The Society for the Protection of Ancient Buildings (SPAB) Old House Project at St Andrews and the Hospitium at Boxley Abbey, Kent
The Society for the Protection of Ancient Buildings (SPAB) Old House Project site is a Grade II* ‘building at risk’ near Maidstone - St Andrews (former) Chapel. When the SPAB bought the building in November 2018, it was hidden behind an overgrown garden; vandal damage had left the site
vulnerable; parts of the roof were leaking. The five-year repair project aims to showcase the best in building conservation whilst providing a live training site for volunteers passionate about old buildings, apprentices and local students.
In 2018 and early 2019, emergency work to the roof and windows was carried out, and a security system was installed. In 2020 bespoke scaffolding was in place, allowing socially-distanced access to the roof so the SPAB could carry out vital roof repairs to the characterful 20th-century post office extension. The SPAB also embarked on extensive archaeological work, with the help of KAS, HAARG, MAAG and KURG, to help them understand the story of St Andrews’ story and how it relates to the nearby Boxley Abbey and much more. In 2021 the Old House Project was named as joint winner of the Museums + Heritage Award for Restoration or Conservation Project of the Year.
Whilst working on nearby St Andrews, the SPAB has held annual working parties at Boxley Abbey.
SPAB volunteers and experts have taken on a range of works, including repairing the abbey boundary wall and timber repairs at the hospitium.
Working parties give those new to hands-on conservation the opportunity to learn from skilled practitioners in a relaxed setting.
Following the SPAB’s acquisition of St Andrews and the initial invitation to the KAS and MAAG to participate in a shared project, KAS Historic Buildings Group members have contributed to the enhanced understanding of the building and the essential associated local history and documentary research, using estate maps and rentals in Kent History and Library Centre especially. Research has been facilitated by invaluable input from Dr Elizabeth Eastlake. Her 2014 PhD thesis on Boxley Abbey (and generously proffered associated documentary translations) is continuing to aid interpretation of the abbey site, local landscape, and related buildings. With the benefit of skills and knowledge of the local history and archaeology, local input by members of MAAG and KAS has included study and review of relevant and essential KAS
Top
Fig 1: View looking west, St Andrews Chapel in the background. SPAB volunteers working on test pits
HISTORY OF ST ANDREWS CHAPEL AND WIDER SITE
by Graham Keevil
The Cistercian abbey of Boxley was founded in 1143/46 as a daughter house of Clairvaux. It thus took its allegiance directly from France rather than any of the English Cistercian houses. Boxley was closed in 1538 as part of Henry VIII’s Dissolution of the Monasteries. Parts of the cloister were converted into a mansion house, but the church was ruinated to prevent re-occupation should the Catholic faith mount a successful revival after the king’s demise. No such reversion took place despite Queen Mary’s efforts, and Boxley has remained in private, secular hands since the middle of the 16th century. Perhaps unsurprisingly, the site today owes most of its appearance to these centuries of domestic use and landscaping – but its monastic origin and history can still be appreciated readily, not least because of its enclosing precinct walls and the great barn-like hospitium.
St Andrews is of 15th-century origin and lays outside the abbey’s precinct wall. It must have been a late afterthought for the monks and may have originated as a chantry or reliquary chapel – perhaps sponsored by a patron who wanted a close association with the abbey. The chapel then passed into secular use after the Dissolution and has been a house for much longer than it was a chapel. The building reflects this history, and we have spent much time untangling how this is represented above and below ground. A dozen test pits have been excavated, a watching brief was maintained when new access was created off Grange Lane to the south, and of course, the building itself has been recorded.
We cannot claim a definitive understanding of how the chapel developed yet, but we know a lot more about it than we did three years ago. Here are a few of our insights.
Our first test pits were dug in July 2019 to examine the foundations of the building, which were of interest to the project engineer and the archaeology team. The chapel itself was built of good quality masonry, with a large offset plinth at the base of its walls. Its foundations continued for just over a metre below the current ground level, and we found late medieval pottery in the trench dug to build them. Much has changed above-ground, but several original doors are still evident (two still in use and one blocked), along with squints – small square windows – which provided a view of the altar
(and perhaps a relic displayed on it)
Top
Fig 2: The brick drain curving away to the northwest. Note the standing water to the east (right) of the masonry, with some present to the left
Bottom
Fig 3: Test pit 3
Other pits had surprises in store. The foundations on the west and part of the south sides, for example, appeared to be almost non-existent. The rest of the south wall and the ‘priest’s house’ to its east were more substantial – albeit of inferior quality to the chapel. However, the two seemed to be bonded together (as do the walls above), which was unexpected. The south wall and west end had been thought of as an addition, probably post-dating the Dissolution. The test pit results, and the dendrochronology, are now calling that into question. It is at least possible that the entire building is, in fact, of late medieval origin (except, of course, for the late 19th-century post office extension on its northwest corner and the inserted first floor).
We have also been excavating in the chapel’s grounds to see whether any archaeology survives there, whether of medieval or later date. Again, this has presented us with a few surprises – pleasant ones for the most part.
For instance, we have found a flint- cobbled surface over much of the area to the east and southeast of the chapel, suggesting that much of this ground was a courtyard. While we were digging this summer, the reason became apparent when severe overnight rain turned the clayey soil into a quagmire. A cobbled surface would have been beneficial in those conditions – no doubt that was the case in the 15th and later centuries as well. We can’t be sure that the cobbles we found were of late medieval date, but this seems likely.
Our best find came from just above the cobbles, where an intact medieval encaustic tile was found in March 2021. This is a much smaller tile than the usual patterned tiles, but it probably dates to the 13th-century. It is a mosaic tile, which relies more on the geometric layout of the tiles than the patterns on them to provide the overall design. Our example is a hybrid, though, because it does feature a simple star on its surface. Remarkably, Peter Tester found an identical tile during his excavations at Boxley Abbey in the early 1970s (published in Archaeologia Cantiana 88 – figure 5, tile 3). Our tile is in excellent condition, with its original glaze intact. It shows no sign of wear, and there are no traces
Top, left
Fig 4: Foundations quarry stone
Top, right
Fig 5: Foundations ragstone
Bottom
Fig 6: Medieval tile St Andrews of mortar on its base or sides. Perhaps it had never been used in a floor, and of course, the tile pre- dates the chapel by around 200 years – but the tile is a tantalising hint of what other discoveries we might make onsite or nearby.
Intriguingly, a tile kiln that had probably been the source of our tile and the one found by Tester was excavated in the 1920s. This seems to have been at Abbey Farm, a short distance to the north of the chapel. The kiln was probably the source of the abbey’s 13th- century tiles and might well have been a profitable enterprise for the abbey: Tester suggested that Boxley was the source for similar tiles at Rochester and Canterbury
Cathedrals, as well as Leeds Priory. Geophysical survey in July 2021 located two kiln-like anomalies in the grounds of Abbey Farm. Is this the same site excavated in the 1920s?
It would be exciting to find out.
BOXLEY AND ST ANDREWS SURVEY WORK
by Kevin and Lynn Cornwell of Hastings Area Archaeological Research Group (HAARG)
The geophysical surveys of these sites were undertaken by Hastings Area Archaeological Research Group (HAARG) and Maidstone Area Archaeological Group (MAAG).
St Andrews Chapel
A detailed survey using Geoscan RM15 Advanced resistivity equipment was undertaken with recordings at 0.5m intervals on 0.5m traverses in a ‘zig-zag’ recording mode. The survey totalled 6,312 readings, covering
0.16 hectares was undertaken on 25 & 27 May 2020. The readings were processed using Geoplot version 4.01. The ground conditions were very dry, much disturbed, with significant amounts of modern debris on the ground surface. Modern structures included a concrete block outhouse and drain to a culvert/ tank located by a metal drain cover. High resistivity readings identified an area of clay extraction which had been backfilled. Readings were high around the edge of the chapel caused by building debris. The line
Top
Fig 7: Grey-scale resistivity survey results within St Andrew’s Chapel curtilage overlaid on a Google Earth image
Bottom
Fig 8: Grey-scale resistivity survey results within Boxley Abbey inner precinct overlaid on a Google Earth image of a high-pressure gas pipeline is present in the results. Due to modern ground disturbance and contamination, a magnetometer survey was not conducted.
Boxley Abbey
Both resistivity and magnetometer surveys were conducted in all suitable open areas. Both survey types were detailed totalling: -
Magnetometer – 6 days – 700,000 readings – 4.41 hectares.
Resistivity – 35 days – 230,000 readings – 5.77 hectares.
Previous investigations projected the floor plan of the main abbey buildings. Resistivity confirmed the layout plus additional buildings and features previously unrecorded.
In the north of the main abbey complex, several features suggestive of buildings, enclosures, water courses, and trackways lie within
Above
Fig 9: Martin Bridge at work
The results for the field to the west of the main abbey buildings (south of the gatehouse) shows a complex overlaying of structures and buildings on differing alignments, which suggests multiperiod occupation and redevelopment. There were additional buildings/rooms attached to the gatehouse. There is a probable wall that is contemporary with these buildings, which crosses the field. A shallow lynchet is apparent at ground level. The footprints of buildings not represented on any old maps are north and south of this probable wall. There is a circular 24m diameter feature present that lies under or overlaps a building. The purpose of this circular feature is unknown; however, it is not present in the magnetometer results. Structures present on the estate map by John Smith (1801), since demolished, close to the hospitium (barn), have been identified on the resistivity results.
Results for the final field to the south-west of the site (south and east of the hospitium) show a series of buildings plus features associated with water management.
DENDROCHRONOLOGICAL INVESTIGATION AT ST ANDREW’S CHAPEL
by Dr Martin Bridge, Oxford Dendrochronology Laboratory
The first investigation established a date for the chapel’s roof, with one timber retaining complete sapwood from a tree felled in the spring of 1484. Other timbers have a similar likely felling date, estimated from their incomplete sapwood. The western extension to this roof was made from trees felled not long after, in the period 1482–1514.
Later investigations made possible as roofs were opened up established that the southeast wing adjacent to the chapel used timbers from trees felled in a slightly earlier period, perhaps a couple of decades earlier, but it isn’t easy to draw firm conclusions from just two dated timbers. However, what was most surprising was that what had been referred to as the extension to the southeast wing contained the earliest felled trees on the site (felled in the period 1418–50).
This throws into doubt the build sequence of the various elements making up the site and requires careful interpretation. Could it be that this southeast extension has used recycled timbers from another building? No evidence of this was seen at the time of sampling, but it remains a possibility.
A single floor beam was the only candidate to date the insertion of the floor into the chapel, but the 74-year long ring sequence obtained failed to date using standard dendrochronology. This has been sent for radiocarbon analysis, establishing a date for this significant change to the building. The bar diagram (overleaf) shows the dates obtained for various elements of the building to date.
The dendrochronological investigations at the chapel were funded by Historic England.
INVESTIGATING THE CISTERCIAN DRAINAGE
by Robert Hall, Kent Underground Research Group (KURG)
Particulars for the Boxley Abbey Estate produced in advance of it being auctioned in the late 19th century contain a tantalising description of extensive tunnelling:
“In the Abbey Wall, on the north side of this [Kitchen] Garden, are Two Entrances to Underground Passages, which are said to lead to Boxley Church on the east, and to Allington Castle on the south west.”
The estate agent somehow omits to mention here that these entrances were probably actually constructed to allow gongfermors to “muck out” the drainage channel under the reredorter – or communal monastic latrine – of the Cistercian monastery.
If, however, the writer was correct in his assertions about the extent of the tunnels, then the current august owner of Allington Castle might be startled to learn that his home was at the wrong end of an implied 2.6 km medieval pipeline of ecclesiastical excrement from Boxley Abbey and Boxley Church.
KURG is somewhat sceptical of such stories and so went to Boxley equipped merely to investigate the few tens of metres of passage that could be directly inspected by the Mk 1 human eyeball, augmented by an assortment of endoscopes, wide-angle video cameras and radio sondes secured on drain rods.
Comparative research was also undertaken at other monasteries and priories: their custodians aghast at our avoidance of the monumental masonry of the abbeys and claustral buildings in preference for the muddy ditches and culverts that formed the remains of the latrines and the associated drainage channels. The usual Cistercian arrangement was for monks to access the first-floor level directly from their “dorter”, or dormitory, into the reredorter to apply their rears to what might be delicately referred to as “drop zones” depositing into the drainage channel perhaps 3 or 4 metres below. The upper level of the reredorters typically seem to be spacious and capable of multiple occupancies with little privacy – perhaps to diminish the risk of dirty habits. Little appears to be known about the function of the ground floor of the reredorters, nor the use of latrines by day or by lay brothers, although in some cases it does seem that corrodians and other more exalted occupants of the monastery had their own latrines also feeding into the same drainage system.
Above
Fig 10: SPAB bar diagram
In the case of Boxley, the paper by Tester in Archaeologia Cantiana LXXXVIII described the remains of the reredorter with much focus on the stonework forming a sluice gate. This comprises a rare survival for a reredorter in such a complete form. Tester also proposed that the drainage channel ran from east to west before turning abruptly south and into a pond. In a pleasing demonstration of inertia, the modern septic tank arrangements seem to have a nearby parallel southern run into the same pond. Based on calcified deposits on stonework, we concur that the most likely direction of flow in the drainage channel was from east to west. However, whereas Tester proposes that the sluice gate was upstream of the “drop zones” with a relatively small reredorter at the end of the east range of buildings, we believe that the visible sluice gate is downstream of the “drop zones” and that the reredorter was a substantial building running east at right angles to the dormitory, similar to that at Cleeve.
The north side of the Boxley Abbey complex is provided with ample water from springs rising on the nearby North Downs. The south side is not so blessed. In winter, surface water run-off might have been an adequate supply to flush the drainage channel when needed, but summer flows are more limited. Rainwater run-off from roofs of buildings might have been used or an engineered channel to direct water from the north to the south side of the complex, but this has yet to be identified.
FUTURE PLANS
Above
Fig 11: Jonny SPAB
ACKNOWLEDGEMENTS:
Following the comprehensive survey work in and around Boxley Abbey, a full report of the findings will be submitted to Historic England in due course. It is hoped that the strong results will help local groups make a case for limited archaeological investigations to assist with understanding the site. The strong team of professionals and volunteers that the SPAB assembled for the project has enabled local archaeology groups to demonstrate their commitment. MAAG and HAARG have a special connection to Boxley Abbey as the late Albert Daniels was part of the Tester excavations in the 1970s, and of course, he was an important member of both groups, as well as KAS, so it seems fitting that we should be continuing to investigate the site.
The SPAB’s Old House Project at St Andrews will continue until 2024, www.spab.org.uk/learning/ working-parties
The SPAB would like to thank AHF and Pilgrim Trust for funding the Old House Project. The SPAB is grateful to SOCOTEC and Terra Measurement for donating their services.
PHOTO CREDITS:
Figs 1–6 courtesy of Graham Keevil Figs 7–8 courtesy of Hastings
Area Archaeological Research Group (HAARG)
Figs 9–10 courtesy of Martin Bridge Fig 11 courtesy of SPAB