MSO9203 - Prehistoric settlement and field system on Codsend Moor (Monument)

A possible hill-slope enclosure is surrounded by a field system covering 8.3 hectares. Other enclosures, a hut circle and numerous field clearance cairns have also been recorded. The remains indicate settlement activity of several phases.

Field system(s), a hut circle, a possible hillslope enclosure, a smaller subcircular enclosure and field clearance cairns positioned on gently sloping ground in enclosed moorland on the south facing slope of Codsend Moor. All features are previously unrecorded. Centred at SS 8875 4034. Archaeological features are visible in three fields, separated by substantial, east-west field boundaries of Post Medieval date. At the time of survey bracken cover was quite thick and coupled with the fragmentary nature of some of the features this made interpretation difficult. This should be noted when reading the following account. (A) SS 8876 4039 - Circular enclosure. This is recorded as MSO10414. (B) SS 8872 4038 - Immediately southwest of the enclosure, a semi-circular platform has been created in a natural declivity by the construction of two stoney banks over it: a north one on the crest of the natural slope and a south, arcing one across the downhill axis of the declivity. The area "inside" has been artificially levelled, perhaps as a solid stance for a building or as an enclosed, agricultural plot. Around the enclosure in all directions is a complex group of fields covering c. 8.3 hectares. Extensive stripping out on the southwest side of the site has made interpretation difficult. There do, however, appear to be three clear elements. Fragments are visible on aerial photographs [1]. Group 1: An irregular group to the south of the enclosure forming an approximately triangular shape with the enclosure at the apex. At (C) these fields are well preserved and bordered by stoney banks or scarps 2 to 3.3 metres wide and 0.3 to 0.9 metres high. The spaces within are cleared of surface stone, with field areas of between 0.03 and 0.09 hectares. Around (D), stripping out has resulted in a fragmented pattern, much more difficult to determine. Traces of irregular fields with stoney banks and scarps of lower relief on their boundaries can, however, still be picked out. Group 2: Predominantly on the margins of the irregular group are two sets of long, striplike fields defined by narrow rickles of small and medium stone, up to 1 metres wide and 0.3 metres high, with an occasional more substantial stoney bank. The sets are similarly aligned at a bearing of 40 degrees, diagonally across the contours heading northeast to southwest, situated immediately northeast of the enclosure and stretched along the southeast margin of the site. A possible third set was noted northwest of the enclosure. This particular layout is reminiscent of the field system two kilometres to the west on Codsend Moor (MSO9193). It is not clear whether these two field types are chronologically distinct. The only hint of a relationship is at (G) where a stripfield bank appears to overlie the hillslope enclosure but the area is disturbed and the observation tentative. There is certainly some integration of plan, with a stripfield boundary, enclosure bank, and irregular field group boundary following a common alignment from (G). It is likely that the visible remains are in fact a palimpsest of many periods, with purely functional differences echoed in field shape. All field remains are of prehistoric types. (E,F): Two large concentrations of local stone, ranging in size from medium to very large boulders. They are unlikely to be of entirely natural origin and are probably enhanced by field clearance of stone not incorporated in boundaries or other structures. (E) confirms the south limit of the irregular fields and in fact impinges in part on them, suggesting an incomplete and perhaps piecemeal field growth. (F) is a denser concentration of stone deliberately heaped into an irregular "cairn" partly overgrown with rush and sedge and c.0.85 metres high at its south end. At least 36 field clearance heaps were recorded, the majority lying along the southeast margins of the site in a distinctly linear concentration. Most are small, stoney spreads as opposed to true mounds, less than 3 metres diameter and usually no higher than 0.3 metres. They clearly relate to robbing of field banks at an unknown date (but probably for post-medieval bank construction) and are often arranged in lines continuing the path of a vanished field boundary. Group 3: At the southwest extremity of the area surveyed, centred SS 8860 4014, is the third group of cultivation remains. The modern field in which they lie has been subject to more improvement than the rest of the surveyed area and consequently the features are smoother and contain less visible stone. A series of short, almost parallel lynchets up to 4 metres wide and 0.8 metres high define a series of terraces descending the hillslope. They are clearly mere fragments of formerly more extensive fields and cannot be linked in any meaningful way with groups 1 and 2. A ruined field wall, now a stoney lynchet or low bank 1.6 metres wide and 0.5 metres high, overlies these lynchets and delimits a small, rectangular area in the north east corner of the present field. It is probably a small, post-medieval field (pecked line on plan). Situated on the fringes of field groups 1 and 2 are three further features. (H) SS 8868 4023 - Hut circle. This is recorded as MSO10417. (J) SS 8871 4047 - Small enclosure or hut "circle". This is recorded as MSO10418. (K) SS 8862 4038 - Oval mound. This is recorded as MSO10419. 200 metres northwest of the main group of features is another prehistoric boundary, running for 283 metres diagonally across the contours from SS 8857 4064 to SS 8877 4082. It is composed of small and medium sized stones, partly turf-covered, and in several places reduced to a southeast facing scarp only. It survives to a height of 0.3 to 0.6 metres and width of 2.5 to 3 metres. A marsh to the southwest and northeast now hides its course. This bank, aligned at an angle of 48 degrees, is probably part of the main group of features, with boundaries of similar alignment. The area of the modern field as a whole has signs of deliberate clearance with occasional patches of cleared stone and some small mounds. It is not apparent when this clearance took place. The following are worthy of note: SS 8867 4062. Rectangular area of concentrated clearance c. 14 metres by 10 metres. SS 8879 4074. Clearance mound, approximately circular, measuring c. 4 metres in diameter and 0.4 metres high. It is composed of medium and large size stones with a central depression. This central hole creates the illusion that the feature could be a small hut circle, but it is not considered to be so. The mound is positoned beneath a steep natural slope. SS 8865 4097. Clearance mound 2.7 to 3.3 metres in diameter and 0.25 to 0.5 metres high, composed of medium and large size stones with a central depression. It is situated on a steep natural slope and, like the previous features, has probably been robbed. [1-5] Elements of the fragementary field boundaries, enclosures, and hut circle of probable of later prehistoric or Romano-British date can be seen as earthworks on aerial photographs of the 1940s onwards. The earthworks transcribed as part of the Exmoor National Park National Mapping Programme project are concentrated in three areas; SS 8860 4014, equating to area E above; SS 8870 4034, part of the triangular enclosure in `Group 1', area B above; and the hut circle (H) at circa SS 8871 4047. The slight difference from the grid references of the above authority probably arise from the greater locational accuracy of the NMP survey. Enclosure `A', at circa SS 8876 4039, along with much of the further field system described above and the associated cairnfield were not confidently discernible on the aerial photographs available to the survey. [6-11] [11] states hut circle (H) or MSO10417 should be sited at SS 8871 4047, not SS 8868 4023 as [3] suggests. However, this first grid reference is the site of feature (J) or MSO10418. It is not clear whether H has indeed been mis-sited or if [11] has mistaken (J) for (H). [12] This record was enhanced as part of the National Record of the Historic Environment to Exmoor National Park Historic Environment Record data transfer project. [13]