Project Eliseg uses modern archaeological methods to investigate the mound on which the Pillar stands and their setting. The aim is better to understand this enigmatic monument and to discover more about the emergence of both Welsh and English early medieval kingdoms on the borders of Wales after the fall of Roman Britain.
Project Eliseg is co-directed by Professor Nancy Edwards and Dr Gary Robinson of Bangor University together with Professor Dai Morgan Evans and Professor Howard Williams of the University of Chester.
Project Eliseg acknowledges generous financial support from the following sources: the School of History, Welsh History and Archaeology at Bangor University, the Department of History and Archaeology at the University of Chester, Cadw (the historic environment service of the Welsh Assembly Government), the Centre for Advanced Welsh and Celtic Studies of the University of Wales and the Knowledge Transfer fund of the University of Chester. The work is being carried out by the directors with students from Bangor and Chester as well as local volunteers.
We are grateful to Cadw for consent to work on a Scheduled Ancient Monument and extend heartfelt thanks to the Davies family of Abbey Farm for granting permission to work on their land and the staff of Llangollen Museum for their unstinting help.
This website gives you information about Project Eliseg, earlier research and the latest discoveries. There are also links to photo galleries and other relevant websites and organisations. We also include information about the historical context of the early medieval kingdom of Powys, early medieval stone sculpture and information about the archaeology of early medieval Britain (c. AD 400-1100). The website includes information and contact details for Project Eliseg’s four directors.
The Latin inscription on the Pillar of Eliseg was recorded by the Welsh antiquarian Edward Lhuyd in 1696 when it was already fragmentary. It is no longer legible, though the area where it was carved on the Pillar can still be made out in oblique sunlight.
The inscription was written in several sections, each beginning with a cross. The Latin for each section is given first followed by the English translation and an explanation of the text.
The mound was excavated and the Pillar was re-erected on the mound in the late eighteenth century. At this time, a new inscription was added at this time, commemorating a local family and their affinity to this ancient monument. That which remains of this old monument, long removed from eyes and neglected, T. Lloyd of Trevor Hall finally restored AD 1779.
The 2010 excavations aim to provide new information to help understand the enigmatic Pillar of Eliseg by:
Discovering more about the mound upon which the Pillar of Eliseg now stands and its date. Is it a reused Bronze Age barrow?
The excavations present a unique opportunity to explore the context of a surviving piece of early medieval stone sculpture found at or close to its original location. Our work is essential to test Nancy Edwards’s 2009 hypothesis about the role of the Pillar as a focus of an early medieval assembly and royal inauguration site.
Given the monument’s location on top of what might be a far older prehistoric mound, the site is a possible classic case of the manipulation of the past over time. Therefore the sequence and date of both the mound and activity around the mound are key questions for understanding the Pillar in relation to the past.
Given the complex medieval and post-medieval history of the Pillar within the Welsh landscape, the project will address key questions concerning the biographies of monuments, their fate within monastic landscapes and (in particular) the nature of antiquarian practice in relation to the designed and idealised landscapes of the eighteenth century.
The main objectives for the 2010 fieldwork are:
We hope that the excavations will allow us to understand the constructional and chronological sequence of the mound, including subsequent alterations.
The excavations will identify the context and relationship of the mound to other possible archaeological features in the vicinity suggested by the geophysical survey by Semple and Turner.
These investigations, in partnership with Cadw, will be followed by works to conserve the mound and better explain the monument to the public. Preliminary Results from 2010
The 2010 excavation focused on the mound (see https://www.tarifs.org/vulcania-tarif). Three trenches were opened to determine its make-up and to try and locate the antiquarian trench. The mound was found to consist of a stone cairn held in place by a well-preserved kerb, mainly made up of large slate slabs and rounded boulders. No dating evidence was found but the construction would be consistent with a burial monument of the Early Bronze Age. These trenches also confirmed that there was no ring-ditch around the monument. No graves or other features were discerned in the immediate surroundings of the monument.
In future research, it is hoped to investigate further possible traces of antiquarian disturbance on the west side of the mound.
A fourth trench was opened in the field to the north of the mound to investigate anomalies suggested by previous geophysical surveys conducted by Sarah Semple (Durham University) and Alex Turner (SAT Surveys), but little of archaeological significance was found. On 31 July a very successful Open Day was held as part of the Festival of British Archaeology. About 200 people were guided round the site and other activities were centred on Valle Crucis Abbey, including displays by the ninth- and tenth-century AD re-enactment group Cwmwd Iâl attracted c. 300 visitors. There was also an artist in residence, Dr Aaron Watson, who has created a DVD photo-animation of the excavation which contains a visualisation of how the cross may originally have looked.
In September 2011, Project Eliseg’s second season concentrated on the west side of the mound and explored an area of possible antiquarian disturbance.
The monument was made from three different types of contrasting stone were employed: large slate slabs, the igneous boulders and the mudstone slabs. All seem to be derived from the immediate locality.
The main body of the cairn consisted of angular/sub-angular slabs of mudstone and it appears the cairn is multi-phased. Phase 1 may be interpreted as a kerbed platform cairn. The kerb comprised of large slate slabs and rounded igneous boulders. Behind this kerb is a tightly packed deposit containing angular slabs of mudstone.
The edge of the cairn was well defined by a substantial kerb made up of two different types of stone, large slate slabs and contrasting rounded igneous boulders. A small rectangular stone cist was exposed and recorded but not excavated. Whilst clearing this feature for recording all associated deposits were sampled for further processing. The processing of these deposits revealed evidence of charcoal and a fragment of cremated cranial bone. While undated, it appears to be a Bronze Age burial cist.
Final stage of the excavation, 2011
In the upper, second phase, a large slab of blue/grey slate was removed from just in front of the south-facing section of the trench revealing the end of a large cist of unknown date. The section was cleaned and a small sample was removed and found to contain small quantities of fragmented bone. Unfortunately the material recovered was too small to assert whether from human or animal.
Excavation confirmed that there was no surrounding ditch associated with the mound and that the dipping in the subsoil was natural confirming the conclusions drawn in 2010.
The area of antiquarian disturbance was distinguishable from the main body of the cairn material by a shallow scoop within the cairn containing small fragmented stones. The fill of this feature contained quantities of post-medieval ceramics.
The most recent episode of cairn construction comprised of a capping of greyish/blue rounded river cobbles, perhaps associated with the remodelling of the cairn at the time when the pillar was re-erected in 1779. This later interpretation is supported by the partial sealing of the area of antiquarian disturbance by this context.
The top-soil removed at the outer end of Trench A revealed a range of post-medieval ceramics and other finds consistent with manuring and other farming activities.
Despite the inconclusive and incomplete nature of the 2011 excavations, Project Eliseg has for the first time revealed important new information about, and raised public awareness concerning, a long-neglected and extremely important historic monument.
The third season of fieldwork by Project Eliseg took place between 26 August – 15 September 2012. The excavations consisted of a single trench measuring 5m × 6m in size located on the western side of the monument. The position of this trench incorporated parts of Trenches A and D, previously opened during 2010 and 2011. Only a portion of the exposed excavation area was taken down to the natural subsoil to establish a full profile through the monument. A standing section was established along the south-facing section of 2011 trench edge. The top western edge of this trench was stepped at no more than 45 degrees and to a maximum depth of 1m. The full extent and depth of excavation was agreed with the Cadw regional Inspector during the course of the excavation.
The purpose of this excavation was to: (1). Resolve the stratigraphical sequence of monument construction. (2). Excavate and record a possible long-cist grave identified in section in the top of the cairn during 2011 (This feature was visible within the section of Trench A and appeared to extend into Trench D). (3). Excavate and record possible evidence for prehistoric cremation burials/cists located during 2011 behind the monument’s kerb. (4). Establish a chronological sequence for the use of the monument through the identification of suitable samples from sealed stratified contexts for radiocarbon dating.
The Method of Excavation
The trench was de-turfed and excavated by hand. The trench was excavated as an open area in order to maximise horizontal control over the excavation area. Vertical control was maintained through the establishment of a series of running/cumulative sections through the excavation area. The trench was positioned to establish a meaningful standing section through the monument. On completion of excavation a geo-textile membrane was placed over the lower sides and base of the excavation prior to backfilling in order that the position of the trench would be easily identified in the future. The trench was then backfilled and the surfaces re-profiled.
The earliest phase consists of a low oval stone cairn of approximately 20 × 15m in diameter on top of the old land-surface on a slight rise. The edge of this cairn was well defined by a substantial kerb made up of two different types of stone, large slate slabs and contrasting rounded igneous boulders. The kerb stones were placed within a foundation trench that cut through the buried land surface and into the natural sub-soil. The body of this primary cairn was formed by large, closely set river boulders, creating what appeared to be a relatively level platform. An attempt to create a level platform was evidenced in the presence of large flat slabs set upon the river boulders. Around the kerb these slabs were set as through-stones that would have provided extra support to the kerb. The cairn can loosely be classified as a kerbed platform cairn and as such should date to around 2000 BC.
Within the body of this cairn, a small rectangular cist was identified. The cist measured 0.4m × 0.3m and comprised four slabs set on end to form a small stone box. The base of the cist was formed by a single slab. The cist had clearly been disturbed by previous antiquarian excavations. The capstone from the cist was missing and its fill consisted of the same loose mixed deposit as the backfill of the antiquarian trench. The fill of the cist was devoid of finds or burnt bone. However, within the backfill of the antiquarian excavation trench around the cist, small discrete deposits of burnt bone and charcoal were identified, suggesting that the contents of the cist had been cleared out and redeposited by the previous antiquarian intervention. At the base of the cist small quantities of burnt bone were recovered from between the base and side slabs of the cist, thus confirming the original use of this cist as a burial structure. The burnt bone retrieved from the vicinity on this cist may provide a date for the primary use of the cairn and basic biological and technological data on the nature of the burial and funerary process.
The secondary cairn
Phase 2 is represented by a heightening of the phase 1 platform cairn. The main body of the phase 2 cairn consisted of angular/sub-angular clast-supported slabs of mudstone. The distinction between these two constructional phases was clearly defined in the south-facing section but was destroyed in the north-facing section as a result of antiquarian disturbance.
A large cist, originally presumed to be an early medieval long-cist grave, was exposed in section in 2011. On excavation this has been reinterpreted as a large Early Bronze Age cist. This was constructed of four large slabs of stone set on end to form a rectangular stone box. The base of the cist was formed of a single slab and sealed by a large capstone. The cist measured 0.5m × 0.3m and was 0.43m in depth. The cist was excavated in quadrants and 50mm spits. Each quadrant/spit was bagged and labelled separately to aid future analysis. Excavation demonstrated that the fill of the cist had suffered from extensive tree root disturbance which had caused substantial mixing of the deposit. At the base of the cist a second undisturbed deposit was identified. No finds or burnt bone were identified during the excavation process from the fill of this cist.
The cist was clearly cut into the body of the secondary cairn and sealed by a layer of large flat slabs that appeared sporadically across the entire surface of the cairn. These slabs formed a discrete sealing layer across the cairn and must have been visually impressive when constructed.
ion To the west of the excavation area and immediately behind the kerb, a second cist was identified. This was 0.33 × 0.4m in size and constructed of four slabs set on end to form a box. The base of the cist was formed by a single slab and was sealed by a cap stone. On lifting the capstone it was apparent that the contents of the cist were intact and that burnt bone was present within its fill. The cist was excavated in quadrants and in 50mm spits, each bagged separately to aid further analysis (Brickley and McKinley 2004; McKinley and Roberts 1993). The deposit was subjected to ‘total earth recovery’ (McKinley 1998, 2000) in order that all potential pyre material could be identified in post-excavation analysis. In excess of 7 kg of burnt bone was recovered from this single context. The weight of a single Bronze Age cremation deposit in the UK varies considerably, but falls within a range of 57g – 2200g (McKinley 1997: 139). It is clear that the cremated bone from cist 3 must comprise multiple individuals. Post-excavation analysis of this deposit is being carried out by Geneviève Tellier (University of Bradford). Whilst detailed osteological analysis is still on-going, it is clear the cremated bone deposit is comprised of multiple adults, juveniles and infants. A bone pin and flint knife were identified in the fill of the cist.
This cist was inserted into the cairn by the removal of a section of kerb and as such, it is unlikely to be part of the primary platform cairn. The cist was overlain by the large slabs that also overlay cist 2. On stratigraphical grounds cist 3 belongs to the same horizon as cist 2, both of which were inserted into the second phase of construction.
The area of antiquarian disturbance was distinguishable from the main body of the cairn material by a shallow scoop within the cairn containing small fragmented stones. The disturbed area was very irregular and its edges were difficult to identify; its dimensions were approximately 2m × 3.8m and it was 0.5m in depth. However, the presence of this intervention was clearly defined in section.
This context clearly cut into the main body of the cairn and would appear to represent an episode of digging and partial backfilling. The fill of this feature contained post-medieval ceramics. These artefacts have yet to be studied by specialists but their presence is consistent with the identification of the feature as the antiquarian intervention made in the mound in 1773 at the instigation of Trevor Lloyd, the landowner, prior to the re-erection of the Pillar in 1779. It was later claimed that they had found an inhumation ‘guarded round with large flat blue stones, and covered at top with the same; the whole forming a sort of stone box or coffin’ (Simpson 1827, 134–5), suggesting that a further Bronze Age burial cist may have been uncovered.
Antiquarian intervention was restricted to the southern part of the excavation area leaving the northern part of the area relatively undisturbed. Confirmation of the nature of the antiquarian intervention was demonstrated by the marked contrast between the south-facing and north-facing sections of the excavation area. The south-facing section, which contained cist 2, was relatively undisturbed and a clear sequence of cairn construction could was identified. In contrast, the north-facing section was comprised of a jumble or stones and voids with no discernible structure. This area of disturbance appears to continue beyond the top of the north-facing section and below the re-erected cross shaft. If this interpretation is correct, it would suggest that the original antiquarian trench was located on the top and western slope of the cairn. This area of disturbance matches well with previous interpretations of the antiquarian intervention (Edwards et al. 2011) and confirms the original interpretation of the hollow present on the west face of the monument as its location (Edwards et al. 2010–11, 2010; Turner 2008).
The most recent episode of cairn construction comprised a capping of greyish/blue rounded river cobbles . This context was found over the entire extent of the excavated area of Trench A, but was only 0.15–0.2m in depth. This capping of the cairn stood in marked contrast to the underlying cairn material which was made up of angular blocks of silt and mud-stone. This context, whilst disturbed in places due to root disturbance, extended into and partially sealed the area of antiquarian disturbance. There are two possible explanations for this context. (1) The capping of cobbles was the final stage of cairn construction at the site and thus potentially of a prehistoric date. (2) The context is associated with the remodelling of the cairn at the time when the pillar was re-erected in 1779. The later interpretation is supported by the partial sealing of the area of antiquarian disturbance by this context.
On the top of the cairn, and surrounding the cross-shaft, a raised lip of re-deposited sub-soil was identified. This deposit was placed directly on top of the cairn and overlay it. This deposit represents a late phase of activity on the monument, either contemporary or later than the re-erection of the cross-shaft. It has been suggested that this deposit represents a deliberate attempt by Trevor Lloyd to manipulate the appearance of the monument. The placement of this deposit hides the coursed stone plinth, onto which the cross shaft and base have been placed and enables them to be viewed from his summer-house which was located a short distance away at Valle Crucis Abbey to the south-east.
Other potential evidence for landscape alterations, carried out to enhance the aesthetics of the monument, are suggested by the sharp break of slope apparent to the south of the monument and no geomorphological reason can be given for its presence. It is possible that a landscaping event may have taken place here with substantial quantities of the ground surface removed. This landscaping gives the impression of the cross shaft and base being surmounted upon a substantially larger and more impressive mound when approached or viewed from the south east. We would tentatively argue that both remodelling events were carried out by Trevor Lloyd at the time of the re-erection of the cross-shaft and base as part of a larger scheme to create a romantic landscape.