For copyright information please contact ELC Archaeology Service
- HER number: MEL1392
- Site Name: White Castle
- Grid Reference: 361338 668606
- Civil Parish:
- Summary: Hillfort
- Description: NT66NW 1 6135 6860.
(NT 6135 6860) White Castle (NAT) Hill Fort (NR)
OS 6" map (1957)
(1-2) White Castle, an Early Iron Age fort, stands on a promontory (1,000ft OD) defended on three sides by steep slopes. It is oval on plan, measuring internally 230ft by 180ft, and is surrounded by triple ramparts which may have been topped by stone walls. At least three possible hut circles, visible at the time of the RCAHMS visit in 1913 were not seen during the marginal lands survey (RCAHMS MSS).
RCAHMS 1924, visited 1913; R W Feachem 1963
(3) This fort is generally as described by the RCAHMS, although on the NE face there are clearly only two scarps, the third being the natural slope. However, around the N and NE, between the inner scarp and the RCAHMS second scarp, there is a further scarp, an added line of defence, no doubt upon which was placed a wall; at this point there is a clear break in the defences, but whether or not this is original cannot be ascertained. Within the interior and between two of the ramparts there are several scoops, representing the site of timber huts.
Resurveyed at 1:2500.
Visited by OS (RD) 26 August 1968
(4) A watching brief was maintained during pipe-laying operations outside the fort on its SW and W sides. With the exception of an enigmatic feature (connected with hill drainage ?) seen in the pipeline section W of the fort, nothing of archaeological significance was noted.
I Ralston 1973
(5) In August 2010 Rampart Scotland undertook the first season of fieldwork in the Hillforts of East Lothian project at White Castle hillfort, Garvald. The topographic survey was undertaken across the hillfort, although dense bracken on the north and west hampered the results. The survey increased the total number of recorded potential hut platforms to at least 17. Within these there are clearly two distinct sizes: c 5m in diameter and c9m and they are clustered to the north-west and west sides of the site. All of the ramparts are clearly more impressive on the southern side and at present there is no clear trace of the inner rampart to the north of the site, or of the outer rampart to the west of the site. In addition, Rampart 3, continues further to the southwest than previously recorded. The putative second south-east entrance identified by previous surveys, is now interpreted as a breach corresponding to the change in nature of the ramparts from the south to the north of the site. Of some considerable interest is the discovery of another defensive feature, Ditch 3, beyond Rampart 3, measuring c 40m east-west by c5m wide and in effect cutting off the causeway. It is argued that this feature must predate Ramparts 2 and 3 as they both serve the same purpose more effectively. Finally, the full extent of bracken coverage on site has been mapped; at present it is not clear if this is increasing or not.
The erosion survey recorded rabbit damage, sheep scrapes, cattle and visitor tracks, areas of mole activity, water run-off damage and a single spade cut hole, assumed to be connected with a camp fire or metal detecting.
(5) In August 2010 Edinburgh Archaeological Field Society undertook a geophysical survey at White Castle hillfort, as part of Rampart Scotland's Hillforts of East Lothian project. The resistivity clearly showed differential features across the interior and the defences. Rampart 1 had a high resistance which may represent the stone foundations identified in Trenches 1-3. Rampart 2 shows both high and low resistance, which may suggest stone footings to the east and looser earth banking to the west, as supported by excavation in Trenches 5, 6 and 7. In addition, the differential resistance may indicate repair work or phasing or the work was undertaken over an extended period or by different work gangs. The gap within the high resistance linear feature is due to the breach in the rampart that is caused by the visitor track. Rampart 3, where surveyed shows low resistance, again suggesting a looser earth bank, again evidenced by excavation in Trenches 10 and 11. The large high resistance anomaly between Ramparts 1 and 2 on the east edge of the survey area, suggests a stone footing in a rectangular pattern. This is approximately the same size as the ‘shieling’ feature that overlies Rampart 1. Within the interior it is unclear whether the pattern of high and low resistance represents geological or anthropogenic features, however, there does seem to be a correlation between topographic features and these geophysical features.
(5) In August 2010 Rampart Scotland undertook the first season of excavation in the Hillforts of East Lothian project. A single staggered trench and 11 test-pits, were excavated. All the features
identified within the trenches were excavated, recorded and surveyed. Charcoal, slag, pottery, burnt bone, possible saddle quern fragments and a single lithic were recovered from the site. Post-excavation analyses including radiocarbon dates are proposed along with three or four further phases of excavation on the site.
(6) Between late July and early August 2011 Rampart Scotland undertook a topographic and erosion survey at White Castle hillfort, as part of Season 2 of the Hillforts of East Lothian project. During Season 2 additional close contour survey was undertaken, building on the results of Season 1. Further surface survey was carried out by the East Lothian Young Archaeologist Club on the location and measurements of hut platforms using a hand held Garmin GPS unit and the previous topographic survey print out. The number of visible hut platforms has now been confirmed at 16 and their diameters recorded; there are clearly three distinct sizes: c 3.5 m in diameter, c 5m in diameter and c 9m. The platforms are clustered to the north-west and west sides of the site. The rampart banks are enhanced on the south and southwest portion of the site, with the north and northwest having no visible bank but steep sided terracing. The entrance gate-ways on the south-western quadrant and the track leading from the causeway to these entrances have been recorded.
The erosion survey re-evaluated the results of Season 1 including rabbit damage, sheep scrapes, cattle and visitor tracks, areas of mole activity and water run-off damage.
(6) Between late July and early August 2011 Rampart Scotland undertook a second season of archaeological evaluation at White Castle, Garvald. Seventeen trenches/test pits were excavated, concentrated on the northeast slope of the monument, across the three narrow terraces. Further trenches were opened across the causeway ditch, the geophysical anomaly that was identified during the 2010 excavation season and also across an old track-way that passes the site. Charcoal, burnt bone, possible saddle quern fragments and two worked lithics were recovered from the site. Post-excavation analyses including radiocarbon dates are proposed along with two further phases of excavation on the site.
(7) In July 2012 Rampart Scotland undertook a third season of excavation at White Castle of a proposed four season research excavation utilising volunteers and professionals. The project involved further topographic survey, combined with geophysical, vegetation and erosion surveys as well as 7 trenches.
The topographic survey identified a number of potential hut-platforms, which were drawn during a detailed walkover and topographic analysis; there are now 18 of these platforms with one rectangular structure to the southwest. The more detailed walkover has demonstrated that the platforms have a chronological depth to their distribution as specific examples overlap others.
The resistivity survey was undertaken by the Edinburgh Archaeological Field Society as part of the Rampart Scotland project using TR/CIA area ground resistance measuring equipment. All readings were taken at 1m intervals in lanes 1m wide in 20m by 20m survey grids, giving a total of 400 measurements in each grid. Continuing the 40x60m grid that had been carried out in 2010 a further 6 20m grids were surveyed to the north and west, while additional partial grids were included to the north. The known geology of the site is a greywacke, which has not affected any readings. The results of the geophysical survey 2012 greatly enhanced the previous extent of data, covering the entire upper enclosed area and most of the banks and terraces, excluding the northern slopes.
The resistivity clearly showed differential features across the interior and the enclosing banks and terraces which may be of archaeological significance. The banked material and underlying geology was clearly recognisable. Within the interior there seems to be a correlation between topographic features and geophysical features. By comparing and superimposing the topographic and geophysical data sets the hut-platforms were clearly visible in the resistivity, which was registering the platform rubble. Re-examination of the resistivity plots based on this understanding of the overlying topography and recognisable hut platforms suggests further locations which are not directly visible on the surface.
The excavation expanded upon three of the 2011 trenches (12, 13 and 18) to confirm the nature of Hut-Platform 08 and nature of the putative palisade identified in Trenches 12, 13 and 18. The presence and nature of the palisade was confirmed and it was clear that it had been cut into underlying midden material. In addition, Hut Platform 08 overlay it – itself covered by a rubble spread from a further Hut Platform up slope.
Other trenches examined Hut Platform 04 and the two of the south-eastern gaps in the ramparts to confirm if they were breaches or entrances. The gaps were confirmed as entrances, although they had been subsequently eroded. Hut Platform 04 comprised a substantial cut terrace into bedrock with a foundation of larger rocks into which were cut a slot and a series of post-holes. The slot feature may represent a drain.
Finally a single 1m by 1m trench was dug over a molehill to determine what if any impact they have on the underlying archaeological deposits. No impact was discovered.
The excavation recovered charcoal and possible saddle quern fragments and three radiocarbon dates were obtained from the site. One further season of fieldwork is proposed.
The accumulation of data from White Castle continues to reveal the monument’s complexity. There are at least two phases of rampart construction (the outer two and inner) and none of them represent a complete circuit, perhaps undermining any defensive function. In addition, some of the visible occupation structures appear to be linked to the inner rampart, although some clearly post-date it. However, it is extremely likely that the hut-platforms represent the final phase of occupation on the site, having been built over earlier occupation debris.
It is also now also seems plausible that the rampart banks were penetrated by multiple entrances and that the entrances in inner and outer banks were aligned to provide direct access. This again undermines any defensive interpretation to the site given that defence would rely of staggered entrances to slow down attack given that entrances are the most vulnerable part of a defensive circuit.
(9) During July 2013, Rampart Scotland undertook their fourth and final season of archaeological work at White Castle. The programme of archaeological work comprised topographic and erosion survey together with the excavation of six trenches. Excavation concentrated on the northern and western sides of the hillfort, and explored Platform 16, the putative north-western Inner and Middle Rampart entrances, a section of the middle ditch (D2) at the south of the hillfort, a post-Inner Rampart structure (S1) on the inside of the Inner Rampart and a mound on the Outer Rampart, previously thought to contain a cist.
The excavation recovered charcoal, an abraded sherd of hand-thown pottery and a piece of unworked lithic. The putative north-western Inner and Middle Rampart entrances were determined to have been formal entrances. The full extent of the middle ditch (D2) was not determined, but it was found to measure at least 3m wide by 1.8m deep and to have been filled with collapsed material from the Middle and Outer Ramparts, together with deposits from an entrance in the Outer Rampart examined in 2010. The earliest deposit excavated within the ditch may have been an initial stone facing to the Inner Rampart, or deliberate infilling to create a causeway. The mound on the Outer Rampart was found to be constructed of discrete episodes of dumping, which concealed five post-holes asscoiated with concetrations of stone, suggesting the mound was a discrete feature. The mound lies adjacent to a gap in the Outer Rampart, which together suggest that the mound functioned as part of a formal gateway in the Outer Rampart.
The results of the excavation identified only a single feature associated with Platform 16, a small sub-oval pit, dated to c. AD 1100. Structure 1 appeared to have utililsed material from the Inner Rampart in its construction. An occuaption layer was identifed within the structure overlain by burnt in-situ twiggy matting, together with a raiseded bank or soil bed and the remains of a possible hearth. Rdiocarbon dating of the twiggy material suggested that Strucutre 1 dates to sowmtime between the 15th and 17th centuries.
Overall, the archaeological work to date suggests that the Inner, Middle and Outer Ramparts were constructed in different styles and may never have formed a continuous whole. Based on the radiocarbon dating, the phasing of the ramparts is suggested to be as follows: the construction of the Middle Rampart (between c. 600 to 400 BC); the construction and subsequent destruction of the Inner Rampart (between c.400 to 350 BC); and the construction of the Outer Rampart (between c.300 to 200 BC). Internal settlement and occupation is suggested to date to two phases: the first associated with the Middle Rampart, and the second associated with either the Inner or Outer Rampart. Two post-hillfort phases are suggested by the dating evidence obtained from Platform 16 (around AD 1100) and Structure 1 (between the 15th-17th centuries).
- For more information contact: East Lothian Council HER
- Related Places:
- Associated Periods:
- (1) Bibliographic reference: RCAHMS. 1924. The Royal Commission on the Ancient and Historical Monuments and Constructions of Scotland. Eighth report with inventory of monuments and constructions in the county of East Lothian. 35, No.52. figs 71, 72.
- (10) Digital archive: Connolly, D & Cook, M. 2014. Rampart Scotland project 001: The Hillforts of East Lothian Season 4: White Castle, Garvald, East Lothian Data Structure Report.
- (2) Bibliographic reference: Feachem, R W. 1963b. A guide to prehistoric Scotland, 1st edition. 123.
- (3) Unpublished document: Ordnance Survey. Ordnance Survey Site Visit.
- (4) Bibliographic reference: Ralston, I. 1973. 'East Lothian, White Castle, Garvald, Iron Age fort', Discovery Excav Scot 1973, p.64. 64.
- (5) Digital archive: Cook, M, Connolly, D & Kdolska, H. 2010. White Castle, Garvald: The Hillforts of East Lothian Season 1 Data Structure Report.
- (6) Digital archive: Cook, M & Connolly, D. 2011. White Castle, Garvald: Data Structure report and site records.
- (7) Unpublished document: Cook, M & Connolly, D. 2013. White Castle, Garvald Data Structure Report.
- (8) Digital archive: Cook, M & Connolly, D. 2013. White Castle, Garvald Data Structure Report.
- (9) Unpublished document: Connolly, D & Cook, M. 2014. The Hillforts of East Lothian Season 4: White Castle, Garvald, East Lothian; Data Structure Report.