April 2008-March 2009

April 2009- September 2009

GGAT 89 Prehistoric and Roman Settlements and other sites

This project is designed to run over two years, the first year covering the western and northern parts of our area, comprising the unitary authorities of Blaenau Gwent, Bridgend, Caerphilly, Merthyr Tydfil, Neath, Rhondda Cynon Taff, Swansea and Torfaen. These Unitary Authorities include most of the uplands, and the sites studied this year area had a relatively high proportion of surviving stone structures, as was initially anticipated.

The dataset compiled last year for the initial assessment phase of this project contained 133 sites in these eight UA areas. Eighteen sites were recorded on the HER as having been noted on air photographs, but with insufficient information to allow them to be assessed properly. The photographs were accordingly viewed at the Central Registry, with the result that eight have been discarded as either destroyed or not falling within the remit of the survey. At eight sites, the air photograph information was so good that copies were obtained for plotting. In total, 278 photographs were viewed. Seventy sites in all were identified as having potential for field visits. Site visits were carried out from the beginning of September 2008 until the end of January 2009.

The sites were located using a GPS handheld device, enabling the NGR for each monument to be corrected if necessary, and where possible as a reliable GPS reading could not always be obtained. Where sites were not found at the given NGR, the surrounding area was searched. In case of discrepancy between the height Ordnance Datum (OD) given by the GPS reader, and the height listed in the NGR database, the latter was considered more reliable, and was therefore preferentially retained.

Each site was recorded using a pro-forma, noting present description, measurements, position, aspect and prospect, and present land use. The survival, amenity value, condition, fragility and vulnerability were assessed on a separate pro-forma. The record forms closely followed the pro-forma developed for the Ritual and Funerary Monuments project (GGAT 72). Horizontal measurements were taken using a measuring tape, and the surviving height over the surrounding ground was assessed using a photographic scale. Photographs were taken using black-and-white print and digital images. A sketch was made for unscheduled monuments where the monument contained distinguishing features and a RCAHMW survey did not already exist. New sites noted during the course of the field visit were recorded using the same methodology.

Some of the sites visited were very well preserved and potentially of national importance. In most instances, the sites are located in a stable environment and are not under threat of being further eroded or destroyed in the foreseeable future. A majority of them are situated on open access land that is used as pasture. As a general rule, upland sites are still upstanding stone built structures, and they in a better state of preservation than sites that are lying on lower grounds. This bias originates in the contrasted land-use, with lowland grounds being more likely to have undergone centuries of ploughing. A few upland sites lie on karst geology, and are vulnerable to collapsing into sinkholes. The fieldwork was carried out under varied weather conditions, which occasionally had a negative impact on the recording quality.

A handful of new sites were identified during site visits, although not all of them fell within the remit of the project. Two new sites were added to the list of Later Prehistoric Undefended Settlements and Roman Rural Settlements, and the HER database has been updated accordingly. These sites consist of two platform houses that were identified within the Glais Brook enclosure complex. There is considerable overlap between the different monument types and many groups are made of a combination of two or more types. An aggravating factor that affected the recording work came from the complexity and extent of a number of settlement sites that were grouped under a single PRN. In the end, we updated/entered into the HER database seven settlement sites, twenty-one enclosure sites, fourteen hut circles sites, four platform sites, three wall sites, two field system sites, two field boundary sites, four bank sites, and two sites have been classified as earthworks, as it was impossible to refine their typology further. It was not possible to visit one enclosure site, as we were unable to obtain access. Two others could not be found. One wall site could not be located during the site visit. Medieval and Post-medieval platform sites were surveyed during an earlier Cadw-funded project (the Deserted Settlement Survey), leaving the prehistoric and Roman sites to be dealt with in the present survey.