe earliest activity at the site comprised a sequence of enclosures in the south-east side of the field, covering an area c 80 x 60 m and dating to the early 2nd century AD. A possible midden (2) within one of the enclosures points to domestic activity in the vicinity, though this may not belong to Phase 1. Further smaller enclosures of c early-mid 2nd century date were located to the north-west, along with a series of long north-south linear ditched boundaries which may belong to this phase. A stone-lined channel (33) in the centre of the site may have been of Phase 1 date. Very limited dating evidence from ditches in WB88 to the south suggests that these also belonged to this phase. EnclosuresEnclosures 1 to 6 in the south-eastern part of the site seemed to form a coherent group, although it is uncertain how many of these were contemporary as the plan suggests a gradual shift in various boundaries. Enclosures 7 to 12 further west were generally smaller and lay on a different alignment. They may have been associated with the long linear ditch 8/13. Further elements of the enclosure groups were found to the south and east in salvage operations, but not excavated. These features included two long linear boundaries, that together with ditch 24 and 8/13, seemed to form a large (c 60 m²) rectilinear enclosure that encompassed many of the smaller enclosures in the south-east. On spatial grounds, this enclosure is likely to belong to Phase 1, although it could have continued in use into Phase 2. Enclosure 1 (10, 20, 21, 27, 28) Sub-rectangular enclosure 1 lay in the south-east corner of the field and measured 11 x 12m internally. Ditch 10 (1.2 m wide, 0.45 m deep), which formed the western and southern side, ran southwards for 10m from ditch 21, before turning east and terminating after a further 7 m, on one side of a 2m wide entrance. The fill was mostly of peaty clay with gravel at the base. It contained a small quantity of Roman pottery and a piece of copper alloy binding (SF 10). The corresponding eastern ditch (20; 1.8 m wide, 0.45 m deep) ran from the southern entrance for c 12 m before turning west and terminating 2 m from ditch 21. Thus a corresponding northern entrance was created, which had an elongated passage like structure formed by a short section of ditch joining the main north-south part of ditch 20. Pottery from ditch 20 was more abundant than ditch 10, and included savernake vessel fragments dated mid/later 1st to mid 2nd century. The only other find comprised a late 1st century AD brooch (SF 95). Both features showed signs of re-cutting, especially around the southern entrance terminals. Internally, the enclosure appears to have been sub-divided by two small shallow curving gullies (27, 28; 0.3-4 m wide, 0.1-2 m deep), neither of which contained any finds. Enclosure 2 (21, 22) A small (6.5 x 4 m) semicircular enclosure was formed by curving gully 22 abutting ditch 21. The gully was 0.6 m wide and 0.3 m deep, with a fill comprising gravelly clay. It contained a small quantity of animal bone and a small piece of copper alloy sheet with traces of gilding (SF 89). The enclosure lay within enclosure 3 and was probably contemporary. Enclosure 3 (3, 11, 14, 21) This sub-rectangular enclosure (c 18 m²) was situated centrally within the south-eastern group, and shared its boundaries with E 1, 2, 4, and 6. The ditches ranged from 1.2 to 1.4 m wide and 0.45 to 0.6 m deep, and were generally V-shaped in section. It appears that they were left to silt up naturally with no evidence of deliberate in-filling. The very small amount of pottery recovered dated from the late 1st to 2nd century AD, and the only other finds comprised an early 1st century AD brooch (SF 67) in ditch 14 and a sling stone (SF 32) in ditch 21. Within its interior lay enclosure 2, attached to the eastern ditch (21), along with a shallow peat filled depression (2) which may have been a midden, almost certainly belonging to Phase 2 (see below). Enclosure 4 (14, 24, 25, 26) Just to the north of E3, and sharing with it boundary ditch 14, was enclosure 4, measuring 18 x 11 m. Ditch 24 (1.3 m wide, 0.3 m deep) was traced running SE-NW for c 48 m along the southern margins of the marshy area through the middle of the site, and seems to have formed the northern limit to the south-eastern enclosure complex. Three small fragments of Roman coarseware pottery and two iron nails derived from the upper fills of this enclosure. Enclosure 5 (12, 23, 24, 26) Enclosure 5 (c 20 x 12 m) lay to the west of E4 and north of E6, and appears to have been open on its north-western side. Ditch 23 to the south was quite shallow (0.25 m deep) and difficult to fully trace, but did contain a small amount of animal bone and 2nd century Roman pottery in its upper fill. Ditch 12 (1.3 m wide, 0.5 m deep) formed the western limit of the main south-eastern enclosure group, and continued into ditch 7 to the south. It was filled with dark clay and gravel and contained a small quantity of 2nd century pottery in its upper fill. Enclosure 6 (3, 7, 23) Enclosure 6 (15 x 13 m) lay south of E5 and west of E3. Ditch 7 on its western side (1 m wide, 0.4 m deep) was cut by Phase 2 ditch 8, and contained northwards to become ditch 12. The only datable pottery was two small 2nd century sherds, which came from its shared boundary with E3 (3). The only internal feature was an unexcavated section of gully (c 0.4 m wide), which may well have been contemporary as it was parallel to ditches 7 and 3, and would therefore have sub-divided the enclosure. Enclosure 7 (38) To the south-west of E6 lay a small sub-rectangular enclosure (c 10 x 4 m), with only one part of its western boundary (38) being sectioned. This was 1.5 m wide and 0.4 m deep and contained a single fragment of 2nd century pottery. None of the others ditches were excavated, although the northern boundary appeared to curve southwards, and the southern boundary terminated 1.8 m short of linear ditch 8 (see below). It is uncertain if they were contemporary, but if so, this gap may have formed an entranceway. A parallel gully lying 4 m south of the southern E7 boundary was located during salvage operations, and may have formed part of a larger enclosure. Enclosure 8 To the north of E7 lay another small enclosure (10 x 6 m) none of which was excavated. Its eastern boundary may well have been linear ditch 8, as this shared the same alignment as that to the west. Enclosure 9 (9) A small enclosure (c 7 m²) to the west of E8 had one ditch section excavated (9), which was shallow and irregular (c 1.5 m wide, 0.3 m deep), and filled with gravel and clay. It was devoid of any finds. During later salvage work, both of the north-south ditches were found to continue southwards, though curving slightly to the east. Enclosure 10 (49, 100) Just to the south-east of - and partially overlain by - the Phase 2 rubble building platform was enclosure 10 (c 7 m²). Its southern entrance was formed by two short gullies, the positions of which created a short angled passage. The terminal of the eastern gully (49) was 0.6m wide and 0.15m deep, and contained no finds. The western boundary (100) was part of a triple ditch system overlain by the rubble platform (see below). Enclosures 11 and 12 (8/13) The southern parts of two enclosures were located to the north of E 8/9. E11 was c 7 m across and traced northwards for 10 m, and had a small (1 x 2 m) sub-rectangular enclosure abutting its southern boundary. Further east, E12 was sub-divided by a narrow ditch/gully, and probably used ditch 8/13 as its eastern boundary. Within an excavated section of this ditch was found part of a triple vase (SF 3). No other parts of these enclosures were excavated. Linear boundariesDitch 8/13 Ditch 8/13 was traced for 75 m running north-south through the central part of the site, although most of this was located only within salvage areas and therefore not excavated. Three sections were dug in the area of the enclosures, where it cut through E6, with dimensions approximately 1.7 m in width and 0.3 m in depth. It is likely that this ditch formed the eastern boundary of the group of small rectilinear enclosures (E7-12) with which it shared a common alignment, although it seems to have still been in use into Phase 2. It may also have formed the western boundary of a large (c 60 m²) rectilinear enclosure (see Fig. 6.1.3: Phase 1). Finds included iron nails, fired clay daub and oven fragments and 29 sherds of pottery dating from the 2nd to early 3rd century. Two further sections of north-south ditch to the north (34, 40) were probably part of 8/13. The only finds recovered were 6 sherds of Roman pottery (O32), unable to closely dated. Triple ditches 100-102 To the west of the main enclosure groups were three parallel ditches, one of which (100) formed the western side of E10. They were not substantial features, ranging from 0.7 to 1.3 m wide and 0.1 to 0.3 m in depth, though a reasonable quantity of 2nd century pottery came from their fills. The only other find comprised a copper alloy ligula. Ditch 102 was connected to a ‘mesh’ of probable drainage channels to the west (18, 52), all lying under the later Phase 2 rubble platform. Water would thus have been drained from the area of the gravel island towards a waterlogged peaty sump in the area of the probable midden (54). A single piece of fired clay daub was the only find recovered from the channels. Ditch 16 To the north of the rubble spread was a U-shaped ditch (0.8-1.05 m wide, 0.4 m deep) which seems to have formed part of an elongated sub-rectangular enclosure. A small amount of pottery (possibly 2nd century AD), an iron plate and a number of iron nails were the only finds recovered. A 1.6 m length of north-south gully (45; 0.5 m wide, 0.1 m deep) was traced within the enclosure. A shallow butt-ended gully (36) ran westwards from 16 for 3 m and was filled with waterlogged peat. As it lay at the lowest lying point of the central depression, it probably acted as a drainage channel. A group of gullies (59) lying to the east of the sub-rectangular enclosure may also have served as drainage channels, taking water away from the north-eastern flank of the main gravel island. A more substantial ditch/gully ran north from this group parallel to the enclosure, and could well be the same as ditch 44 to the north. Triple ditch/gullies 41-3 Two ditches and a closely spaced double gully, all running north-south, were traced for a short distance in the north-eastern part of the site. Furthest east lay a shallow double gully (41; 0.45 m width, 0.1 m deep), while to the west lay a more substantial ditch (42; 1.9 m wide, 0.4 m deep). 1.3m west of this was ditch/gully 43 (0.6 m wide, 0.2 m deep), which was traced for longest in the salvage area to the north. No finds were recovered from these feature and their function is uncertain, although they were on the same general alignment as the other ditches in the area and therefore presumably contemporary. Stone-lined channel (33; Fig. 6.2.2: Stone channel)Feature 33 ran east-west north of the occupation area through the central depression of the field. This comprised a well constructed stone-faced channel (1.5 m long, 0.25 m wide, 0.3 m deep) made up of five levels of drystone walling and a roughly cobbled floor sloping north-west. It was set within the eastern end of a purpose-built trench, and abutted a ponded area (32; 7 m across, 0.5 m deep) in such a way to suggest that the stone structure may have acted as a sluice mechanism to control the overflow of water away from this hollow depression. No finds were directly associated with the stone channel, but a small amount of 2nd century pottery came from black peaty clay of the ponded area. The structure could therefore belong to either Phase 1 or 2 or indeed both. A number of other features were partially excavated in the south-eastern area of the field, but did not form any readily identifiable pattern. A shallow curving gully (15; 0.7 m width, 0.1 m deep) to the west ran into an area of grey silty gravel, with other possible gullies also emerging from it. Further east lay what appears to have been a triple north-south gully (17, 103-4), possibly related to the other triple ditch line, 18 m to the east (100-102). Lying between the central rubble platform and ditch 16 was a c 2 m long section of east-west ditch (37: 1.1 m wide, 0.4 m deep), with a small gully leading southwards from it. Both lay on a similar alignment to the enclosures to the south (E10) and were probably associated. The ditch contained over 1200g of pottery (probably 2nd century AD) and a group of iron nails, though this was from the upper layer and may derive from Phase 2 activity associated with the rubble platform. A shallow curving ditch (35; 1.6 m wide, 0.2 m deep) was located just north of the central marshy depression. It’s overall form and function were not established and no finds were recovered. A number of evaluation trenches in the north-eastern side of the WB88 field revealed Roman ditches and gullies sealed beneath alluvial material that were undoubtedly a continuation of the enclosure system from WB85 to the north. This, along with the recovery of eight sherds of 1st-2nd century pottery, suggests that they belonged to Phase 1. Phase 2 (c mid/late 2nd to early 3rd century AD; Figs 6.1.4: Phase 2, 6.2.1a: Plans of excavations: southern area and 6.2.1b: Plans of excavations: northern area )SummaryThe final development consisted of a small rubble platform (4) established over the top of the highest area of the site, which would have formed a relatively well-drained island of gravel. This platform presumably served as the base for an essentially timber-framed building resting on stone footings, and was associated with large quantities of Roman fine and coarseware pottery, dating primarily from the late 2nd to early 3rd centuries AD. Other probable Phase 2 features comprised a number of likely middens (2, 54, 105, 106). Linear ditch 8/13 was also probably still in use at this time. It is uncertain if there was any occupational gap between Phase 1 and Phase 2. Building Platform (4, 30, 51, 55-8; Fig. 6.2.3: Rubble platform )An area of ditches (100-102, 18, 52) approximately 10 m² was surfaced by rough limestone rubble paving and light stone footings (4). This building platform seems to have served as the base for a timber-framed unit resting on, rather than cutting into, the gravel. Two pits (50, 53) cutting through the earlier Phase 1 drainage gullies were also covered by the rubble, and one of them (50) contained a stone packed posthole (51; 0.5 m dia, 0.2 m deep). Large sherds of a late 2nd century pottery vessel were recovered from the base packing of the posthole suggesting that it was part of the building structure, although no other postholes were identified. Four areas (55-58) of limestone paving were located within the rubble spread, which may have formed part of the foundations for the building, or it’s interior. Three of these paved areas (55, 56, 58) had definite evidence for faced edges, though the exact shape and dimensions of any building remain unknown. Another area of more rough paving to the north-west (30; not on plan) may well have been part of an external courtyard. A prolific quantity of pottery, tile and other domestic debris was recovered from layers above, below and within the rubble platform, thereby helping to confirm the presence of a domestic building. These finds included anomalously high levels of fine wares, both of British and of continental origin, along with quern fragments, iron structural fittings and two coins (late 2nd and early 4th century). Diagnostic metalwork and pottery all date it clearly to the late 2nd/early 3rd century AD. A number of waterlogged depressions were located across the site that seem to have served as middens for the disposal of domestic waste, mostly dating from the 2nd up to the early 3rd century AD. Feature 2 within E3 was a shallow peat filled depression (c 2.6 m wide, 0.15 m deep) within which was a small gully (5). The depression contained large quantities of 2nd-3rd century pottery in its upper fill, along with smaller amounts of flint and animal bone. The large unabraded sherds were similar in nature to those from other probable midden deposits. A much larger shallow depression (19; 8 m across, 0.35 m deep) was located within the central marshy area to the west of ditch 16, and contained 2nd century pottery, two coins (2nd and 3rd century), structural fragments, animal bone and a copper alloy ring with intaglio (SF 106). The most extensive midden deposits came from 54, a large area of peaty black gravel just to the south-west of the rubble building platform (4). Finds included large volumes of 2nd and 3rd century pottery, whetstones, quern fragments and an array of metalwork, mostly miscellaneous iron fragments. Many animal bones were also recovered and it is clear that this represented the main dumping area of domestic refuse for the inhabitants of the central building. Two other middens (105, 106) were located during salvage work to the south and north-west of the main domestic area. Both produced quantities of 2nd and 3rd century pottery along with a range of metalwork, recovered by metal detecting. Midden 106 contained eight lead weights, representing nearly all such objects found at the site. Other probable Phase 2 featuresA small amount of late 2nd-early 3rd century pottery from the upper fills of ditch 8/13 suggested that this linear boundary may still have been in use during Phase 2, perhaps representing the eastern limit of the main area of occupation, and forming the western boundary of a large rectilinear enclosure (see above). It is possible that parts of the western enclosure group (E7-12) may also belong to this period, although none of the minimal amount of pottery recovered can be confidently dated beyond the 2nd century AD. During salvage operations to the north-west, a square enclosure (107; 16 m²) was located just south of the field boundary. The north-eastern side produced most of a 2nd to 3rd century AD pot, suggesting that it was in use during Phase 2. Other unexcavated ditches surrounding this feature were on a similar alignment and have therefore been tentatively assigned to the same phase. Un-phased features from salvage workDuring salvage work in 1983 and 1985, there were many features which although apparently Roman in date, could not be assigned to a specific phase. A trackway (7m wide) running NW-SE was traced for over 60 m in the far west of the site, along with a number of ditches, some of which were clearly not contemporary (Fig. 6.1.2: Location of archaeological investigations). A further series of linear ditches found in 1985 to the north running NE-SW into the depression in the middle of the field. They were on the same alignment as other dated features in the area and therefore presumed to be contemporary. |