|
|
Archaeologists’ Report
HOLY TRINITY CHURCH, STREET, SOMERSET
ARCHAELOGICAL CLEANING AND RECORDING IN THE NAVE BEFORE REFLOORING SPRING 2003
The complete re-flooring of the nave and north aisle of Holy Trinity Church, Street allowed access to preserved archaeological deposits beneath. Investigation revealed the remains of the north wall of the medieval nave (demolished in the 1830s when the north aisle was added) and provided an opportunity to view the remains of the features illustrated in an 1829 sketch bv the Rev d John Skinner. Evidence from the north wall, combined with the discovery of an early, narrow, opening to the chancel suggest that the present church was built during the Norman period and remodelled in the 15th century. Within the nave a sequence of earth floor surfaces of probable medieval date were found to survive over 60% of the exposed area, where not destroyed by later burial vaults and the surface of the former graveyard was exposed beneath the floor of the north aisle. Sleeper walls, probably indicating tlie location of 17th or 18th century box pews, were found to be overlying the medieval deposits throughout the original area of the nave, as well as small areas of contemporary lias slab floor. Beneath the medieval church some hints of the former use of the site were identified, possibly relating to the Early Christian monastic foundation that is traditionally associated with the site.
The front cover shows the simple gravestone of
Prudence Kemp
dy'd
1738
which was found used as base for scaffold posts during the 1830's construction of the north aisle. This stone is likely to have been one of many cleared from the northern part of the graveyard at this time.
1.0____Introduction
In the spring of 2003 work commenced on the complete reflooring of the nave and north aisle of Holy Trinity Church, Street. As this necessitated the total removal of the existing 19th century suspended timber floors it was seen as a prime opportunity to assess the extent and nature of any surviving archaeology beneath. This could provide information relating both to the development of the medieval church and its possible Early Christian origins. The investigation consisted of cleaning and recording the exposed remains within the nave with a single small area of vault infill removed to examine the sequence of deposits. The work was carried out by James Brigers, Alan Graham and Lorrain Higbee with help from Bob Croft. It was funded jointly by the Parish and Somerset County Council. The work revealed significant archaeological remains and the design of the new floor, by architect George Chedbum, ensured that no disturbance took place and the deposits now lie preserved beneath the new floor.
2.0 The Site
2.1 Location
Holy Trinity Church stands in the north-east comer of a large, oval churchyard c500m to the north east of the core of the medieval village of Street in Central Somerset. The building appears to have remained isolated from the settlement it served until Crispin School and Strode College were built to the south and the areas to the north and west were developed for housing in the mid-late 20th century. The church occupies low ground on the southern edge of the flood plain of the River Brue,
looking north across the river to the slopes of Wearyall Hill and Glastonbury. The underlying stratum is a compact grey marl derived from the weathering of the underlying Blue Lias Limestone.
2.2 History
(The following is a precis of the Historical Report on the church produced by the architects, Chedbum Design Ltd. in September 2002).
The site of Street Church would seem to have been in religious use since at least the 7th century. It is thought that the site can be identified with the land at Lantocaii granted by Winchester to Glastonbury in 680, as the name 'Lantokay' has been associated with the area for many years. It has
been suggested that the name Lantokay is derived from the Welsh or Cornish prefixes for an enclosed burial ground 'llan' and 'lan' combined with the name of the little known celtic St Kea. It is therefore possible that St Kea was the founder of a monastery or hermitage there in the 6th century. An indication of the contemporary importance of the site is that it became one of the first possessions of the Anglo-Saxon monastery of Glastonbury.
It is uncertain whether the site remained in continual religious use from the 6th century onwards but it would seem likely that there was a church there in the 12th century, known to have been dedicated to St Gildas in the 13th century. The dedication of the church was changed sometime after the Reformation to the present Holy Trinity.
The earliest part of the existing building appears to be the chancel possibly dating to the 13th or 14th century. The south wall of the nave was rebuilt in the 15th century with the south porch and the tower being added during the same period. In 1826 faculty was granted to the church to construct a north aisle and the work was completed in 1834 to the design of John Ralphs of Warminster. This architect had probably been employed for the works because the patronage of the church had, by this time, passed to the Marquis of Bath at Longleat.
The removal of the medieval north wall of the nave to allow an arcade to be inserted destroyed the curiously shaped and probably early north door that is shown on an 1829 sketch by the Revd. John Skinner (Fig.1). The new layout of the church necessitated the erection of a pulpit close to the south wall of the nave, as members of the congregation in the new north aisle were unable to see into the chancel. The 19th century layout of the interior resulting from these works including pews and suspended floors and appears to have survived with little modification until the present day.
3.0____The Archaeology
3.1 Method
The 19th century timber floors and pews were removed from the entirety of the nave and north aisle leaving only the flagstoned gangways in situ. This divided the nave into four areas of investigation. Area 1 in the north east; Area 2 in the south east; Area 3, a small area in the south west and Area 4 in the north west. Masonry and surfaces within these areas were cleaned and planned at a scale of 1:50. General photographs were taken in colour and black & white with more detailed shots being taken where deemed necessary As the process of construction of the new floor was unlikely to disturb any of the surviving remains very little intrusive excavation was carried out. This was limited in Area 1 to the partial removal of loose backfill [104] to provide a section through earlier deposits against the south side of the medieval north wall footing. In Area 4 a limited investigation was earned out of surfaces in the likely vicinity of the former north door. Sections of these investigations were drawn at 1:20 and incorporated into the two north-south profiles across the width of the nave. All field records have been deposited with the Somerset Record Office.
3.2 Results
Pre-conquest.
Deposits visible in section following the removal of infill from a vault in Area 1 were found to have been cut by the construction trench for the north wall of the medieval nave and therefore appear to pre-date the present building (Fig.2). These deposits lay immediately above the local geological subsoil [123], here discoloured to dark grey-green compact clay with thin layers of weathered lias. Above this material and at the south end of the section a small thickness (0.22m) of dark loamy clay [121] was visible. This may represent the much-truncated remains of a former land surface. Layer [121] was in part cut by a feature containing three small lias slabs [122]; two of these slabs were set upright on edge, one with its face parallel to the section, with a third apparently laid horizontally across the top. These appeared to have been laid in a shallow cut into the surface of the subsoil. Without further detailed investigation it is impossible to arrive at an objective interpretation of this feature. However, it does appear to have been deliberately constructed and may represent the eastern end of a cist grave of Early Christian date.
North of [122], layer [120] was a compact grey-brown clay loam with inclusions of lias fragments and mortar 0.32m in thickness, again apparently the infill of a shallow feature or features. To the north it was clearly cut by the construction trench [116] for the north wall of the medieval nave, and it can be postulated that it is the infilling of early graves, predating the medieval church.
Medieval (c1100-1600)
i) The Structure of the Church.
It was during this period that the present church was established. The investigation allowed the full length of the stump of the former north wall of the nave [117]/[417] to be exposed in plan (Fig.3 and plates 1, 2 and 3). In the side of the vault at the east end of area 1 it was also possible to establish its depth (section, Fig.2). In addition, the north and south footings for the medieval chancel arch [124} [212] were revealed along with the east side of the footing for the west wall of the nave, [304} [418] (Fig 3 and Plate 4). From this evidence the overall internal dimensions of the medieval nave could be determined as 15.00m long by 6.00m wide, with openings at both ends for the chancel arch and presumably the west door.
The stump of wall [117]/[417] ran the full length of the present nave, a total of 15,00m. Its upper portion consisted of five courses of large, faced lias slabs bonded by hard, cream coloured mortar 1.10m in width at the top stepping out twice to the N beneath the upper course to an overall width of 1.40m. This faced and coursed element was found to stand upon an irregular footing of roughly coursed, bonded lias rubble (Fig.2). Limited excavation revealed the overall depth to be 1.10m, including footing. At a point 6.00m E of the junction with the W end of the nave the upper surviving course of this wall is interrupted by a 1.00m wide gap (Fig.3). This appears to be the location of the threshold of the former north door as shown on Skinner's sketch of 1829 (Fig.1 and Plate 2).
In Area 1 this wall had been built in a steep-sided construction trench (Fig 2, [116]). This cut was backfilled with two deposits of mixed clay- loams, lias rubble and mortar flecks [118] & [119] to a level 0.24m below the existing top of the wall. Though in Area 1 the top of the construction trench was obscured by a later deposit of loose loam [112], in Area 4, it line was apparent in one place, cutting through the earlier layer [420] (Fig.3).
Beneath the piers for the present chancel arch two opposed, squared wall termini were found, [124] to the north and [212] to the south (Fig.3). The visible portions of these two wall footings were about 1.10m wide and constructed of large lias slabs bonded by hard cream sandy mortar comparable to the north wall. The space between these two footings indicates a width for the original medieval chancel arch of cl 75m as opposed to the 3.00m of the current arch. At the west end of the nave wall footings [418] and [304] were found with their east edges set 0.30m to the east of the existing west wall (Fig 3). These were of identical construction to those at the east end of the nave although it was impossible to see their overall widths. An apparent comer was, however, seen on wall 304, indicating an opening at least two metres wide in the original west wall of the nave. This could have been for an ornate west door, or an arched opening into a contemporary tower.
A rectangular foundation of at least two courses of bonded lias slabs [113] was found butting up to the north face of wall [117] and overlying its stepped-out lower courses (Fig. 3 and Plate 3). This masonry measured 0.75m by at least 2.00m with its full eastern extent not determined. The location of this feature corresponds with a poorly defined possible projection from the north wall of the nave, possibly a buttress, illustrated in Skinner's sketch of 1829 (Fig.1) and presumably represents its foundation.
ii) Internal Surfaces.
Although heavily disturbed by later activity in many areas several intact surfaces of probable medieval date were identified. The largest expanses of these were made up of [211] at the west end of Area 2, [303] almost over the entirety of Area 3 and [410], [413] & [414] in Area 4 (Fig 3). In Area 1, it remained as layer [115] which sloped away beneath the later deposit [112]. The layers were predominantly compact dark grey-brown loamy clays laminated with yellow sand and, occasionally, mortar. There is strong evidence, especially in Areas 2 and 4, that these represent a sequence of earth floors overlaying one another indicating a process of repair and renewal throughout the period. Unfortunately no specific dating evidence could be found associated with these deposits, however, all the surfaces observed appeared to have been laid against the medieval stone walls suggesting that the sequence begins after their construction. The only exception to this pattern is a small area of compact dark grey-brown loamy-clay, [420] sealed beneath later surfaces [413] & [410] in Area 4. This material exhibited a reddish brown patch, consistent with the effects of contact with intense heat and appeared to have been cut by the construction trench for wall [417]. This evidence tends to suggest that this deposit may represent a surface associated with an earlier phase of use of the site.
Immediately to the south of the probable site of the north door in wall [417] the surface of [410] exhibited a shallow sub-rectangular depression measuring 1.30m x 0.75m and a maximum of 0.10m deep. This hollow was filled with very compact dark grey loamy clay [409] up to the level of the surrounding surface. It would seem highly probable that this depression indicates the effect of wear to the floor of the church inside the north door, this wear being subsequently repaired by the laying of [409].
iii) The Graveyard.
A very compact deposit of dark grey clay-loam was found to be built-up against the north face of the north wall of the nave. This material [114]/[419], represents the truncated surface of the former graveyard prior to the construction of the north aisle in the 19th century. Limited excavation of a section through this deposit in Area 4 produced no further information to aid with date or method of deposition and no evidence for grave cuts was found.
Post Medieval
During this period there is evidence for substantial changes to the interior layout of the nave with the probable installation of seating in the form of box pews and the insertion of burial vaults resulting in the total destruction of medieval deposits at the east end. In between the pews it would appear that a floor of lias flagstones was laid, remnants of which survived, sealing the earlier deposits (Fig. 3)
At the east end of Area 1 a layer of loose loam mixed with mortar is probably the backfill of a collapsed or demolished vault as is a similar deposit in Area 2 (Fig.3). The top of an intact brick vault was visible occupying the south-east comer of the nave. The loose infill in Area 1 [104] was partially excavated and produced pottery dating from the 18th and 19th century along with fragmentary human bone. It is impossible to cite an accurate date for the construction of these vaults but it is probable that they were inserted in the 18th century.
In all areas of the nave parts of walls consisting of a single course of unbonded rough lias blocks up to 0.50m in size were found running east-west (Fig 3, 18th Century sleeper walls). These appeared to be earlier than the duct and sleeper walls of the 19th century expansion and reflooring of the church, and contemporary with remnants of lias slab paving at both the east and west ends of the nave. The walls probably supported the timber platforms for box pews along each side of the nave, with a narrow, paved gangway between. During the dismantling of the timber floor prior to this investigation several lengths of chamfered panel frames were found to have been re-used as joists. These are probably elements of the dismantled box pews.
The 19th Century Alterations.
Most of what is known of the alterations of the early 1830s was self-evident in the standing church prior to the commencement of the current project. However, some further evidence was obtained from the investigation, mainly relating to the disturbance caused to earlier deposits.
In the newly constructed north aisle the surface of the former churchyard appears to have been reduced to maximise sub-floor space with joists being suspended on E-W sleeper walls of has rubble (Fig.3, 19th century sleeper walls). Throughout the nave a large number of loose stone blocks was employed to support the floor, some of which were medieval architectural moulding fragments of oolitic limestone probably salvaged from the demolition of the old north wall. To the east of the south door a roughly laid area of mortared lias rubble was found, [202]. This lay immediately above the medieval floor surfaces in this area and measured 1.90m x 1.00m with a mortar pad towards its north end measuring 0.50m x 0.50m. This feature was almost certainly constructed to support the pulpit when it was moved to aid the view of the congregation in the new
north aisle following the alterations.
Along the north side of the nave a line of three flat-laid lias slabs were found, set m shallow cuts, [111], [411] & [412], These were spaced 2.30m apart parallel to the old north wall and 1.10m to the south of it (Fig.3). Slab [412] proved to be part of a re-used memorial slab and bore the inscription 'Prudence Kemp dy'd 1738' (front cover) providing a convenient 'terminus post quern' for its deposition here. It is probable that these slabs were laid to provide firm bases for vertical timbers, possibly as a temporary measure to support the nave roof during the demolition of the north wall of the nave. The grave slab is likely to have been one of many cleared from the area of graveyard that was to disappear under the north aisle.
4.0____Conclusion & Discussion.
The association of the site of Street Church with that of Lantokay and the implication that it may therefore be the location of a monastic foundation of early date is compelling indeed. However, there is very little solid evidence with which to buttress this assertion. Most of what is suggested for the early history of the site tends to be based on tradition, and it is well known that the monks of Glastonbury had a good track record in the manufacture of 'tradition' for the purposes of financial gain in the highly competitive medieval pilgrimage and patronage market. Although it may be true that tradition had been manipulated at some time in the later medieval period, the grant of land at Lantokay to Glastonbury as early as 680 implies an eagerness to obtain the site implicit of its pre-existing importance. This latter fact in combination with the potentially telling size and shape of the churchyard tends to sway the balance in favour of early Christian occupation of the site.
This investigation, limited as it was in terms of excavation, can provide few answers to these questions relating to the site's early history. However, glimpses in section in Area 1 (Fig. 2) hint at the survival of remains pre-dating the construction of the medieval church. The evidence here defies objective interpretation, and, therefore it can only be stated with all due wariness that these may represent possible grave cuts and the end of a cist of potentially early date cutting into the natural subsoil and a humic deposit that could be the remains of a pre-existing ground surface. In Area 4 a small amount of the surface of a deposit was visible where the medieval surfaces had been eroded, the deposition of this also seemed to pre-date the construction of the medieval church. Apparent heat discolouration on the surface of this material may hint at the fate of a possible earlier structure in the vicinity.
Although the evidence obtained from this investigation relating to the early occupation of the site is tantalismgly frugal, at least success has been gained in proving that these deposits seem to survive intact beneath later developments. Medieval floor surfaces appear to remain undisturbed over two thirds of the area within the original nave, indicating that the potential for the survival of a similar quantity of earlier material is high. The possibility remains, therefore, for an excavation at some future date to provide significant evidence relating to this enigmatic period. This investigation was principally concerned with the area of the medieval nave the remains of the N wall of which was found to survive to a height between 0.60m and 0.70m above its foundations. Skinner's sketch of 1829, although of poor detail, seems to remain the only pictorial record of the north elevation of this wall prior to its demolition shortly afterwards. This sketch clearly shows the former north door and, less distinctly, a possible projection from the east end of the wall, close to its return towards the chancel. Skinner's representation of the door (Fig.1) shows it as immensely tall and narrow with a possibly pointed arch at its head. These proportions are not at all typical of any later medieval architectural style and imply Norman or possibly earlier construction A 1.00m wide gap in the upper surviving course of the north wall was revealed during the investigation immediately opposite the existing south door of the nave. The surviving medieval surfaces immediately to the south of this gap were found to exhibit a considerable degree of wear corresponding to the width of the opening concordant with the focussed passage of traffic over a considerable period of time. There can be little doubt that this gap is all that remains of the threshold of the N door as depicted by Skinner.
The projection illustrated on the sketch would appear to be represented archaeologically by a rectangular foundation of mortared lias [113] against the north side of the north wall (Fig.3). This would appear to be the remains of a broad buttress constructed at some time later than the main fabric of the wall. Once again, this work may have been carried out during the later medieval period.
At the east end of the nave opposed wall stumps [124] & [212] were found beneath the existing late medieval columns for the chancel arch. The current width of the opening of the chancel arch is 3.00m but the space between these two walls is only 1.75m. It is unlikely that the columns would have been intentionally set this far back on their footings suggesting an earlier phase with a chancel arch possibly only 2.00m in width. The proportions of this opening when considered with those of the north door point towards a possible 12th century date for the construction of the church. Unfortunately, no dating evidence was found associated with any of the exposed wails to help confirm this assertion, however, the massive construction of the north wall would fit well into a Norman context.
At the west end of the nave, an apparent gap in the early medieval wall footing suggested a wide opening, up to 2 metres, central to the west wall. This could be a large west door or an arched opening into a contemporary west tower. As the present tower is considered to be of 15th century construction, it may therefore be a rebuild. The floor surfaces of the medieval nave were found to be extant over some 60% of the area exposed during the investigation. The floor levels encountered consisted of layers of compacted clay separated by bands of sand and mortar. No evidence could be detected to suggest that the nave ever had a tile or stone floor until the post-medieval period when lias flagstones were laid in the narrow central space between the newly installed box pews.
Probably during the 17th or 18th century the medieval floors were sealed by the construction of box pews on timber floors suspended across the low sleeper walls that ran in four east-west lines along the entire length of the nave. The positions of these walls indicate that the new seating was constructed in two principal banks 2.30m in width separated by a 1.30m central passage surfaced with lias flagstones. The south bank of pews would have been interrupted to allow access to the south door and font that currently stands to the west of the doorway and has probably not moved since this period. These box pews probably survived until the 1830s alterations when they were swept away in favour of open pews with some elements of their framing becoming re-used in the new suspended floor.
The principal purpose of this investigation was to ascertain the extent and preservation of archaeological deposits surviving beneath the 19th century floor of the church. Also it was hoped that some evidence could be found to aid in the understanding of the development of the medieval church and provide some information to increase the knowledge of the potentially important earlier use of the site. The results obtained show an unexpectedly high level of preservation of the internal surfaces of the nave and hint at a similar degree of preservation of the potentially important earlier material beneath. Examination of the foundations of the medieval nave, suggests that the present church originated as a substantial two cell structure (though possibly with a tower) and may date to the 12th century or earlier. Alterations took place throughout the medieval period, principally the 15th century, including the addition (or replacement) of the west tower, the construction of the porch, the widening of the chancel arch and re-fenestration of the nave. Further to this, the features illustrated on Skinner's sketch of the former N wall have been identified archaeologically enabling them to be placed in their correct locations in the overall development of the church. On the whole this investigation has proved to be a successful and worthwhile exercise providing information that can be used as a firm platform for further work on this important site.
|
Thursday 28th of August 2008
