SECTION 5.1: SOMERFORD KEYNES NEIGH BRIDGE INTRODUCTIONGEOLOGY, TOPOGRAPHY AND LAND USE Background to the excavationsThe site at Somerford Keynes Neigh Bridge was originally identified in aerial photographs as a dense complex of cropmarks covering an area of some 2.5 hectares (RCHM, Somerford Keynes 2). The archaeological potential of the site was enhanced by metal detectorists, who recovered large numbers of finds from the location, and subsequently informed OA staff of their discoveries. Proposed gravel extraction at the site led to a rescue excavation undertaken by OA between November 1986 and May 1987. The work was funded by Amy Roadstone Corporation (AMC), Manpower Services Commission (MSC) and the Cotswold District Council. Further intermittent salvage recording took place during the early stages of gravel extraction, up until the spring of 1988. The site was investigated in order to test theories resulting from the HBMC funded Cotswold Water Park research program (see section 1). Despite limited available resources, the combined excavation and salvage recording revealed a substantial part of a late Iron Age and Roman settlement. The exact nature of the site remains somewhat enigmatic, although it is clear that during the early 2nd century it was transformed from a series of sub-rectangular ditched enclosures to a rectilinear system of ditched boundaries and trackways, along with a substantial aisled building. A strong indication of official or even military control is suggested, possibly in the context of a tile distribution depot, while religious activity is also indicated. Evidence for further activity of some kind is suggested in the 4th and 5th centuries, although this could not be associated with any specific archaeological features. Location of the siteThe site is located just to the south of Somerford Keynes village, within the Somerford Keynes parish in southeast Gloucestershire (NGR SU 019945; Fig. 5.1.1: Site location plan ). It lies approximately 100 m south-west of the river Thames, and is now part of Neigh Bridge Country Park, within the western Cotswold Water Park. Geology, topography and land useThe site is located on the immediate floodplain of the river Thames, with the central area (trench 5) occupying a slightly higher level. Geologically the site rests on part of an extensive first gravel terrace to the Upper Thames Valley, overlying Jurassic Oxford Clay, and was overlain by a narrow band of alluvial clay flanking the river. The site appears to have been primarily pastureland prior to gravel extraction. Archaeological background (Fig. 5.1.1: Site location plan )Somerford Keynes Neigh Bridge is located within the westernmost part of the upper Thames valley, a region that has produced much evidence for archaeological activity from the Palaeolithic to the post-medieval periods (see section 1). The site lies just over 5 km south of the Roman city of Corinium and 8 km west of Ermin Street Roman road, while in the more immediate vicinity are a number of known Iron Age and Romano-British settlements along with a series of undated sites known from cropmarks (Fig. 5.1.1). Lying 1 km to the south was a Romano-British settlement spread over 14 hectares, from which fragments of samian pottery were recovered (SMR 2404), while 1 km to the east was another Romano-British settlement which was partially excavated in 1971 (SMR 2406). Work by Oxford Archaeology at Cotswold Community 2 km to the north-east revealed a Roman farmstead and trackway (OA 2004), while a further 1 km north-west of this is a probable Roman settlement as revealed by cropmarks (SMR 2368). Iron Age activity in the area is less well know, although a middle Iron Age settlement was partially excavated at Spratsgate lane (SMR 2361) just 1 km north-east of the present site. The Neigh Bridge site is located in the midst of this fairly dense pattern of Roman rural settlement. Its location at a crossing point of the upper Thames may have contributed to the importance of the site. Excavation methodologyExcavations at Somerford Keynes Neigh Bridge took the form of a salvage operation with very limited funding, using personnel from the MSC scheme. The topsoil across the site (c 0.4 ha) was stripped mechanically, and a number of small trenches were put in, in order to gain a greater understanding of the archaeology as revealed from soil marks (Fig. 5.1.2: Site plan showing trench locations). The initial location of trenches was also partly influenced by the highest concentrations of metalwork as plotted by metal detectorists. A number of these trenches were expanded as necessary, especially in the highest part of the site (trench 5) which contained the most concentrated amount of archaeology with the clearest stratigraphic sequences. In this way it was hoped to increase understanding of the nature and chronology of the site. Many of the lowest parts of the site to the north and east were subject to flooding and so excavation here was very limited. To the east of the site, near the river Thames, were visible earthworks, which appeared to relate to the ditches on the main site (Fig. 5.1.2). A few trial trenches were dug in this area to observe this relationship but unfortunately no subsurface features or finds were revealed. Chronology and phasingThe archaeology of Somerford Keynes Neigh Bridge comprised a mass of inter-cutting ditches, gullies and pits, along with at least one substantial aisled building. Pottery from the whole site indicated occupation from the early 1st to later 2nd or early 3rd century AD, and the phasing of features within the site is based upon this material. Later 3rd and 4th century coins and small finds do suggest later activity of some kind, although this cannot be related specifically to any of the features. Trench 5 contained the most extensive archaeological deposits, and it is only in this area that it was possible to present a coherent system of phasing. However, certain features from other trenches can assigned to either phase 1 or 2/3 with a reasonable level of certainty, based either upon pottery dating, or stratigraphic/spatial relationships with other phased features. Below is a phasing summary for the site. Phase 1: Early 1st century AD - early 2nd century AD (Fig. 5.1.3: Trench 5 phase 1)The earliest phase of activity within trench 5 comprised a sequence of sub-rectangular ditched enclosures and sub-enclosures, varying in size and form, belonging to the later Iron Age and early Roman period. The number of Iron Age and early Roman coins and brooches does suggest a possible religious focus at or very near to the site, although most were recovered from metal detecting and have no secure context. There are a number of smaller ditches and pits from other trenches that contain higher quantities of 1st century AD material, but these do not form any coherent pattern. Phase 2: Early - mid 2nd century AD (Fig. 5.1.4: Trench 5 phase 2)At some point in the early 2nd century AD, the enclosures and sub-enclosures were replaced by a rectilinear system of ditched boundaries and trackways and a substantial aisled building was constructed. This building has been tentatively identified as a tile depot due to the large quantity and variety of tile found in the vicinity and feature 301, a possible tile stack lying just outside the building. In the western part of the site, a substantial curvilinear ditched boundary (16) was dug, with parallel ditches running westwards from it, defining the northern and southern edges of the main occupation area. It is uncertain how many of these ditches were contemporary, as dating evidence is slight. Sculptural fragments of an eagle and shield - possibly relating to the Capitoline triad - were found on the top fill of the curving ditch in trench 13, but may not have been directly associated with the use of this feature. Behind the ditch lay a series of boundaries probably representing successive phases of an enclosure system, although nearly all datable features indicated that activity was restricted to the 2nd century AD. Further to the south-east, another substantial curved ditch was located (9), which - if contemporary - may have acted as an inner boundary. Only a very small part of this feature was excavated, in trench 19. Large quantities of small finds were recovered from metal detecting in the area to the east of this ditch, although none could be specifically related to any features. In trench 5, it was possible to further sub-divide phase 2 on a stratigraphic, if not chronological basis.
Ditch 135 and gully/beam slot 70 may have defined part of an enclosure around the aisled building, with a large 19 m long opening directly to the south of the building. Substantial occupation debris from this phase onwards suggests much more intensive activity than in phase 1.
During this phase, north-south and east-west trackways were built through the site, running just to the south and east of the aisled building. The junction point between the two trackways was blocked by part of ditch 172, although there may have been a causeway across. The trackway ditches acted as boundaries within the site, with the aisled building being located within the south-east corner of the newly created north-eastern enclosure. It is uncertain as to whether the eastern ditches (135, 122, 112) were still open at this time, although the western end of 135 was cut by ditch 52 of the north-south trackway. Phase 3: Mid 2nd - late 2nd/early 3rd century AD (Fig. 5.1.5: Trench 5 phase 3)The mid 2nd century saw the main trackways in trench 5 go out of use, to be replaced by another set of north-south and east-west boundaries. New trackways also appear to have been constructed. Most of ditch 172, and possibly ditch 101, may still have been utilised as boundary markers, and a sub-rectangular enclosure was created in the middle of the trench, with an entrance to the north. Part of ditch 172 was used to define the western boundary of this enclosure, and located within it was a structure thought to be a corn-drying oven. The aisled building seems still to have been in use during this phase, as tiles stacked against its south wall were positioned above filled-in ditch 318 of the phase 2b east-west trackway. Activity associated with these features seems to have ended by the end of the 2nd century or start of the 3rd century AD. Most of the features from other trenches within the site could not be dated any more closely than 2nd century AD, and so could fall into either phase 2 or 3. Phase 4: Late RomanAlthough there does not appear to be any further structural phases within the site, the overall quantity of later 3rd and 4th century coins and small finds suggests continued activity of some kind in the area. Furthermore, the nature of the small finds indicates a continued official state presence. The largest concentrations of late Roman small finds from the site occur to the east of the large curvilinear enclosure (Fig. 5.1.6: Distribution of finds from metal detecting ). The finds suggest that activity continued into the early 5th century. |